We are nearly done working through Drexel’s bogus badness about Bayh-Dole. We are considering commercialization. Drexel says Bayh-Dole requires Drexel to commercialize inventions. And Bayh-Dole doesn’t say that. Someone’s gotta be wrong. Oh, hey! I think it’s Drexel!
The basic points are these: Bayh-Dole mandates a working requirement but does not mandate commercialization. Whoever owns a subject invention has to use the patent system to promote utilization of the invention and not to suppress utilization. Utilization could be promoted by contributing the invention to a standard, or by widespread non-exclusive licensing so everyone can make their own. Think about that for a bit–it can be done: any enforcement of patent rights in a subject invention has to show that the suppression of use promotes use. For instance, when Stanford sued reluctant licensee Roche for infringement, Stanford was not using the invention and Roche was. If the invention had been made under Bayh-Dole (it wasn’t, but Stanford claimed it was), then Stanford should have been prevented from suing Roche anyway–it was not within Stanford’s patent property rights to sue Roche for using the invention Stanford claimed to own–that suit would suppress, not promote, use. Yes, think about it. What is the public policy at work here?–a policy that Congress would intend–or what Sen. Bayh, or Norman Latker, or Drexel would intend?
The University must attempt to develop and commercialize the invention. If an attempt is not made, the federal government retains the right to take control of the invention. The government also may take control of the invention for other reasons, such as a need to alleviate health or safety concerns. This provision is referred to in the law as the government’s “march-in” rights.
This bit sort of works. But Bayh-Dole does not mandate either development or commercialization. “Development” itself is problematical and has a long political life that’s divorced from practice. Many research inventions involve research tools or methods–and these are used without any “development.” Not all inventions require “development.” And some inventions that might benefit from “development” also may be used without development. All one needs is to understand the invention and implement it.
“Development” in the history of Bayh-Dole is cover for what a pharmaceutical company does to take a compound that shows biological activity and formulate and test it for use as a clinical product. Much of this “development” is testing to show that a compound is safe and effective–meaning it works for its intended purpose. In effect, “development” is the “first actual reduction to practice” for an invention that claims a clinical purpose. Until one has that first actual reduction to practice, one may have a patentable invention, and even a patent, but really the invention has not even yet been “made” until there’s both conception and actual reduction to practice. Yes, it’s technical, but also yes, it matters, if we are serious about Bayh-Dole, which it should be if Bayh-Dole is important enough for you to read this far.
The Drexel statement here gets rather silly. The university is not obligated to develop or commercialize a subject invention. Rather, the university may lose exclusive control of a patent if it fails to achieve timely practical application, whether itself or through its licensing arrangements. Bayh-Dole does not dictate how the university achieves practical application, or even that it must achieve practical application. The point of Bayh-Dole is that if a university doesn’t timely achieve practical application, then the university has no business hanging onto its patent monopoly. The Drexel claim then messes everything up.
The government does not “take control of the patent”–the government can require Drexel to license the invention on reasonable terms. Drexel still gets any income from such licensing. Drexel still can enforce the patent (presumably to set up more non-exclusive licensing rather than forcing settlements or judgments against infringers for “damages”). Drexel still has control of the patent, but for government-required licensing. Fine point, but important. Drexel overdramatizes march-in and makes it sound like something bad when, if march-in would ever happen, it would be because Drexel was being bad enough that the funding agency had to step in and do the licensing work for Drexel Applied Innovation.
The University must provide the U.S. government with a nontransferable, irrevocable, paid-up, nonexclusive license (“confirmatory license”) to use the invention.
The license requirement is set out in 35 USC 202(c)(4). The license is worldwide and in effect divides the sphere of activity for any subject invention between the government (all government purposes) and the contractor (any purpose that is not a government purpose). In the Kennedy patent policy from which Latker copied text for Bayh-Dole, the markets are laid out expressly. Contractors’ exclusive rights are for the “nongovernmental” market. Bayh-Dole specifies that the license is “to practice or have practiced for or on behalf of the United States.” In the Kennedy patent policy, the scope is “the Government of the United States (including any agency thereof, state, or domestic municipal government).” In the Nixon patent policy, the scope is “the Government of the United States (including any government agency) and States and domestic municipal governments.”
The Bayh-Dole license is to “the Federal agency” but the scope is “the United States”–not merely the U.S. government–and the license is not merely to “use” the invention but “to practice and have practiced.” The Kennedy and Nixon patent policies agree that “practice” means “to make, to use, and to sell.” To have practiced, then, is to authorize others to make, to use, and to sell for any governmental purpose. While a contractor may also authorize others to make, to, to use, and to sell for a governmental purpose, the contractor has no exclusive right to the governmental market. The government, too, has the right to deploy the invention for its purposes, in its “market” as it were. Drexel makes it appear that the government license is limited to “use” when the license is much broader. For some inventions, there may be little or no government market–an invention directed at game controllers, say. Others, such as those that might become medicines, may have a large governmental market, and if so, then Drexel under its Bayh-Dole patents does not have an exclusive right to that market.
In granting a license to use the invention, the University also generally must give priority to small businesses, while maintaining the fair-market value of the invention.
The first part of this “key provision” is sort of true. The provision presumes a single license, as in the special case that a small business and a non-small business are both asking for an exclusive license. Here is Bayh-Dole (35 USC 202(c)(7)(D):
a requirement that, except where it is determined to be infeasible following a reasonable inquiry, a preference in the licensing of subject inventions shall be given to small business firms;
This provision also appears to presume exclusive licensing, but basically makes the preference rather whimsical. If a contractor determines that it is “infeasible” to give a preference to a small business firm, well, then that’s that. The implementing regulations weaken the requirement to whimsy. After a long fuss about the preference, the standard patent rights clause (37 CFR 401.14(k)(4)) lapses into
The decision whether to give a preference in any specific case will be at the discretion of the contractor.
So not much of a preference at all. In the original Bayh-Dole, nonprofit exclusive licenses were limited in duration except if made to small companies. That provision–an actual preference for small companies–was amended out three years after Bayh-Dole came into effect.
The last bit of Drexel’s provision, though, is nonsense. There’s nothing in Bayh-Dole having to do with “maintaining the fair-market value” of any subject invention. That’s more than nonsense, actually. Drexel would have us believe that Bayh-Dole insists that universities extract rents from patent positions, and there’s nothing in Bayh-Dole like that. Bayh-Dole does anticipate that federal contractors might receive payments for licenses (such as in the reporting requirement 37 CFR 401.14(h), “gross royalties”–though reporting royalties is not in Bayh-Dole, for which see 35 USC 202(c)(5)). But the idea that an invention has a “fair-market value” that must be maintained in licensing or small businesses need not apply is Drexel fantasy. That would be fine if Drexel were merely stating its own policy on the matter. But here Drexel makes it out to be federal law, and it’s not, and that makes Drexel’s account a misrepresentation.
When granting an exclusive license, the University must ensure that the invention will be “manufactured substantially” in the United States.
Not quite. Bayh-Dole’s requirement is specific to certain forms of exclusive license–“an exclusive right to use or to sell any subject invention in the United States” (35 USC 204). If the exclusive license is for a non-U.S. jurisdiction, there’s no Bayh-Dole U.S. manufacturing requirement. If the exclusive license is “to make, use, and sell” then broadly the exclusive license functions as an assignment of the invention and Bayh-Dole control, for nonprofits, passes to the assignee under 35 USC 202(c)(7)(A), which requires that the assignee comply with the nonprofit patent rights clause, even if the assignee is not a nonprofit. But, as an assignee, the new owner of the invention is not subject to the US manufacturing requirement, because the license was not an exclusive one “to use or to sell”–it was an assignment. Otherwise, it’s a nice thought. I don’t know of any university that limits its exclusive licenses to “use or sell.” Here’s a bit from Drexel’s template patent license—
Drexel grants to Company an exclusive, world-wide license (the “License”), under the Patent Rights, to make, have made, use, import, offer for sale and sell Licensed Products in the Field of Use during the Term (as such terms may be defined in Sections 1.2 and 6.1). The License includes the right to sublicense as permitted by this Agreement. No other rights or licenses are granted by Drexel.
Except for the right of enforcement:
Company may prosecute any infringement of the Patent Rights at Company’s expense. Company must not settle or compromise any such litigation in a manner that imposes any obligations or restrictions on Drexel or grants any rights to the Patent Rights without Drexel’s prior written permission.
Drexel’s template assumes exclusivity. The grant is for all substantial rights in the invention including sublicensing and the right to enforce the patent. In effect of the license, even with the various qualifications, is to convey ownership of the invention to the “licensee” while retaining for Drexel “title” to the patent on the invention. It’s clever, in its way, but clearly an assignment. So much for the US manufacturing requirement.
Excess revenue must support research and education.
How hard is it to get any of these provisions right? Bayh-Dole restricts the use of “any royalties or income earned by the contractor with respect to subject inventions” (35 USC 202(c)(7)(C)) to “the support of scientific research or education.” Bayh-Dole allows for the use of royalties or income earned to be used to pay “expenses (including payments to inventors) incidental to the administration of subject inventions.” Drexel uses “excess revenue” rather than pointing out that Drexel cannot use any of its subject invention revenue to fund other parts of its technology transfer operations. But Drexel’s own royalty sharing schedule ignores Bayh-Dole’s requirement. Drexel sets aside 5% for its “Office of Tech Comm.”–presumably that’s now Drexel Applied Innovation and folks have not bothered to update their policies, what with how names change so often for tech transfer offices, as if administrators think a name change is what’s needed to make their licensing operations suddenly effective. And of course Drexel leaves out the “scientific” modifier used by Bayh-Dole. Maybe this part is just loose and sloppy, and we should let it go. And we should, after pointing out that it is all loose and sloppy. Is that what we should expect?
The University must share a portion of the royalties with the inventor(s).
Spot on! 35 USC 202(c)(7)(B).