Negotiating a share of royalties under Bayh-Dole

In his amicus brief to the Supreme Court in the Stanford v Roche case, former Senator Birch Bayh made an odd claim–that inventors had the right to negotiate royalties with the universities that took ownership of their inventions:

Rather, Congress rewarded individual inventors by requiring their employers to provide them with a share of royalties to be negotiated with the universities or nonprofit organizations. See 35 U.S.C. 202(c)(7)(B).

But at 35 USC 202(c)(7)(B) we find only this:

(B) a requirement that the contractor share royalties with the inventor;

There’s nothing in the law about negotiation. And nothing in the standard patent rights clause. And nothing in most university patent policies–in fact, those policies are written, now, to preclude negotiation. Some policies even declare such negotiation to be an ethics violation–since an inventor would then have a financial interest in the university’s management of contracts under which the invention was licensed.

Even Sen. Bayh’s own argument regarding the standing of inventors (last in line for rights, which vest conditionally with their employers) appears to eliminate any prospect for negotiating royalties. What leverage is there to negotiate, if the law takes your property away, hands it to your employer, and leaves it with your employer to “share” royalties however it decides? [It’s even worse when it’s not even an employer that gets the rights, but an entity releasing you from your “official duties” to do research in the public interest with public funding the entity receives on your behalf.] That’s not much of a negotiation. You don’t have your property, can’t get it back, are constrained to sign paperwork to permit patent applications to be filed (if the (f)(2) agreement requirement has been complied with), and exactly what else is there that a university must negotiate with an inventor about?

The Supreme Court ruled that Sen. Bayh was wrong about how his law operated–that the words of the law mattered, and not so much the statements about intentions of folks who wanted the law to work the way they said it did. Words matter–my, that could be the subtitle for a number of articles here at Research Enterprise.

But let’s consider this issue of inventor negotiation a bit further. If Bayh-Dole simply requires federal agencies to allow inventors to assign to their university employers, then the assignment transaction itself is subject to a negotiation. That is, there must be “a valuable consideration” provided for the assignment, per 35 USC 261, at least if the assignment is to take precedence over any later assignment. An assignment without consideration is, apparently, voidable by a later assignment. That’s interesting.

Universities, however, have implemented practice under Bayh-Dole by assuming that federal law vests invention ownership with the university. The universities then go about stipulating that they own all inventions made with federal support, or they have a right to take title to such inventions as a condition of federal law, or some such variation. In some cases, the university claims that the right to take title is a condition of employment, or a condition of the use of any university resources involved in making or developing the invention, or a condition of participation in federally supported research.

But making something a condition of employment is not necessarily the same thing as providing consideration for assignment. It depends, among other things, on whether the invention was made within the scope of employment or used university resources as defined by policy. That is, one might find that the consideration for assignment turns out to be continued employment–payment that comes after the assignment. That is, if the inventor quits with their invention, then there’s no consideration for assignment, but if the inventor does not quit, and accepts pay, then the inventor has received consideration. Even that consideration depends on the inventor having invented within the scope of employment–within the scope that an employer can reasonably claim the right to own an employee’s invention.

In some cases, universities have changed policies and asserted the changed policy applies to all employees, even if they have previously agreed to a different policy. That happened in the Shaw case at the University of California, when the university changed its royalty sharing policy and insisted that Shaw accept a lower royalty share. Shaw challenged the university, and the court agreed, giving the university a good spanking, after which the university went back to doing the same thing, having learned nothing. But the court made it clear that if the university has a contract with an employee, while the university may change its policy at will, it cannot change its policy with regard to that employee without the employee’s agreement:

We find no merit in the University’s suggestion that, as a public employee who is employed pursuant to statute, not contract, Shaw has no vested contractual right in his terms of employment, such terms being subject to change by the University.

When a public employer chooses instead to enter into a written contract with its employee (assuming the contract is not contrary to public policy), it cannot later deny the employee the means to enforce that agreement.

Although the University is entitled to revise its Patent Policy, it cannot do so with respect to Shaw because of its written agreement with him.

What’s interesting about the university’s position is that it was arguing that any royalty due to Shaw was not consideration for assignment, but merely the working out of an administrative policy in which the university authorized itself to share out a portion of its proceeds. While it might be said, then, to share royalties with an inventor, it was arguing that it had no obligation to do so as a result of assignment of the invention–that the sharing of royalties was not a “valuable consideration” for assignment. The university argued that the purpose of the patent policy was to vest ownership of inventions directly with the university. The court rejected this as well:

By its terms, the patent agreement embodies Shaw’s promises to disclose any inventions he may create in the future so that the University may “examine[] … and determine rights and equities therein in accordance with the Policy” and, if the University “desires … to seek patent protection thereon,” to assign his interest in the invention to the University.

The clear language of the patent agreement does not, as the University argues, effect a contemporaneous and “complete transfer of plaintiff’s rights to the University.”

The patent policy’s language requires reporting, anticipates a review of circumstances, and if certain conditions are met, then the university may (but does not have to) require assignment of an invention. That assignment is left open by policy. The policy simply makes an agreement to assign in certain situations “mandatory.” But that agreement itself, to be enforceable, must take the form of a contract–offer, acceptance, and consideration. It’s the consideration part that’s lacking. And the university argued that the royalty sharing in the same policy was not, essentially, consideration for the agreement to assign. The university has since gone further and inserted a present assignment into its “patent acknowledgment” document, which all employees must sign, making it appear that assignment takes place before the review required by patent policy–and thus, apparently without consideration, other than as might be argued a condition of employment.

We might find, then, that many university assignments of inventions do not carry consideration. The royalty sharing that takes place is purely administrative policy, a kind of perk if you wish, and not consideration for the assignment. When a university changes its policy, and wants that policy to stick, then it must find a way to get employees to accept the new policy. One way is to fire those that refuse–no continued employment. That may be more difficult for those employees with tenure–not something that one can quickly turn to company case law to find analogs. But if the university does not fire the employee, then the university needs to show some additional consideration that the employee accepts for the change in policy. Without that additional consideration, it’s an open question whether the assignment is binding. That’s especially true if continued employment is already the full consideration that binds an employee to the former policy. Continued continued employment does not reflect the added benefit to the employer of the new policy.

If we turn to Bayh-Dole (hello, darkness my old friend… ), we find that universities have other problems with royalty sharing. Nonprofits are expressly called out for restrictive treatment in the standard patent rights clause. A nonprofit may only deduct certain allowable expenses from income on a subject invention (37 CFR 401.14(a)(k)(3)):

The balance of any royalties or income earned by the contractor with respect to subject inventions, after payment of expenses (including payments to inventors) incidential [sic] to the administration of subject inventions, will be utilized for the support of scientific research or education

Thus, if a university arbitrarily deducts a set amount off the top of licensing income for its technology transfer office, it is not in compliance with the standard patent rights clause unless that set amount is *less than or equal to* the actual allowable expenses. It can only deduct expenses related to subject inventions–not all inventions, not stuff that’s not inventions (like copyrights or biomaterials). Note, as well, that a university could allocate to the inventors all income after its allowable costs and not take a dime otherwise–that would be compliant with the standard patent rights clause. If a university deducts more expenses than allowable before calculating a royalty share due inventors, it has shorted the inventors under its own royalty sharing policy by failing to comply with the standard patent rights clause’s limitation on what expenses it can deduct.

The money “earned” from subject inventions must be fully accounted for–either it is used to pay expenses on the administration of subject inventions (including sharing with inventors) or it goes to scientific research or education. There are no other allowable items in the royalty stream budget. Sharing with inventors is, by law–since the language is in Bayh-Dole and repeated here in the standard patent rights clause–an expense. It’s not a perk.

There’s more, natch. It gets better. Nonprofits must share royalties with inventors (37 CFR 401.14(a)(k)(2)):

The contractor will share royalties collected on a subject invention with the inventor …

We might notice the difference in language between

(k)(2) “royalties collected on a subject invention” and

(k)(3) “any royalties or income earned by the contractor with respect to subject inventions.”

In (k)(2), the target is singular–invention; in (k)(3), it is plural–inventions. In (k)(2), we find only “royalties” but in (k)(3) it appears that there can be income other than royalty that must be accounted for. In (k)(2), the money must be “collected.” In (k)(3) it is “earned.” That’s a big difference. Money may be earned and not collected. What’s the difference? For the use of any money not collected but “earned”–whoever holds that money is to be under the obligation to use that money only for scientific research or education. If, for instance, a university’s research foundation does the licensing and receives the money, but doesn’t turn that money over to the university, the research foundation is still stuck with compliance under (k)(3). It can’t hold the money and, say, invest it. That’s not using the money for scientific research or education, but rather to make more money by betting on the movement of markets.

Why is the language between (k)(2) and (k)(3) so inconsistent? What income might there be that’s “earned” on a subject invention but not a royalty, given that a “royalty” is generally any amount paid in consideration for a patent license. Certainly one could earn income from a subject invention in ways other than licensing–one could sell the invention and what one receives is not a royalty. One could also sell product based on the invention, and the income received is not a royalty–it is sales revenue. And notice that in (k)(3), the language is broad–“earned with respect to”: so, one might sell services associated with the invention, such as technical assistance, and that income, too would not be a royalty and would come within the scope of the requirement at (k)(3). Even sponsored research income would appear to come within the scope of “earned with respect to” if the research is directed at a subject invention.

I should remind you that virtually no-one gives a rat’s ass about any of these details. Few are the universities that bother to distinguish any of these, or care. But now take a look at (k)(1):

Rights to a subject invention in the United States may not be assigned 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 will be subject to the same provisions as the contractor;

I’ve highlighted the text to consider. If a university assigns a subject invention, then it must pass through the same nonprofit requirements the university has–that is, (k)(1) to (k)(4), and so including both of the money restrictions we have just discussed.

If a university wants to reduce its exposure to inventors, here’s how it does it. First, it assigns inventions (or makes inventors assign inventions) to an affiliated “research” foundation. Now that foundation, often operated by the same people from the university, goes out and does a license. The licensee pays the foundation, and the foundation pockets a share of the money (often a set amount, in the cases I know about, costs plus around 40%) and passes the rest through to the university. The university then shares what it “collects” with inventors–by applying its patent royalty schedule. Often the inventor gets around 30%–sometimes up to 50%, sometimes less. But you can see the difference–the inventor’s net is actually not 30% at all, but rather 30% of 60%, or 18%–a little over half of what is promised. When a university controls the research foundation, then the negotiation of the foundation’s fees is simply an exercise in deducting money from the revenue stream for the university’s benefit, before the university deals with inventors. But the research foundation has the same obligations with regard to deducting only allowable expenses and sharing royalties with inventors–it does not matter what the deal might be between the university and the research foundation. The federal obligations take precedence over state-enforced obligations.

This practice is a variation on a university deducting its expenses before paying inventors any share. Given that a university can run up quite a patenting bill, in such a scheme inventors ,may get nothing, even if the university collects some royalty income. The standard patent rights clause requires a university to share, but does not come out and say when or how much–but $0 is not sharing. Thus, if a university deducts its costs from a royalty payment and does not share anything, it breaches its (k)(2) obligation. Sharing with inventors is included as an expense with other expenses. It’s not an afterthought and the requirement to share means that a university can’t pay itself first, if that means leaving nothing for the inventor.

We might go so far as to argue that “share” here means something like “tenants in common”: whatever comes in gets shared as agreed, and if not agreed otherwise, gets split evenly. If there’s an agreement on sharing, well, that might involve negotiation, regardless of a policy royalty sharing schedule, since the obligation to share is a matter of federal agreement and does not depend on, and is not subordinate to, a university’s patent policy. In fact, the federal agreement represented by the patent rights clause takes precedence over a university’s patent policy. The patent rights clause comes first. A university agrees to the patent rights clause. It therefore cannot at the same time not agree to the patent rights clause in order to enforce its patent policy. Many universities have research policies that make this order of precedence clear–the research contract overrules policy. That includes, in the case of federal funding, a university’s rules on royalty sharing. “Share” is the requirement, not “share after the university has swallowed all it wants to.” That may well mean *equal sharing* until an inventor agrees otherwise.

Back to (k)(1). If a university assigns a subject invention, it passes as well the obligations in paragraph (k) of the standard patent rights clause. That is, the assignee must “share royalties collected” with the inventor. It is not just the university that must share–it is also the assignee. Let’s take two cases. First case: university assigns to a research foundation that then does the licensing. The research foundation, under (k)(1), accepts the same obligations as the university. Thus, the research foundation must share royalties with the inventor, in addition to sharing royalties with the university. Both organizations have this obligation. The university’s obligation is not met by the research foundation sharing with inventors, nor is the research foundation’s obligation met because the university shares what it receives with inventors. 

Yeah, I italicized all that to put some tone into it. University attorneys will jump up and down about how wrong it all is, but they’ve been consistently wrong about plenty of things involving Bayh-Dole and that’s a hobgoblin they have chosen to live with. What matters are the words–the requirements of the written, federal agreement between university and the federal agency. More importantly, of course, what matters is the intent behind the words, and there we have a battle of wills. University attorneys–for whatever reason– routinely intend to limit what the university should share. I haven’t met a university attorney yet who thinks the “client” is the inventor and that the intent of policy should be as equitable as possible with regard to the inventor’s interests.

The way I’ve put things, above, takes the position of inventors–why should they agree to a lesser share just because the university uses a machination to make it appear there’s less to share? Why should the share from the research foundation be anything less than half, unless voluntarily negotiated otherwise with the inventors? There is nothing in Bayh-Dole or the standard patent rights clause that permits a research foundation or other assignee of a subject invention from “sharing” royalties with anyone other than the inventors. The options for a research foundation are: deduct allowable expenses (including sharing royalties with inventors) and using the balance for scientific research or education–so any transfer of money from royalties to a university must stipulate the money must be used only for these two purposes. Not for also “sharing” with inventors. The foundation’s money is not the university’s money. When the foundation has received assignment of a subject invention, then it is the foundation’s royalty sharing agreement with inventors that applies, not what it is in a university’s patent policy regarding what the university will share. And absent an agreement with inventors, “sharing” ought to be a matter of 50-50 unless agreed otherwise. There would appear to be a basis for negotiation with inventors–but it is flipped. If the research foundation does not want 50-50, then it must decline the assignment or work something out with the inventors that offers them something else they want other than 50% of the royalties received.

Thus, a university cannot reduce its obligation to share by assigning subject inventions for management to an external foundation.

Second case: a university assigns a subject invention to a company, under the cover of an exclusive license that grants substantial rights in the patent to the company–telltale signs: the license grants an exclusive right to all rights of the patent, reserves only noncommercial, educational, and government rights, and gives the company the right to enforce the patent and to sublicense. These are typical terms for a university exclusive license. In that case, regardless of the label put on the document, the transaction involves an assignment of patent rights. If that’s the case, then the company ends up being “subject to the same provisions as the contractor. The consequence of assignment is that the company accepts limitations on assignment and on the use of royalties (as well as the requirement to prefer small businesses in any licensing–which would apply to any sublicensing clause). 

As an aside, you can see how universities avoid compliance with (k)(4), which requires a preference for small businesses in licensing–just grant an exclusive license to the subject invention, and give the licensee the right to sublicense, and leave it at that. The university has given up the right to license small businesses (the license is exclusive but for noncommercial uses) and the university does not require the exclusive licensee to give a preference in any sublicensing to small businesses. In this way universities mock the small business preference objective.

If a university assigns to a company, then the company ends up with an obligation to share royalties with the inventor, quite apart from whatever obligations the company may have to the university. It may be that the company, as an assignee, does not have any obligation to share income “earned … with respect to subject inventions” with inventors, but it will have the obligation to use such income, after allowable expenses, only for scientific research or education. That money doesn’t go back to the shareholders, doesn’t plump executive salaries. That’s the gesture of owning a patent on a subject invention, an invention made in a project to further the public interest. While the public interest might not be overtly protected by price restrictions (those fickle “reasonable terms”), it certainly is in the form of restrictions on the income from exploiting a subject invention. This is a great realization. Except, of course, no university I know of admits that they assign patent rights in their exclusive licenses, even when courts rule they have.

The requirements on assignment and royalties in paragraph (k) do not apply to inventors directly or to small companies. Inventors, if they retain their invention rights, are treated as small business contractors. They may assign freely and happily as they will and pocket whatever they happen to earn, collect, or otherwise get paid.

Which brings us back to an odd idea. Is it possible that the sharing requirement itself is subject to inventor negotiation? Perhaps Sen. Bayh was not entirely mistaken about his own law. If federal agreements take precedence over university policies (and they do), then the standard patent rights clause takes precedence over university patent policies and their royalty schedules. The patent rights clause becomes effective with each new federal funding agreement. (That’s one reason that it is a gross mischaracterization to claim that universities are subject to Bayh-Dole, as if the law applies to universities and inventors alike, requiring them to do things. Bayh-Dole applies to federal agencies and sets the conditions for a standard patent rights clause to be used unless an agency can justify an exception.) Universities agree to the patent rights clause. It takes precedence. In that patent rights clause, universities agree to require inventors to make a written agreement to protect the government’s interest in subject inventions. That’s the (f)(2) requirement. That requirement comes from the university, not from the government, not from the law. The university makes inventors parties to the funding agreement–contractors.

Here’s where it gets interesting. When the university gives its inventors standing as contractors, it gives up its own claims to the inventors with regard to its own patent policy. It has to. The (f)(2) agreement is the agreement that the university has agreed to substitute for its own requirements. If the university wants to acquire a subject invention from an inventor, it has to negotiate for that right–it can’t make that right a condition of the federal funding, because as a condition of the federal funding, it has just made the (f)(2) agreement the requirement, and the (f)(2) agreement authorizes the inventors to sign papers to establish the government’s rights in subject inventions, something they can do only if they have rights in their own inventions. They would have no right to license or assign their inventions if university policy could at the same time claim all inventions outright. In this way (f)(2) is not made void and unnecessary by university policy–just the opposite: (f)(2) makes university policy unnecessary and at the same time protects inventors from predatory university practices.

If (f)(2) must operate, and universities must agree that it operates, then each time the university accepts a federal funding agreement, if it wants to acquire rights in a subject invention, it must negotiate with the inventors. It cannot simply take any invention it wants. It cannot treat a share of royalties as an administrative perk to be anything it wants. It must acquire an invention by assignment, and for that assignment to be binding, there must be “a valuable consideration.” A university may be inflexible and assert that it will only share as its policy states. But that does not mean that a university has any right under (f)(2) to demand assignment.

Of course, NIST proposes to change all this and turn Bayh-Dole into the vesting statute that university attorneys have lusted after for years by “clarifying” that (f)(2) is supposed to require assignment, not just reporting, of inventions–even though there is absolutely no authority for doing so in Bayh-Dole and the disposition of ownership is exactly what Bayh-Dole addresses as federal patent policy. The Supreme Court ruled that Bayh-Dole did not disturb common law rights of inventors, and there’s then nothing that NIST can do in administrating the requirements of Bayh-Dole to make federal patent policy for all agencies require that inventors assign subject inventions to their employers.

In this reading–one that no university attorney I know of will accept (so you’ll need $200K or so to beat them down on it), the (f)(2) written agreement requirement sets up an inventor’s right to negotiate with the university over sharing of royalties. Small businesses have the same (f)(2) requirement, but have no obligation to share royalties, so for them this is all different. A university inventor is faced with a choice: assign to your employer for whatever share of royalties you can negotiate, or deal with the federal agency, which may or may not request title. If the university won’t negotiate and you don’t like the royalty share it offers, then take your chances with the federal government. If the federal government lets you retain title, you can license or assign to anyone, negotiate what you want with anyone, for whatever you can get. And if the federal government requests title, then you’ve taught the university a lesson and you likely will have access to your invention if you leave the university, since the government still may license non-exclusively and perhaps even royalty-free.

If all of these feels like a trip through Detail Land, well, it is. But if we put detail things front and center, we might say boldly that:

“Bayh-Dole requires universities to allow inventors to negotiate a share of royalties as a condition of any assignment of their inventions to the university or any other organization the university might require the inventors to assign to, and such share shall be required, as well, of any organization to which the university assigns the rights it acquires from the inventor, unless the inventor agrees to a different share.”

That, perhaps, is what Sen. Bayh meant when he wrote that Bayh-Dole allowed inventors to negotiate a share of royalties. That’s the effect that (f)(2) should have–if anyone complied with it. That’s the basis for the requirement that nonprofits share in paragraph (k) of the standard patent rights clause. Inventors have no chance, now, to negotiate royalty sharing, just as for 30 years they were told that they had no right to negotiate assignment. Well, the Supreme Court showed that they did have a right to negotiate assignment. Pity that it will take another Supreme Court case to prove that inventors also have the right to negotiate royalty sharing, and should have that sharing not just for the university but for any organization that the university assigns rights to, whether the university calls the assignment an exclusive license or otherwise.

The financial liability exposure to such litigation for universities and their assignees runs to the hundreds of millions of dollars. A fine class action lawsuit would do the trick. All that’s needed is a law firm willing to work on contingency and a few plaintiffs wanting to get their fair share from what their universities have taken from them under the false color of law.

This entry was posted in Bayh-Dole. Bookmark the permalink.

Leave a Reply