Reagan’s Bayh-Dole cluster bungle

In 1987, President Reagan signed an executive order (12591) that aimed to legitimize his 1983 memorandum that instructed heads of executive branch departments and agencies to “promote the commercialization . . . of patentable results of federally funded research by granting to all contractors, regardless of size, the title to patents made in whole or in part with Federal funds, in exchange for royalty-free use by or on behalf of the government.”

This 1987 executive order is nothing like Bayh-Dole. Bayh-Dole has nothing to say about promoting commercialization. Bayh-Dole does not grant title to patents to contractors. Bayh-Dole does not even grant title to inventions to contractors. Bayh-Dole’s scope is inventions arising from federally supported research or development–not just research, and not just “results of” but “arising from.” Bayh-Dole requires a government license to practice and have practiced, not merely use–much broader.

It’s even hard to understand how an executive order to agency heads could result in federal patent law being superseded, so that somehow title to patents could be granted by the federal government to contractors rather than to inventors, without the active concurrence of the inventors. More than that, it is hard to understand how Reagan could by an executive order overturn Bayh-Dole. Bayh-Dole sets the requirements for standard patent rights clauses used in federal funding agreements. For Reagan’s executive order to be effective, agencies would have to determine, for nonprofits and small companies, exceptional circumstances and substitute Reagan’s requirements in place of the defaults specified by Bayh-Dole.

Determining exceptional circumstances in turn would merely force federal agencies to exploit a pathway available within Bayh-Dole to circumvent Bayh-Dole. To keep Bayh-Dole “uniform,” federal agencies would have to determine exceptional circumstances for *all* circumstances. In other words, Bayh-Dole’s default could not be the default–and could not be used at all, apparently.

Of course, Reagan’s executive order includes the qualification “to the extent permitted by law” and since Bayh-Dole doesn’t permit willy-nilly changes in its defaults, federal agencies would have to go through the determination of exceptional circumstances to substitute Reagan’s contracting requirements for Bayh-Dole’s defaults. Even then, it’s not clear at all how a federal agency has any right–permitted by law–to force title to inventions to contractors. Perhaps federal agencies could be instructed to take title to all patentable inventions as a condition of federal funding and then assign that title to contractors. But that would be a change in executive branch patent policy, and would operate only when Bayh-Dole did not preempt executive branch patent policy.

Since executive branch patent policy allowed–and still does–non-small company contractors to retain title to inventions if they have a commercial market position and capability and the research or development is not directed at public health or things that federal agencies intend to develop to the point of practical application before releasing for public use, Reagan’s executive order has the effect of requiring the federal government to take title to such inventions (from inventors or whomever they may have assigned to) and assign that title to those non-small company contractors, but now without regard for their commercial market position, capability, public health, or agency purposes. Doing so, apparently, will promote commercialization. That’s a puzzle. We are not given a clue what “commercialization” means here. If all it means is “making products to sell” then how does granting title to patents do anything that licensing patents doesn’t do, or for that matter not having patents at all or granting royalty-free, non-exclusive public licenses doesn’t do? If commercialization means rather “investing private funds to make inventions useful that otherwise would not be useful” then we are hung up on the standard for patentability–which includes a demonstration of use. If use is already implicit in patentability, then commercialization must somehow mean development for a profit-making use that’s distinct from the use already necessary for a patent to issue. Thus, perhaps commercialization means “mass production” of an invention. Even if so, why does a contractor have to hold title to patents on the invention? Why should a contractor have a right to prevent non-mass production practice of an invention–such as for research or custom internal non-product use or use as a standard? And even if there’s mass production of one instance of a given claimed invention, why should a contractor be permitted to exclude mass production of a different instance of a given claimed invention? How doe suppressing the commercialization of many other instances of a given claimed invention promote the commercialization of the invention? No, doing so suppresses most commercialization of an invention in favor of the profit-seeking interests of the contractor, which may not even be in a good position to mass produce anything.

No, it’s nonsense–or cover for something else, designed to fool anyone who isn’t in on the trick.

Reagan’s 1983 memoradum attempts to extend Bayh-Dole to non-small company contractors. It also gets Bayh-Dole wrong–“federally funded research and development” contract rather a “federal funding agreement” as defined by Bayh-Dole. The Reagan memorandum requires that “agency policy” with regard to the disposition of inventions “made in the performance” of a federal contract be “the same or substantially the same” as Bayh-Dole. But this could never be. First, Bayh-Dole’s contracting provisions apply only after a contractor has acquired title to an invention “made in the performance of work” under a federal contract. Thus, Reagan’s memorandum is directed at the wrong thing–the invention rather than the subject invention–the invention the contractor has somehow acquired. Second, when Bayh-Dole’s contracting provisions apply, Bayh-Dole preempts other federal law having to do with ownership of inventions made in work receiving federal support. Reagan’s memorandum cannot preempt federal law. Thus, it can never be the case that, if a federal statute provides for federal ownership of inventions made in federally supported work, a federal agency can ignore that federal law simply because there’s a memorandum that tells them to ignore that law. A non-large contractor cannot merely elect to retain title to an invention required by federal law to be owned by the federal government. There has to be a determination that the public intereset will be better served by contractor or inventor ownership of a given invention (and its associated patent rights) than by the federal government making the invention available to all–including to both the inventor and contractor, and that’s something that would not be the case if the contractor has demanded that the inventor assign the invention to the contractor. Reagan’s memorandum purports to do away with the determination of public interest. Federal agencies would not have to require contractors to explain why their exclusive control of inventions made in federally supported work would serve the public interest as a condition of waiving the federal government’s statutory right to take ownership.

What does this mean? For non-small companies, it means that even if there’s a federal statute the requires federal ownership, the executive branch is instructed to refuse to accept that ownership if a contractor otherwise wants ownership. It’s not even the case that the contractor has to first obtain ownership (as is the case under Bayh-Dole). Rather than preempting federal law, which is how Bayh-Dole does it, Reagan’s memorandum requires federal departments and agencies to abandon the “determination of greater rights” for a contractor and with it any interest in how the public interest might be best served. Contractor ownership of inventions is declared to best serve the public interest. This is very weird, in that federal patent law holds that the public interest is best served by inventors owning their inventions, and the US Constitution provides for federal authority to reserve exclusive rights in inventions for inventors, not for companies that happen to contract with the federal government to provide research or development services or (in the case of most university research) to nonpprofits that agree to handle federal money provided to support research projects that are themselves deemed to be conducted to serve the public interest. Under Reagan’s memorandum, the public interest is served when contractors get title to inventors’ inventions. That’s just weird.

Reagan’s memorandum aims to change the Nixon patent policy. But the Nixon patent policy required codification of its provisions. That means that to change those provisions requires an executive order or law, not some stinkin’ presidential memorandum. Thus, Reagan has to come back around to things in 1987 with an executive order–which gets most everything wrong about Bayh-Dole.

What do we get to? Well (to use a Reagan “well”), the Reagan executive order for non-small companies eliminates the determination of greater rights when a contractor acquires ownership of an invention made in federally supported research or development. The “for all contractors regardless of size” part doesn’t work because some contractors are nonprofits, small business firms, or even individuals to be treated as small business firms (for which see 37 CFR 401.9). For these contractors, Bayh-Dole applies. If Reagan’s executive order were to apply, it would deny these contractors their Bayh-Dole rights without the due process of each agency for each funding agreement making a determination of exceptional circumstances. In effect, Reagan’s executive order then also eliminates determination of exceptional circumstance determinations. The executive order itself becomes the only permitted exceptional circumstance.

According to Reagan’s executive order, if a patent is made with federal funds (and what the heck does that mean?), then until the claimed invention is owned by a contractor, that patent is subject to Reagan’s executive order even though it is a patent that is not on a subject invention. Reagan’s executive order requires federal agencies to grant title to the patent to the contractor. For this to happen, the federal agency must take title and then must grant that title back. If a federal agency does not take title, then it has no standing to grant title. It might waive its right to receive title, but that’s not “granting the title.”

Now turn to Stanford’s problem in Stanford v Roche. Stanford has a patent policy that provides that inventors own their inventions unless Stanford is required to take ownership. Stanford allows its inventor-to-be to assign rights in inventions to a company that Stanford allows–indeed assigns–its inventor to visit for nine months to learn PCR. To be able to sue Roche (which acquired the company), Stanford has to find a way to be required to take title to the invention and turns to Bayh-Dole. But that’s all wrong. Stanford doesn’t have ownership of the invention, so the invention is not a subject invention, and therefore (as the Supreme Court ruled), the invention is not a subject invention and Bayh-Dole’s contracting requirements don’t apply. Bayh-Dole does not vest ownership of inventions, does not require contractors to take ownership.

But Reagan’s executive order does require federal agencies to grant title to patents to contractors. There’s no conditional, such as “but only if the contractor wants title.” What if Stanford had demanded that the federal government comply with Reagan’s executive order rather than make up some goofus claim about Bayh-Dole? Then the NIH would have had to demand assignment of the invention–from Roche, not under Bayh-Dole, but under claims made in Reagan’s executive order. To grant title, the NIH would have to take title. And to take title, the NIH would have to have a second patent rights clause in the funding agreement–one that applied whenever an invention was not a subject invention. No doubt Roche would point out that while Reagan’s executive order required federal agencies to do various things, for a requirement to be imposed on contractors, those agencies would still actually have to do what they were required to do. The executive order does not act on its own but requires agency heads to act.

Oh, it’s worse than this. You see, the standard patent rights clause authorized by Bayh-Dole required contractors to require inventors to make a written agreement to protect the government’s interest in inventions (see 37 CFR 401.14(a)(f)(2) in old style citation–take out the (a) for now). That written agreement is where inventors agree to sign paperwork to establish the government’s rights in subject inventions. But that written agreement also becomes part of the patent rights clause, and thus part of the funding agreement itself–and that makes the inventors into contractors, and their inventions necessarily become subject inventions. And if an inventor that’s a contractor assigns a subject invention, then the assignee also becomes a contractor under the funding agreement. As a contractor, then, Roche would have the right to elect to retain title to the assigned invention, but only after Roche’s patent personnel had received a disclosure of the invention and had in turn disclosed that invention to the federal government. If Roche’s patent personnel did not receive a disclosure of the invention, then they had no obligation to disclose the invention to the federal government, and did not yet have any deadline to elect to retain title.

You follow? No, of course. This is a dump of a law all over the place. Nothing needs to be this convoluted. Such convolutions alone are enough to say please no more Bayh-Dole. But onward. Stanford never required its inventor to make the written agreement to protect the government’s interest. Thus, the invention never was a subject invention. Bayh-Dole doesn’t apply. And no vesting even if it did. But also the NIH never complied with the Reagan executive order and so the funding agreement did not have a second patent rights clause applicable to non-subject inventions made in grants to nonprofits, so Stanford’s inventor had no obligation to assign inventions to the NIH and the NIH had not taken any action to create for itself the right under contract to require assignment of the invention so that it could grant title to patents to Stanford. So it was one big cluster of non-compliant handwaving to try to claw back what everyone (but Roche) had bungled.

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