Bayh-Dole–enforce or repeal the patent rights clause?

Bayh-Dole is such a strange law. Consider:

After 35 years, there is no public data supporting the claim that the Bayh-Dole Act has been successful. Bayh-Dole makes that data a government secret. Universities don’t report it. AUTM licensing surveys do not report it and don’t even break out government funded inventions. What gets reported is proxy metrics that don’t address Bayh-Dole’s requirement for the use of inventions arising in federally supported research. These metrics, even when bloated for effect, represent activity, not outcomes.

Bayh-Dole does not operate. Universities refuse to comply with the standard patent rights clause in most of its substantial elements; federal agencies are allowed to waive many public protections, and beyond that, federal agencies routinely do not enforce the patent rights clause and have never “marched-in” in response to nonuse or unreasonable use; and furthermore, federal agencies routinely make no use of their non-exclusive royalty-free license to make, use, and sell (and have made, have used, and have sold) subject inventions. And beyond that, universities ignore Bayh-Dole’s policy limitations on patent property rights in subject inventions–they use their patents to exclude use rather than promote it, and to suppress free competition and enterprise.

Whatever data anyone might produce claiming Bayh-Dole is a success (or failure!) has be taken with a pint of beer. If Bayh-Dole doesn’t operate, then the data is about something else, not Bayh-Dole.

In particular, universities routinely

  • do not comply with the (f)(2) written agreement;
  • misconstrue what constitutes a subject invention;
  • misrepresent their rights to inventions made with federal support;
  • do not prefer small businesses in licensing;
  • assign inventions without getting federal approval; and
  • misspend royalties and income relating to subject inventions.

All this on top of problems with bureaucratic requirements–failure to timely report or notify the government of decisions to elect to retain title or provide updates of publications or other potential bars to patenting or place a notice of government rights in patent applications.

If we want to see what Bayh-Dole does, federal agencies must require university compliance with the standard patent rights clause, must not routinely waive requirements, must march-in on all nonuse and unreasonable use, and must act on the license rights Bayh-Dole provides for them. Furthermore, federal agencies must require annual reports of invention use and data from these reports must be made public, redacting only that information that meets FOIA’s requirements for exclusion of trade secret, commercial, and financial information that is confidential or privileged–but it cannot be trade secret or confidential that an invention is not being used or that it has been licensed exclusively and still is not being used.

If universities comply with Bayh-Dole’s standard patent rights clause, then

they will stop claiming ownership of inventions as a condition of federal funding or any university resources made available as a condition of that funding–that’s the effect of the (f)(2) requirement, which is drafted to pre-empt such claims, just as the standard patent rights clause forbids such claims in other forms of subcontracting.

they will stop constructing “exclusive licenses” that assign subject inventions. They will limit licenses to less than all substantial rights in inventions. Specifically, they may grant an exclusive license to sell, but will grant only a non-exclusive right to make and use, permitting broad access to publicly funded inventions by the research community and by the public and industry that has no need to wait for a commercial version of the invention in order to make and use and benefit from it.

any university or nonprofit exclusive license for all substantial rights in a subject invention will also carry the nonprofit standard patent rights clause–requiring the company to devote all income relating to the subject invention, after allowable expenses incidental to the management of subject inventions, to scientific research or education. Most companies will reject this condition of Bayh-Dole and will request a license to less than all substantial rights.

There are two obvious pathways to address this situation, among other pathways.

1. Repeal the non-compliant parts of Bayh-Dole. If people are happy with present practice, which mostly ignores Bayh-Dole while touting the wonderfulness of Bayh-Dole, then repeal the parts of Bayh-Dole that universities and federal agencies ignore. Bring the law into conformity with non-compliant practice. Declare that federally supported inventions are no different from any other inventions. Declare that the federal government needs no non-exclusive license–it can get a shop right if it operates the shop, and it can have an equitable right if it provides the specification and resources–it doesn’t need any special laws or regulations for such stuff. Besides, the government almost never (almost never) uses its license. The government can infringe any time it wants, anyway, and a patent owner can file a claim for reasonable compensation in the Court of Federal Claims. So just do that. And there need be no extra special “protections” for the public–declare that the patent system and antitrust law are sufficient. Trust small business firms and nonprofits (indeed, all companies) to do the right thing, the reasonable thing. And there needs to be no public accounting–declare that public accounting is a waste of time and effort and a burden on innovators, especially if the government never marches in and doesn’t report the data to inform public debate regarding federal invention policy.

The repeal would involve 35 USC 200, 35 USC 202(c)(1)-(8) but for, say, (6)–the requirement that the contractor include a federal funding notice in patent applications–and 35 USC 203 (march-in, never used), 204 (American manufacture, so narrow its nonsense), and 205 (confidentiality of invention disclosures and patent applications provided to the government). There would be no need for march-in because patent law takes care of things. No special protectionism involving US manufacture, as trade laws and policy manage that. And there’s no point to nonprofit preference for small business–it’s really quite silly. License to a small company that sells to a big company? Easy work-around. And there’s no special need for government confidentiality because the federal agencies won’t receive copies of invention disclosures or patent applications. And no policy restrictions on what a patent owner might do–no working requirement, no pro-competition requirement. Let patent owners sit on inventions, or troll industry with them, or work monopoly deals short of antitrust–in short, release the krakken.

Bayh-Dole could then be reduced to the following bits:

Each nonprofit organization or small business firm may elect to retain title that it acquires to any invention arising in federally supported research or development; Provided, however, That a funding agreement may provide otherwise, following approved procedures.

A nonprofit organization or small business firm, or any assignee thereof, shall include within the specification of any United States patent application and any patent issuing thereon a statement specifying that the invention was made with Government support.

That should do it. I’ll leave off any discussion of the part of Bayh-Dole dealing with inventions owned by the federal government. With this change, most of the standard patent rights clause at 37 CFR 401.14(a) drops away. We are left with:

The contractor may retain the entire right, title, and interest throughout the world to each invention that the contractor acquires and which was made in performance of work under this contract, provided that the contractor shall include, within the specification of any United States patent application and any patent issuing thereon covering any such invention, the following statement, “This invention was made with government support under (identify the contract) awarded by (identify the Federal agency).”

I know, adding the government funding notice does nothing for innovation, but it does create an exchange of sorts–the government agrees not to claim inventions as deliverables, and the contractor agrees to acknowledge the government’s financial support–as any reasonable contractor ought to do anyway. In reality, if inventions made with federal support are just ordinary inventions, all Bayh-Dole has to do is prevent federal agencies from asserting any interest in inventions as a condition of federal funding, unless there are extraordinary circumstances (such as, say, work on naval nuclear propulsion systems). For that, Bayh-Dole would simply forbid federal agencies from using a patent rights clause in any federal funding agreement unless extraordinary circumstances were determined to require a patent rights clause.

Then Bayh-Dole would read, adapting 35 USC 202(a):

No Federal agency shall include in any funding agreement any provision securing for  the Government any interest in any subject invention; Provided, however, That a funding agreement may provide otherwise . . . [list of exceptions].

No need for policy limiting patent property rights; a few definitions remain; no bothersome standard patent rights clause; no worthless paperwork; no fuss about march-in, confidentiality for unnecessary reports, no worry about antitrust law. Still assert precedence and exclude educational grants. Leave federal patent practice alone for now. That is, exactly the way universities and federal agencies practice under the law now. 

And the second obvious pathway:

2. Enforce Bayh-Dole’s patent rights clause, restrict waivers of compliance with the standard patent rights clause, report all waivers granted, and act on the government’s license for all commercial products based on subject inventions and used by or on behalf of the government. If folks persist in the idea that inventions made with federal support are not ordinary inventions, and that the government should have rights in these inventions (and should intend to use those rights), and that the public should be protected from nonuse and unreasonable use of the patent system with regard to these out-of-the-ordinary inventions, and the patent property rights in inventions made with federal support should be limited, and there should be a public accounting of the use of such inventions and of the use of royalties and income relating to these inventions, then Bayh-Dole’s standard patent rights clause should be enforced, waivers of the provisions of the standard patent rights clause should be restricted and made public, and the owners of patents on subject inventions held to account.

Enforcing these things does not require new legislation, though new legislation might clarify, simplify, and motivate–most everything can be done by executive order, since it is the action of federal agencies in response to the standard patent rights clause that must be changed. But it’s important that those involved understand the Bayh-Dole Act as it exists, and aren’t just more advocates of the faux Bayh-Dole, ready to twist the statutory language so that it makes an appearance of public oversight and accountability but doesn’t really do any such thing.

Two areas, especially, require clarity.

(1) The standard patent rights clause (f)(2) written agreement provision requires contractors to subcontract, in effect, with certain of their employees to delegate to those employees the right to establish the federal government’s rights in inventions made with federal support. That written agreement does multiple things: it makes inventors parties to the funding agreement, so that the inventions they make are subject inventions and come with the various provisions of the standard patent rights clause; it causes the contractors to set aside their own claims on inventions made with federal support as a condition of the federal funding in favor of the agreement they agree by federal contract to require of their inventor employees; and it establishes that inventor employees are “small business contractors” if they do not assign their inventions.

An executive order should make clear that under the (f)(2) agreement, universities and nonprofits have no right to subject inventions, just as they do not for subcontracting under section (g) of the standard patent rights clause. Universities and nonprofits should obtain rights to subject inventions only by a voluntary agreement with the inventors. Only in those circumstances in which a university or nonprofit operates a contract research program, consistent with business contracting–assigning and controlling the work–should the employer be permitted to have a patent agreement that is not displaced by the (f)(2) requirement.

(2) It must be made clear that Bayh-Dole does not exclude all invention use reports from FOIA disclosure. Bayh-Dole excludes only what FOIA permits to be excluded–and thus only those portions of reports that involve trade secret, commercial information, or financial information that is confidential or privileged is excludable from FOIA. Bayh-Dole cannot be interpreted to declare that all use reports must be treated as if they are confidential and privileged even when they are not confidential and privileged. In particular, use reports information provided by instruments of state government (such as public universities) must be fully disclosable.

With inventor employees free to manage their inventions without the intrusion of institutional claims, we get the full benefit of research conducted in the context of academic freedom–the “free play of free intellects” as Vannevar Bush called it. That alone is a great liberating element of Bayh-Dole that must come into play to protect inventors from the crush of institutional claims, procedures, requirements, and capabilities.

With public information regarding the use of subject inventions, we will be in a position to understand how Bayh-Dole actually operates, and consider what if anything needs to be done to improve outcomes.

Messing with innovation is difficult for institutional thinkers, who tend toward institutional control and power, and have rational arguments for the resources available to institutions and their claim to be positioned to put the public interest first. But innovation is often not so respectful of institutional needs, roadmaps, timing, resources, or reputation. It is all too easy for institutional thinkings to conflate their loyalty to the institution with their service to the public and their support of innovation. If only we own the invention, they reason, we can serve the public and make money and build fame for our institution. Innovation will have none of that. Innovation finds ways around institutional policies and practices, outclevers institutional attorneys, leaves primarily the helpless, the foolish, and the desperate to work within institutional boundaries–and often is the reason that inventive folks become helpless, foolish, and desperate.

There are other things that can and should be done with federal invention policy. Procurement from commercial firms should be separated from the operation of government labs by contractors, and these activities should be separated as well from grants to promote research deemed to be in the public interest at nonprofits and universities. An arbitrary policy serves no-one well. An arbitrary policy set up to capture patent monopolies for the pharmaceutical industry should be broken out and established on its own merits and not be hidden in abstractions that make it appear all agencies, all agreements, all purposes, and all subject matter must benefit from these same procedures.

These matters, too, can be addressed within the existing framework of Bayh-Dole, as “exceptional circumstances.” We have to recognize that “exceptional circumstances” are not “rare” circumstances but simply those that lie outside of the requirements set down by Bayh-Dole for a standard patent rights clause. As it is, there are four standard patent rights clauses. There should be at least three more.

One for university grants led by faculty investigators.

One for procurement from commercial concerns and nonprofits who choose to participate for use by government.

One for research and development directed at public health and potentially involving the development of commercial products.

For decades, policy makers were adamant that a single federal patent policy was not appropriate and would not work. Even those advocating for the government to grant some exclusive licenses or allow contractors to own inventions (and grant exclusive licenses) did so with the argument that such freedom would be exercised rarely and only in the public interest–you know, when no public benefit was possible from an invention other than through the development of a commercial product that the government refused to fund and which was otherwise so expensive to develop and so easy to copy once developed that only wealthy investors might be attracted to provide the risk capital to undertake the effort, and they would only participate if assured of a patent monopoly. That special case.

And then, somehow, with Bayh-Dole, everything went flip-flop without so much as a meaningful discussion. A “uniform” policy was necessary, was just the thing. Patents on federally supported inventions were just ordinary patents, obtained at the expense of a bunch of bureaucratic procedures, most of which were just for show and ignored. Perhaps three and a half decades of this “experiment” is enough and we can get back to sensible government practice. If we divide up the territory, we can have a focused discussion on each proposed patent rights clause.

For instance, if it is really so important that pharma be provided with government-subsidized patent monopolies, then why not just have the government nominate for each university the big company that will own any inventions relating to public health, along with the negotiation on indirect costs? Or let the companies bid to be each research project’s patent buddy? Why waste everyone’s time with gawd-awful university patenting and licensing, when it is clear that exclusive licensing of inventions performs much more poorly than does ownership of inventions by capable companies?

And if it is not so important that pharma be provided with such patent monopolies, then let’s have the public discussion about how the government should encourage other approaches to the use and development of inventions made in the public interest and directed at public health. Open innovation, commons, libraries, cumulative technology, consortia, standards–there are plenty of options that are viable alternatives to commercial product development.

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