NIST’s Chief Counsel on Bayh-Dole, 4

One last slide from NIST’s chief counsel’s talk from 2013. Much to discuss.

How to unwind this assertion? The Bayh-Dole Act requires federal agencies to use an arbitrary default patent rights clause. In the absence of Bayh-Dole, executive branch patent policy permitted nonprofits and small businesses to deal in patents on inventions made with federal funding. Bayh-Dole did not enable anything new that way. Bayh-Dole changed the process of making a determination whether, for any particular invention, that it was in the public interest for a nonprofit to deal in patent monopolies.

Before, the nonprofit had to make a case. After, the nonprofit had an arbitrary right to keep ownership, and federal agencies could (but are not required to) act to correct any of the nonprofit’s failings.

Before, a federal agency could wait to see what was invented before making a determination. A given grant could give rise to very different sorts of invention. Some inventions might be better served as standards rather than as enticements for speculative investors; some inventions might be better managed as research tools available to all rather than given over to any one organization to exploit for profit; some inventions may only be meaningful if a number of other inventions are also always available and on common terms, rather than permitting ten or twenty or thirty nonprofits to each claim a tiny portion of the whole and attempt to license that tiny portion exclusively. After, federal agencies must declare an alternative patent rights clause before an award is even granted, and face a lengthy review and appeals process if they try this approach.

For all that, the effect of Bayh-Dole has not been to permit nonprofits and small businesses to “out-license” inventions. The point is that they are able to “out-license” exclusive rights because they can deal in patents. Small businesses, we might note–and now all businesses–have no particular reason to “out-license” anything. They can exploit the inventions they acquire for their business purposes. And certainly companies can choose to license an invention that has been acquired for a business purpose exclusively to another company, but in the general case, this is a rare thing. A more expectable use of licensing by companies is cross-licensing or licensing non-exclusively within a standard or licensing within joint ventures or strategic research partnerships–that is, not exclusive licensing.

As for non-profits “out-licensing” inventions, consider that prior to Bayh-Dole, many non-profits could license out these same inventions under the NIH and NSF IPA programs–and did a reasonably poor job of it. When, by contrast, the federal government obtained ownership of inventions made in federally supported research, it made nearly all of these inventions available broadly for domestic use. It out-licensed nearly everything, whether by dedicating the invention to the public domain or acquiring a patent on the invention and then acquiescing in use without the bother of formal licensing. Any federal exclusive license in effect restricted public access in favor of a chosen company. That was rare–and makes sense.

By contrast, Bayh-Dole arbitrarily gives nonprofits the right to deal in patent monopolies for any inventions made with federal support that they can acquire. There’s no requirement that the nonprofit have a decent patent policy or treat its inventors nicely or have a history of effective practices and publicly beneficial outcomes. Sure, there’s an argument that a nonprofit might do a better job identifying companies that ought to want a given invention–but even that argument falls flat. Even if that’s the case, there’s no need for the nonprofit to hold monopoly patent rights if the companies involved could just as readily get non-exclusive rights from the federal government, often without any formal paperwork. The nonprofit could offer consulting services, say, to generate income, but would have no need to “out-license” as well. It’s only in the case of “out-licensing” a monopoly right that Bayh-Dole changes the situation.

Bayh-Dole makes it difficult for federal agencies to decide what is in the public interest for the results of the research they fund, placing the default with nonprofits regardless of their performance, and changing out making most inventions freely available (via the federal government) for withholding nearly all inventions for hoped-for exploitation by single, favored companies (which, when they do obtain their exclusive license, generally fail to do significant exploiting). That’s a true highlight! Why don’t the advocates of Bayh-Dole want to point it out?

You have read all this way to get to this fundamental misstatement of Bayh-Dole. Bayh-Dole creates no presumption of ownership by the contractor. The Supreme Court was adamant. There is no presumption, no vesting, no special privilege or right, no mandate. Priority of claims between the government and a contractor, after the contractor has acquired ownership. Nothing more.

A twisted reading “presumption of ownership” might be “presumption that a contractor that goes out of its way to acquire ownership might want to keep that ownership and not wait in a state of uncertainty while a federal agency determines whether it is in the public’s interest for the nonprofit to deal in patent monopolies, motivated by the hope of financial return.” But that’s twisted.

The clear reading of Wixon’s point here is that he apparently doesn’t know Stanford v Roche–the single most important interpretation of Bayh-Dole’s management of ownershipand is more than happy to repeat political talking points about Bayh-Dole that floated around before the Supreme Court made clear what anyone who read the law carefully would know. Bayh-Dole cannot possibly be a vesting statute, does nothing with regard to “presumption of ownership.” Presumption does not show up in Bayh-Dole. “Presumption” of ownership is what the charlatans promoting Bayh-Dole used to make it appear that Bayh-Dole stripped inventors of their invention rights and handed those rights to organizations whose only merit is that they handled the federal funds allocated for projects proposed, for the most part, by university faculty.

We might observe that the Constitution authorizes the federal government to secure for inventors exclusive rights to their inventions for limited times. There’s nothing that gives the federal government the authority to secure exclusive rights for its favorite universities rather than for inventors. Bayh-Dole is part of federal patent law. How could it possibly be that Bayh-Dole could reserve for universities ownership of inventions made with federal support? The Supreme Court observed that had Congress intended such a “sea change,” “it would have said so clearly.” But had Congress said so clearly, we might add, it would have acted outside its Constitutional authority. Whether patent law vests ownership directly with a university, or requires inventors to assign to the university, or requires a university to require inventors to assign to the university–it’s all of the same thing: the force of federal patent law to place ownership of inventions made with federal support somewhere other than with the inventor.

We may distinguish this line of argument from the situation in which the federal government contracts with investigators to receive any inventions they make as deliverables. Such an arrangement is not a matter of federal patent law. The government contracts with the inventor to acquire the personal property of the inventor in exchange for acceptable consideration. Even those statutes that provide for the ownership of inventions made with federal support to vest in the federal government can be understood to mean that the government limits how it will issue patents to inventors of technology the government intends to control, just as the government may do for reasons of national security.

NIST’s chief counsel does not indicate what starts those clocks ticking. That’s the disclosure of a subject invention by the inventors to personnel designated by the initial contractor as responsible for patent matters. A subject invention is one that the contractor already owns. There are three possibilities:

  1. Voluntary reporting. The inventor reports an invention made with federal support to the initial contractor–so far, the invention is not a subject invention, and the inventor reports it voluntarily, or is required to report on some basis other than the patent rights clause in a funding agreement; if the initial contractor acquires the invention, then it becomes a subject invention, has been disclosed already, and the clock apparently starts at the point that the invention becomes a subject invention.
  2. Equitable title. The initial contractor through its actions acquires equitable title in an invention made with federal support–the initial contractor may have hired an employee to invent, or directed an employee to engage in experimentation, or the employee is a senior officer with a fiduciary duty to the contractor; thus, the invention is owned by the contractor, although formal title is not perfected until assignment, and so the inventor must report the invention as a subject invention and afterwards is required to assign. The clock starts when the invention is disclosed, since the initial contractor already has equitable title.
  3. Inventor subject invention. The initial contractor complies with the requirement at (f)(2) of the standard patent rights clause, requiring each inventor to make a written agreement to protect the government’s interest and thus become a party to the funding agreement and therefore a contractor; any invention made then becomes a subject invention, which the inventor has agreed to report, but which is not owned by the initial contractor unless the contractor has some claim entirely outside Bayh-Dole and the patent rights clause of the funding agreement under which the invention has been made. The clock starts when the invention is disclosed to the initial contractor, but the right to elect to retain title is with the inventor under the inventor patent rights clause.

When an initial contractor subcontracts its own inventors as required by the standard patent rights clause, it creates a federal contractual obligation for both itself and its employees. The initial contractor cannot have a claim on inventions those employees make outside of Bayh-Dole and at the same time subcontract with those employees to have the right to establish the government’s rights in those same inventions. You see the problem. The written agreement requirement forces the initial contractor to give its inventors the benefit of the inventor patent rights clause–unless the initial contractor has the right to hire and/or assign employees to the work that’s federally supported (and thus gain equitable title).

This is a point that doesn’t come up often, but only because those that talk most about Bayh-Dole want to suppress it. Why does it matter? Bayh-Dole was created specifically to deal with university-based inventions. There was virtually no small business contracting for research. The SBIR program had not been launched. There were other nonprofits doing federal research, but many of those were contract research operations that had no problem giving up ownership of federally supported inventions, since they made their money on the research contracting itself.

The focus was the university, and universities operated in two research modes. One mode was the research institute, in which employees were hired to conduct research, directed by the university. The other mode was faculty-led research, which could be department based (without a sponsor, but perhaps using specially allocated university resources) or “extramural” (“outside the walls,” with a sponsor, but outside a faculty member’s university duties). For faculty-led research, universities then and now assure the faculty of the freedom of research and freedom to publish. Universities disavow any institutional right to direct faculty research or control publication.

Thus, university faculty, unless they voluntarily give up their freedom of research, cannot be assigned or directed to conduct specific research or to invent for the university. The circumstances cannot create a claim of equitable title for the university that hosts their research. With the written agreement subcontract in place, a university gives up any prior contractual claim it might have to require faculty to assign inventions made in faculty-led research. When university faculty invent in a project with federal support, they are subject to the inventor patent rights clause, not to the nonprofit patent rights clause.

This result may be distinguished from the outcome for inventors working for small businesses (and by executive order extension, all businesses). In the case of businesses, employers do not assure employees freedom of research or publication, and do have the right to assign and direct (and approve and suppress the publication of) employees’ research work. Thus, the analysis that takes into account faculty privilege does not arise, and the conventional patent agreement is not displaced by the company’s compliance with the written agreement requirement, and in the absence of any such conventional patent agreement, the company may still claim equitable title based on the circumstances of invention.

This result that’s favorable for university faculty ownership of subject inventions is confirmed by the patent rights clauses. In the case of inventor ownership, the patent rights clause does not expressly require the inventor to file any patent application, nor does the government have the right to acquire any invention for which the inventor does not seek a patent. This result is entirely consistent with faculty freedom to publish. A patent system creates an exchange: an inventor receives an exclusive right for a limited time in exchange for fully publishing the technical details of the invention. A patent is a publication that carries the inventor’s name. Forcing faculty to publish in the patent literature violates their freedom to publish. While a company does not offer such a freedom to its employees, generally, universities do, for faculty members, who are not ordinary employees.

This understanding of Bayh-Dole is anticipated as well in other areas of the law. Look at this wording in the march-in provision (35 USC 203):

With respect to any subject invention in which a small business firm or nonprofit organization has acquired title under this chapter,

A subject invention is one that a contractor already owns. It would be impossible for a contractor to then acquire a subject invention “under this chapter” [i.e., Chapter 18, where Bayh-Dole has been placed] unless the invention was already a subject invention before the contractor acquired it. The only way it could be a subject invention before the contractor [the initial contractor] acquired it was that it is owned by the inventor, subject to the inventor patent rights clause. Otherwise, there would be no reason at all for the “in which a small business firm or nonprofit organization has acquired title under this chapter.” All subject inventions would necessarily be owned by the initial contractor. But here section 203 differentiates between subject inventions generally and those that have been acquired by a small business firm or nonprofit organization. So a distinction is intended–and sure enough, there is one. If no distinction were possible, the wording of section 203 could easily be:

“With respect to any subject invention for which a contractor has elected to retain title . . . “

All this could be said simply, were it not for the widespread misinformation regarding Bayh-Dole and its patent rights clauses:

Bayh-Dole respects the academic freedom of university faculty when conducting research supported by federal funds. Faculty are assured of ownership of any inventions arising in such research, and so long as they report and choose to retain title to these inventions, they are not required to assign their inventions or to file patent applications.

But this, apparently, is not a highlight of Bayh-Dole–even though it is, and perhaps is the most intelligent thing in the law, buried in the implementation, made necessary by the basic logic of the statute and the effect of the standard patent rights clause. To adopt the rhetorical form of the Supreme Court, had Congress wanted to repudiate the value of academic freedom in a law directed almost entirely at university faculty inventions, it would have done so clearly and not with innuendo and an idiosyncratic use of the definition of “subject invention.”

Our chief counsel adds one additional claim in his depiction of Bayh-Dole’s clocks:

While it is true that if a contractor fails to disclose an invention, or does not choose to keep title, or file a patent application (except for inventor-contractors), the government has a right to request title (“loss of rights”). As Wixon’s wording makes evident, federal agencies do not have to request title, even for these acts of non-compliance. For instance, in the Stanford v Roche case, Stanford apparently missed its election of title “clock” deadline, but the NIH didn’t bother to act on the breach.

But Wixon adds that there’s a “clock” “ticking” on report of commercialization progress. That’s not the case. First, re[porting is on the “utilization” of inventions (35 USC 202(c)(5)):

require periodic reporting on the utilization or efforts at obtaining utilization that are being made by the contractor or his licensees or assignees

(Note as well how Bayh-Dole fails here to use “practical application”–part of the great disconnect in the statute). Utilization is not commercialization. Utilization does not even mean public availability. All that is required is that a federal agency have the right in the standard patent rights clause to request reports on utilization or efforts toward utilization. Not commercialization, not practical application. And there’s nothing in Bayh-Dole or the standard patent rights clause that threatens a contractor with a loss of rights for failing to report on utilization. Failing to report on utilization cannot even trigger a march-in proceeding, though a march-in proceeding for nonuse might force a contractor to gin up some account of the “effective steps” it is taking to achieve “practical application” (which in turn will require some gesture toward utilization).

No clock starts ticking at invention for reporting on utilization. Reporting on utilization comes at federal agency request, later and perhaps never. Reports are on utilization, not merely “commercialization.” The remedy for a failure to report is not “loss of rights.” In a lawyer world, there is a way to twist the chief counsel’s claim into something almost true in some narrow sense. But in reality, no. It’s simply another statement that misrepresents the law.

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