[Bayh-Dole’s section 204 (35 USC 204) permits federal agencies to waive the requirement that exclusive licenses to use or sell in the United States must require product sourced from United States manufacturing. The waiver provisions are easy to meet and work against potential United States licensees. Section 204 ends up being an empty gesture.]
We have looked at the first sentence of section 204 of the Bayh-Dole Act, which purports to dedicate subject inventions to American industry for manufacturing, but actually does almost nothing for American industry while making a big flourish about it. Section 204 lays out a narrow framework for requiring American manufacture of products based on subject inventions–but only for a limited form of exclusive license–to use or to sell. There’s nothing in Bayh-Dole requiring subject invention *owners* making American when *they* use or sell in America. An owner of a subject invention can make product wherever it wants.
One would think that if a purpose of Bayh-Dole is to promote the commercial use and availability of subject inventions, then the US manufacturing requirement would apply to owners of subject inventions, and not just to a limited form of exclusive licensing. There’s nothing in Bayh-Dole requiring *non-exclusive* licensees to manufacture substantially in the U.S. So, if you own a subject invention, you can avoid the US manufacturing requirement by doing just one non-exclusive license and waiting a decade to see if you need to do another. Or do two non-exclusive licenses–one to a company that wants an exclusive commercial position and one to a company that, say, does research or, ahem, likes to collect licenses. Just because a company gets a license does not mean they have to *act on it.* The other non-exclusive license then has the effect of an exclusive commercial position–but isn’t an exclusive license, and so Bayh-Dole’s requirement doesn’t apply. Or a subject invention owner might just license non-exclusively to a bunch of companies, but none of them in the U.S., for sales to the U.S. Bayh-Dole in such a case does not require any U.S. manufacturing. Section 204 ignores all this practice obviousness.
Even with the requirements on those limited exclusive licenses to use or to sell, Section 204 is full of qualifications and ambiguities. Even the restriction on exclusive licenses can be waived by the federal agency that funds the research if a patent owner “shows” that they tried and failed or that they decided it wasn’t worth trying.
Of course, section 204 was drafted by lawyers, so they couldn’t use plain language like that. Let’s take a look at their work.
The Second Part: Waiver
Now for the major walk-back in section 204. For all that–the prohibition on exclusive licenses in the United States, except that yes, one can grant exclusive licenses in the United States but only if the licensee agrees to an ambiguously worded tie to the domestic manufacture of products–section 204 also authorizes federal agencies to waive the requirement altogether, but to do so the agency must meet at least one of two requirements. Here’s the start of the waiver in 204:
However, in individual cases, the requirement for such an agreement may be waived by the Federal agency under whose funding agreement the invention was made
You might wonder why it is that Bayh-Dole delegates decisions about this most important provision in the law to the federal agency that grants the funding. Why should that federal agency have any particular say in waiving the default requirement?
One might also begin to think that by placing all these provisions into a patent rights clause for each funding agreement and then allowing the granting federal agency to waive any of the public interest apparatus of that patent rights clause, Bayh-Dole lets federal agencies who don’t care for the public interest apparatus simply waive it–Bayh-Dole requires them to use the defaults in their funding agreement (unless they want to go through a heck of a lot of hoo-haw), but does not require the federal agencies to enforce those defaults. You see how it works. NIH attorney Latker drafts Bayh-Dole to make it appear that the public interest is protected, but knows that with this architecture the NIH can quietly choose not to enforce those provisions.
The public interest apparatus was necessary to get the law passed, but Congress foolishly assumed that federal agencies would enforce the required default patent rights clause. But they don’t. What would Bayh-Dole be without any of those public interest provisions? Pretty much a patent monopoly pipeline from federal funding to favorite companies, stripping inventors of rights, no public accountability, no basis for appeal–a multi-billion dollar industry public subsidy.
And to grant a waiver, each agency must have a procedure involving “a showing”:
upon a showing by the small business firm, nonprofit organization, or assignee
Of either of two conditions. First, “we tried”:
that reasonable but unsuccessful efforts have been made to grant licenses on similar terms to potential licensees that would be likely to manufacture substantially in the United States or
Second, “we didn’t bother trying because”:
that under the circumstances domestic manufacture is not commercially feasible.
These conditions are remarkably odd. In the first condition, what is that “likely” doing?
to potential licensees that would be likely to manufacture substantially in the United States
We might expect that if the owner of a patent is to grant an exclusive license to use or sell in the United States, the licensee must agree to manufacture the product substantially in the United States. There is no likely about it. Section 204’s waiver standard is significantly lower than the licensing standard. To invoke the waiver, a federal agency does not have to see documented that a potential licensee agreed to source product manufactured in the United States but otherwise rejected the terms of the license. The patent owner does not have to try that hard. The “efforts” to grant a license “on similar terms” to potential licensees could mean simply announcing an invention is “available for licensing” without any company showing up to take an exclusive license. What constitutes “reasonable” efforts? In whose judgment is an effort “reasonable”? Here, it would appear to be the federal agency that would determine “reasonable”–but then the law ends up being whatever a federal agency decides, and given different federal agencies may make very different determinations, the law becomes anything but “uniform.”
Put it another way. Section 204’s waiver may operate if a patent owner offers an exclusive license to use or sell in the United States and no company agrees to source America-made product for that using or selling. That could be because the patent owner does not make American-made product and has not licensed anyone in America to make product. Or it could be that the patent owner offers an exclusive license to make and to sell (say), and no company that would manufacture in the United States is willing to take the offer–for whatever reason.
We might redraft the provision to make this clear: “that reasonable but unsuccessful efforts have been made to grant licenses . . . to potential licensees that otherwise would agree to use or sell product manufactured substantially in the United States.” Then this waiver language would make clear that the problem is not the agreement to use product manufactured substantially in the United States–it is the on similar terms. There is no requirement that the offered terms are “reasonable” or promote “free competition and enterprise.”
Perhaps you now see the shape of the waiver requirement: the agency can grant a waiver for a licensee (i) who has not agreed to manufacture substantially in the United States but (ii) is willing to take terms that potential licensees who would likely manufacture in the United States won’t accept (or are–thought experiment here–unlikely to accept). All one has to do to qualify for the waiver is to make the starting offer beyond what a United States licensee would be likely to accept, perhaps because the cost of manufacturing or working conditions or regulatory compliance–but a company manufacturing in another country may readily bypass these costs. The “on similar terms” removes the teeth from the condition, since it makes it easy for the owner of a patent on a subject invention to be unsuccessful in finding an exclusive licensee who will agree to use or sell (“substantially”) domestic manufactured product.
There is nothing in this first condition that requires the “terms” on offer to be “reasonable.” The “reasonable” is placed to modify “efforts”: “We tried reasonably to get a company that was likely to manufacture substantially in the United States to accept our unreasonable terms, but gosh, they just wouldn’t–but here’s a company that manufactures in another country and can afford those terms–cheaper labor, government subsidies, taking environmental shortcuts, crappy working conditions–hey, they sure look reasonable.” Or, even more simply, “We found a foreign company willing to pay more than any American company would pay–so it is unlikely that any American company would bid for this license on similar terms.” Consider: a foreign company is so eager to preclude American manufacture that it is willing to accept outrageous terms (what a university would call “favorable” terms). As long as no American manufacturer is willing to accept those outrageous terms–similar terms–the waiver condition goes true. (And the company obtaining the license can always renegotiate the terms later without re-visiting the waiver–yet another workaround). All the waiver condition states, then, is that foreign companies may outbid domestic ones, and if so a federal agency may grant a waiver to 204 compliance. That’s some preference for U.S. manufacturing, now, isn’t it?
In the second condition of the waiver clause, no efforts are required at all–just an argument that “domestic manufacture” (not even substantially manufactured) is not “commercially” feasible. What might “commercially feasible” mean? Perhaps “commercially feasible” means “with an acceptable profit.” Put loosely–“we don’t think any American company will see enough value in manufacturing product for us that it is even worth our time to try to find one. It’s just not commercially feasible.” Obviously, if foreign companies are happy to manufacture product, then the manufacturing is of course commercially feasible. The problem has to be that companies willing to manufacture are not willing to build the manufacturing facilities in America or shift work to existing facilities in America. But this is the condition that the authors of Bayh-Dole were expressly addressing–make that federally supported invention with United States industry and labor. And it is exactly here that Bayh-Dole offers a waiver of the requirement.
Again, the point of the provision is to meet the policy objective of Bayh-Dole to use the patent system to promote “to promote the commercialization and public availability of inventions made in the United States by United States industry and labor.” But here Bayh-Dole walks back that objective by reversing it–if an owner of a subject invention makes a “showing” that United States manufacturing of product is not “commercially feasible,” the owner doesn’t even have to make an effort to find licensees.
Between these two waiver conditions, there’s essentially no substance to the substantially manufacture requirement for exclusive licenses to use or to sell. Sure, a patent owner could grant a license to manufacture only to a company that does manufacture in the United States–as well as licenses to others to manufacture in other countries. But there’s nothing in section 204 that requires any patent owner to manufacture in the United States or require any licensee of the right to “make” to manufacture in the United States. Isn’t that just weird? One might expect that section 204 might require that for any product based on an invention made with federal support to be used or sold in the United States, the patent owner must make the product in the United States or must have a standing offer to license for manufacture in the United States on fair, reasonable, and non-discriminatory terms. That is, if there’s product used or sold in the United States, there must be at least one United States manufacturer–either the patent owner or a licensee, or at least an available right to obtain a license.
204’s Empty Gesture
The point of the preference for United States industry in Bayh-Dole is to use the patent system “to promote the commercialization and public availability of inventions made in the United States by United States industry and labor.” This part of the law directly addressed Senator Bayh’s argument that the US was falling behind other countries in innovation and Bayh-Dole would make the difference. Yet at just this crucial point, the most important point of Bayh-Dole, the part that takes precedence over all other parts of Bayh-Dole, over all other parts of federal patent policy (perhaps even including Stevenson-Wydler), Bayh-Dole wilts way.
Instead of requiring US manufacture of subject inventions, Bayh-Dole narrows the requirement to exclusive licenses to use or to sell, and within that narrow scope, further narrows the requirement to “substantially” manufacture, and if that weren’t enough, gives federal agencies the authority to waive the requirement altogether on a determination of “we tried and failed” or “we talked ourselves out of trying” where “tried and failed” can mean no more than “we got a deal with a company who will source product overseas and no American manufacturer would accept that deal on similar terms.”
And if that isn’t enough, Bayh-Dole then anchors the requirement in a mere agreement to comply, offers no procedure by which agreements are documented or reported, places no responsibility on the patent owner to enforce the agreement or report non-compliance, and buries federal enforcement of section 204 compliance in a march-in procedure (section 203(a)(4)) that was designed to be so burdensome that it has never been successfully used in the history of the law, and the remedy is generally to compel additional licensing, which then removes the obligation to substantially manufacture in the United States. If the march-in licenses are to use or to sell, then there is no longer an exclusive license to use or to sell in the United States, and section 204 no longer applies. If the additional licenses are to manufacture in the United States, but not to use or to sell, then the march-in doesn’t address the problem that an exclusive licensee to use or to sell is not using American-made product. And what is the good of a license to manufacture in the United States without also having the right to use or to sell? It would appear that march-in here is an entirely empty gesture.
Section 204 is a wonderful flourish. The most important provision in Bayh-Dole. And it is an empty mess of a gesture that does nothing where it ought, merely creates uncertainty and bureaucracy, and does absolutely nothing to support American industry and labor. It is a microcosm of Bayh-Dole itself. Bayh-Dole is a political machine to convey monopoly patent rights to pharmaceutical companies under the cover of a general policy promoting the use of inventions and the virtue of involving universities and other nonprofits as the complicit partners. The utter lack of substance in section 204 is just more evidence for the only meaningful purpose of Bayh-Dole.
[To show you just how empty section 204 is, no one ever reports the net number of new U.S. manufacturing jobs created through the use of section 204’s U.S. manufacturing requirement, nor the number of new products manufactured exclusively in the U.S. Not universities, not AUTM, not the federal government. Section 204 on its own express claim is the most important provision of Bayh-Dole, and no one bothers to report anything beneficial that happens when section 204 is not waived. If section 204 is empty, the claimed primary rationale for Bayh-Dole–revitalization of American technology leadership–is empty.]
[Hey–if you are reading this article for a class and have questions, feel free to send me an email. Glad to discuss how Bayh-Dole works and doesn’t work in practice as well as in policy.]