The NIH’s View of Bayh-Dole Compliance

In 2015, Ann Hammersla gave a talk at an NIH Regional Seminar that include Bayh-Dole. There are numerous problems with Hammersla’s treatment of Bayh-Dole, but we’ll leave most of those for the attentive reader to pick through.

Let’s look at just one slide, dealing with NIH compliance, mostly involving Bayh-Dole:

This slide is as close as Hammersla’s talk gets to general Bayh-Dole (patent rights clause) compliance. The title of slide and the list of bullet points don’t mesh. Some things are indeed reporting, but others are not–employee agreements, file patents, U.S. manufacturing. More importantly, look at what compliance means–administrative actions that don’t have much at all to do with “public benefit.”

Let’s work through the list and audit things against Bayh-Dole and the standard patent rights clause. It’s clear that the list is a shorthand–and there’s only so much detail one can fit onto a slide. That’s not the problem (although, given this is already a 58 slide deck, one might think that reporting could be separated from other responsibilities).

Implement Employee Agreements “as needed”

Bayh-Dole has nothing to say about employee agreements. It’s not there. Bayh-Dole does not stipulate that the standard patent rights clause have employee agreements (see 35 USC 202(c)). But the standard patent rights clause (37 CFR 401.14(a)) does require contractors to require certain employees to make a written agreement to protect the government’s interest. That’s the (f)(2) written agreement, and it is not contractor-optional. It’s required, not “as needed.” And (f)(2) is required not merely to provide a link directly between the federal government and individual inventors, but also to make those inventors parties to the funding agreement–thus, also contractors (see 35 USC 201(b) and (c)), and thus every invention they make in the performance of work under a funding agreement is a subject invention, owned by a contractor–by each inventor who has made the (f)(2) agreement. The “as needed” is simply wrong. Here, the NIH is all but saying, “you don’t have to comply with (f)(2).” For that to be the case, the NIH should have to declare “exceptional circumstances” and create their own standard patent rights clause.

Of course, if what Hammersla means by “as needed” is “as needed for the institutional contractor to require assignment of inventions from inventors,” then the matter is one of just what inventions an institutional IP manager requires. Have employee agreements for those, advises Hammersla. But that’s not at all a matter of compliance with Bayh-Dole or with NIH requirements. At best, it’s a matter of exploitation of an opportunity, and even there, it is only the opportunity of “if you can find a way to take an invention from your inventors, then Bayh-Dole limits the NIH’s ability to take it away from you.” To make this rather strange opportunity appear to be a regulatory requirement is also rather strange. Perhaps that’s why “as needed” settles folks’ administrative consciences.

There’s an argument–a good one, even–that Bayh-Dole pre-empts nonprofit institutions operating with a policy of academic freedom from imposing assignment agreements as a condition of participating in work under a funding agreement. Bayh-Dole pre-empts federal patent policy and states as that policy the conditions that federal agencies must require in any funding agreement. Those are the conditions, then. When the standard patent rights clause then transmits these conditions to an institutional contractor for acceptance, the institutional contractor accepts these conditions in place of any conditions that the institutional contractor otherwise has or might seek. It’s just like the subcontracting requirement in the standard patent rights clause, also not in Bayh-Dole. The employee agreement that a nonprofit institution must have is one that protects the government interest. It cannot have one that at the same time prevents the (f)(2) agreement, the required agreement, from operating.

The “as needed” bit is simply wrong.

Disclose Each Invention “within 60 days”

The requirement is directed to each subject invention. In the standard patent rights clause, it’s “two months” not “60 days” and the clock starts when the inventor discloses the subject invention to “contractor personnel responsible for patent matters.” In Bayh-Dole, the requirement is “within a reasonable time.”

Resolve Election or Waive [sic] of Title “within 2 years”

The statutory wording is “elect to retain title” not “election of title.” “Elect title” does also get used, but only in reference to the section in which “elect to retain title” is the formal requirement for the standard patent rights clause. There are plenty of conditionals on the timing of the elect to retain notice. The federal agency can extend the notice term. The notice term may be shortened by a federal agency if there’s a statutory bar created. (But of course, in typical Bayh-Dole fashion, how would an agency know if there’s a statutory bar unless the contractor properly reported each bar as it occurred? And even then the agency has to take an express action to shorten the term–administrative overhead designed not to operate).

File Patent “within 1 yr. of election”

Well, technically it’s file a patent application. In Bayh-Dole the filing is to be “prior to the expiration of the 1-year period referred to in section 102(b)” [102(b) has to do with exceptions to the loss of the right to a patent based on publication, public use, or available to the public before filing a patent application]. It’s not anchored to the election of title at all. The standard patent rights clause, making things up as it has a habit of doing, turns this requirement into “within one year of election of title, or, if earlier, prior to the end of any statutory period wherein valid patent protection can be obtained” [with a list of some 102(b)-like conditions].

Provide License to the Govt. “upon title election”

Bayh-Dole provides that a federal agency “shall have a nonexclusive, nontransferrable, irrevocable, paid-up license to practice or have practiced for or on behalf of the United States any subject invention throughout the world.” It does not say when this license is to be “provided” to the government–just that the government “shall have” it. This same language is repeated in the standard patent rights clause. The standard patent rights clause states that the contractor will “promptly deliver” “all instruments necessary to . . . establish or confirm the rights the Government has.”

Indicate Govt. Support on Patent “with patent appl.”

Yes. But the purpose of the statement is more than to indicate government support–it is also to announce that “the government has certain rights in the invention” (see 37 CFR 401.14(a)(f)(4)). In a sense, the government rights legend is equivalent to the government license. The “certain rights” include those of the government license (as well as the right to receive title in some circumstances).

Product Manufacturing in U.S. “required”

Not true in general. Manufacturing in the U.S. is required only for an exclusive license in the U.S. to “use or sell.” There is no requirement for U.S. manufacture for non-exclusive licenses, and no requirement that a patent owner itself manufacture in the United States the product it uses or sells. Even the non-exclusive license to the federal government does not require U.S. manufacturing. We have here a common overstatement of the requirements–and one reason for the overstatement is that some people cannot imagine any institution using non-exclusive licensing strategies. The NIH apparently assumes licenses will be exclusive–in keeping with being the origin within the federal government of the circumvention of federal patent policy that expected non-exclusive licensing.

A subsequent slide highlights the exclusive license element. That same slide also highlights that a federal agency can waive the U.S. manufacturing requirement and provides a link to an NIH web site where there’s a form for requesting a waiver. One might consider that the NIH is eager to grant such waivers–one more indication that the public benefit apparatus of Bayh-Dole is there only for appearances.

Report on Invention Utilization “annually”

Well, for subject inventions, and no more frequently than annually, and only if requested by the federal agency. “The Contractor agrees to submit on request periodic reports no more frequently than annually on the utilization of a subject invention or on efforts at obtaining such utilization that are being made by the contractor or its licensees or assignees” (37 CFR 401.14(a)(h)). Perhaps the NIH always requests annual reports.

Final Invention Report “at award close out” 

This is a funding agreement requirement, but does not arise in Bayh-Dole or the standard patent rights clause. One might wonder if regulations pertaining to “final invention reports” are pre-empted by Bayh-Dole, which stipulates the invention reporting that federal agencies may require.

The impression then of working through this summary slide is that things are presented in a sloppy fashion. Perhaps in the talk itself Hammersla provided the necessary nuance. The things that stand out, however, are the things that are left out:

How does a contractor come to own inventions made with federal support? Apparently, by those “as needed” employee agreements. Not by complying with the (f)(2) written agreement requirement to protect the government’s interest.

Why should a contractor even be involved in dealing with inventions? That, apparently, was already answered by the poem about the high calling of institutional IP managers in the service of creativity and innovation.

What limitations are there on the property rights in patents on subject inventions? It would appear that such patents are ordinary patents, and that the policy statements for Bayh-Dole, placed in federal patent law, are surplusage.

But here’s the greatest emptiness in the slide–that the issues that matter are ones taken up with institutional ownership of subject inventions, what is necessary to comply and keep ownership. By contrast, the purpose of the “IP system” that Hammersla aspires to is “public benefit.” In that IP system, institutional ownership of patents is assumed–unlike the U.S. patent system, in which personal ownership of patents is assumed. Somehow, the Hammersla envisioned IP system is envisioned to be better than the U.S. patent system–that institutions are better than inventors in deploying their own work. It is this belief that lies at the heart of championing an aspirational IP system in place of the statutory framework we have.

In Bayh-Dole, that public benefit is established by Bayh-Dole’s statement of policy–“use the patent system to promote the utilization of inventions arising in federally supported research” and in Bayh-Dole’s definition of “practical application.” These elements create a working requirement in federal patent law for subject inventions and place limitations on the exploitation of patent property rights specific to subject inventions. The march-in provisions make evident some elements of the working requirement. The provisions on nonprofit exclusive licensing other than to small businesses–almost immediately removed from Bayh-Dole, in 1984–showed as well the working requirement. But none of this is evident in Hammersla’s talk. Yes, she cites the AAU statement on university licensing, but that’s just more administrative fluff about intention–universities should focus on efforts to “advance the public interest and public good.” Whatever does that mean? Universities should have “high-level policies” to align IP management with “public interest” and “core missions.” So, policy consistency in wording involving abstractions. Neat. Nothing in the list of points about actual practical application of subject inventions. It’s all about the administrative process of aligning administrative processes with administrative processes.

Bayh-Dole is an instrument designed to allow the pharmaceutical industry to gain monopoly access to inventions made with federal support. It operates by creating the conditions for university IP managers to take ownership of inventions, obtain patents, and pass these patents to companies as monopolies, sharing in their exploitation and passing some portion of their share to inventors, to make a show of the “reward” for participation.

The text for Bayh-Dole originated in the NIH, with help from WARF. It is a creature of IP brokers with a sure belief in the value of monopolies, and an even surer belief that monopolies in the area of public health have the potential to generate significant royalty income. And in these things, the IP brokers have been spot on. Monopolies do have value. IP brokers can make good money trading in patent monopolies. Monopolies where there is pain and suffering can be lucrative. And it may even be the case that institutions are a better front for squeezing monopoly value out of pain and suffering than any individual academic inventor, who might not have the stomach for the effort.

But in all of Hammarsla’s discussion of compliance with Bayh-Dole, she does not provide the definition of “practical application” nor that to meet this definition using the patent system, an institutional owner of a patent on a subject invention will have to do more than have aspirational policy statements or comply with petty requirements about how to keep ownership of patents on subject inventions.

It’s what’s left out of Hammersla’s discussion of Bayh-Dole compliance that’s telling. Even in the utilization reports–government secrets as they are–the point is that a contractor should report the date of first commercial sale or use. It’s not that the contractor file the report annually that matters, but that there’s actual performance, subject inventions actually being used in commercial settings. That’s the express interest of Bayh-Dole. In the law as originally passed, the date of first commercial sale or use mattered, as it started a time limitation on the duration of an exclusive license from a nonprofit to any non-small business.

a prohibition against the granting of exclusive licenses under United States Patents or Patent Applications in a subject invention by the contractor to persons
other than small business firms for a period in excess of the earlier of five years from first commercial sale or use of the invention or eight years from the date of the exclusive license

But that prohibition was quickly removed by amendment. Removing the prohibition, however, did not change Bayh-Dole’s general policy requirements on patent properties pertaining to subject inventions. Except, of course, if the representatives for federal agencies choose not to bother with practical application.




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