Not a Flying Toy–Extended

This is part of the discussion I posted previously at Techno-L on 11/3/09. I’m putting it up here for the sake of assembling the various takes on this issue. It lays out some of the diverse ways Bayh-Dole could be implemented by universities.

[I have revised text for clarity–after six years, the argument holds up well and the 2011 Stanford v. Roche decision makes it clear that this is the correct argument, despite what the “amigos” might claim. I haven’t heard a mea culpa from them yet–and don’t expect to. It would take a degree of moral fiber to stand up and say, “we sure got it all wrong.” I expect they would rather try to bite my legs off before they confess up. 10/2/15]

Okay. Long post alert. This is important stuff. Check my work. Toy Story bit is at the end.

In 37 CFR 401: “electing title” pertains to the disposition of respective interests in the funding agreement. When a contractor “elects to retain title” by notice to the funding agency, the contractor is neither “claiming ownership” nor “obtaining ownership” of patent rights, but rather is interposing university (contractor) interest ahead of the agency’s own claims (whatever they may be). That is, the contractor is informing the federal agency that the agency cannot require assignment of title to an invention that the contractor owns or intends to own. The notice to elect to retain title is a federal contract action that depends on a condition in the federal funding agreement. Think of it as an agency flow down based on an option: if the inventor assigns or agrees to assign a subject invention to you, Dr. University, then you can keep that invention if you notify the federal agency involved; otherwise, the agency can request assignment from either the inventor or you, depending on who owns the invention when the request is made. The disposition of ownership is not an issue in Bayh-Dole except for whether a federal agency has standing to request assignment to the federal government.

There is no issue in Bayh-Dole about inventor ownership of patent rights in subject inventions. The matter is not even addressed in Bayh-Dole. Invention ownership is a matter of federal patent law. Nothing in Bayh-Dole changes that. As far as patent law is concerned, University inventors own any patent rights in the work they do. What matters is how those personal rights are managed in the context of federal research awards. Bayh-Dole makes the agency-university interface (mostly) uniform, but it does not require agencies all to take the same line with regard to their ownership interest in sponsored inventions. Some agencies pay close attention to whether they can obtain ownership, such as ONR and NASA and DOE, and some agencies appear not to care much, such as NSF.

The standard patent rights clause authorized by Bayh-Dole requires written agreements to protect the government’s interest. This is laid out in section 37 CFR 401.14(a) (f)–headed clearly enough: “Contractor Action to Protect the Government’s Interest” (my bold):

(2) The contractor agrees to require, by written agreement, its employees, other than clerical and nontechnical employees, to disclose promptly in writing to personnel identified as responsible for the administration of patent matters and in a format suggested by the contractor each subject invention made under contract in order that the contractor can comply with the disclosure provisions of paragraph (c), above, and to execute all papers necessary to file patent applications on subject inventions and to establish the government’s rights in the subject inventions. This disclosure format should require, as a minimum, the information required by (c)(1) [disclosure of subject inventions to the government], above. The contractor shall instruct such employees through employee agreements or other suitable educational programs on the importance of reporting inventions in sufficient time to permit the filing of patent applications prior to U.S. or foreign statutory bars.

Simply: “The contractor agrees to require its employees to disclose promptly each subject invention and to execute all papers necessary to file patent applications and to establish the government’s rights.”

Read it for what the law says. Notice that there is no indirect object for disclose or execute–who receives the disclosure? on whose behalf are papers executed for patent applications? and who requests papers to establish the government’s rights? As with any sophisticated text, there are many things potentially implied, not just the one that might come first to mind. The standard patent rights clause is written this way for good reason.

Consider: Nothing in paragraph (f)(2) requires employees to disclose their inventions or assign their patent rights to the contractor or, for that matter, to the government. Why not? Why should not the draft requirement be that the inventors must disclose and assign their inventions to the contractor? Would not that be clear and direct? Actually, no. At the time of Bayh-Dole’s passage, many universities had policies that directed university inventors to disclose their inventions to a research foundation–sometimes a local one, often Research Corporation, and in a few cases inventors could choose from a number of agents. So the (f)(2) agreement has to leave room that a university might require an inventor to disclose and assign to a research foundation rather than to the university directly. In those cases, in practice, if not formally, the university has delegated certain of its responsibilities to the research foundation. By operation of the standard patent rights clause definitions, such delegation makes the research foundation also a “contractor,” just as the (f)(2) agreement delegation makes “contractors” of university inventors.

Since (if the (f)(2) agreement is properly implemented) everyone involved ends up being a “contractor,” why not just put “contractor” into the (f)(2) agreement? This question runs to the heart of the Bayh-Dole law. The reason contractor cannot be specified is that the contractor (none of them, including the inventors), may desire to pursue a patent on the invention. In that case, it is the federal agency that then has the choice of whether to pursue a patent. If so, then assignment is directed to the government, and the papers necessary to file a patent application are requested by the government–not by the contractor. So the (f)(2) agreement leaves out the indirect objects, because these can be either a contractor or the government. And that’s the essence of Bayh-Dole, that there is this choice.

The (f)(2) agreement goes even further and leaves out who the (f)(2) agreement is *with*. Is it an agreement between the contractor and its employee with the government as a third-party beneficiary? Or is it an agreement between the employee and the federal government, with the contractor as a third-party beneficiary? The (f)(2) agreement becomes part of the funding agreement, and joins the employee personally, if she invents, as a party to the funding agreement. As an additional party, the employee-inventor has obligations to the other parties to the funding agreement–the university-contractor and the federal government. Because the university-contractor could assign or substitute or subcontract obligations to other parties as well, the (f)(2) agreement text does not specify who the agreement is with–it is an agreement that joins the one who signs to an existing federal contract, without regard for what other parties may be or become involved.

Very nice drafting we thinks it is. Just as agencies may take different approaches to whether they require ownership absent a university claim, so also universities may take different approaches with regard to their employee-inventors. Some universities use a related “research foundation” to manage inventions. Some don’t. Some use multiple agents–though that practice has all but disappeared. The law is amazingly, even brilliantly liberal on this point. Shame to waste all that brilliance if no one appreciates it.

Bayh-Dole permits universities to designate personnel to manage subject inventions without requiring those personnel to be employees of the university–they could be attorneys in a law firm or employees of an invention management firm. The drafting of the (f)(2) agreement permits a university to designate a *whole list* of acceptable agents to manage its disclosure and patenting affairs–different personnel at different organizations for medical inventions, nanotech, stem cells, software, and alt energy, for instance. Under Bayh-Dole, an inventor at University A could even disclose and invention to University B’s research foundation, or to a patent law firm, or to a company set up to manage inventions in a given area, so long as University A designates the personnel at the firm to handle the patent matters. It’s at the discretion of University A.

Further, under 37 CFR 401.14(a)(k), university A can assign “rights” (broader than and different from the simple equation title=ownership) to any organization “which has as one of its primary functions the management of inventions, provided that such assignee will be subject to the same provisions as the contractor.” That is, University A can assign its Bayh-Dole right to elect title to any such organization, OR if it has exercised its right to elect title, it can assign the rights it has obtained (from the government, not from the inventors!) to such organization, OR if it has obtained assignment of patent rights from inventors, it can assign those patent rights to such organization, provided it also assigns all the provisions of the federal funding agreement that apply to the organization. *Any* of these are anticipated in 37 CFR 401.14(a) (k)(1). It is up to whomever accepts these rights to do those things necessary to protect the government’s interest, following the flow down in (k)(1): “provided that such assignee [of rights obtained through the operation of Bayh-Dole, the funding agreement, and the standard patent rights clause of whatever sort] will be subject to the same provisions as the contractor.”

This is one of three major flow downs in Bayh-Dole. The others are the subcontracting flow down in 37 CFR 401.14(a)(g), which I won’t discuss here, and what is more like a by-pass in 37 CFR 401.9, which is important to this whole discussion and shows what the university intervention permitted by Bayh-Dole really is. Read:

§ 401.9 Retention of rights by contractor employee inventor. Agencies which allow an employee/inventor of the contractor to retain rights to a subject invention made under a funding agreement with a small business firm or nonprofit organization contractor, as authorized by 35 U.S.C. 202(d), will impose upon the inventor at least those conditions that would apply to a small business firm contractor under paragraphs (d)(1) and (3); (f)(4); (h); (i); and (j) of the clause at § 401.14(a).

Here is a summary of the sections referenced in 37 CFR 401.9:

(d)(1) convey title to agency on failure to disclose, elect title, maintain patent rights
(d)(3) govt rights in foreign patents if contractor doesn’t pursue or maintain
(f)(4) notice of govt rights in patent specification
(h) reporting on utilization
(i) preference for US industry
(j) government march-in rights.

If a university does not require its employees to assign their inventions to the university (or to a research foundation, etc) *and* the funding agency permits contractor inventors to retain rights, then the deal in Bayh-Dole is directly between those inventors and the agency, and at minimum consists of the above. Note: provisions in the standard patent rights clause directed at universities and research foundations are missing from 37 CFR 401.9, including section 401.14(a) (k) with its clauses about assignment and royalty sharing and use of remaining funds for research or education. Also missing are provisions (b) and (c), but the duties indicated there are taken up and implied by the (d) clauses. The inventors have it easy compared to the university-directed apparatus.

The agency depends on the contractor-university fulfilling its commitment to have an (f)(2) written agreement with its research employees to protect government interests. Beyond that, once the university and its agents are not involved, it is up to the agency to stipulate its interest directly with the inventors. The inventors are obligated to that stipulation through their (f)(2) written agreements. Think of an (f)(2) agreement as making a public, irrevocable, enforceable commitment to do whatever someone authorized under the standard patent rights clause requests–

• it could be the university,
• it could be personnel designated by the university,
• it could be an invention management agent to which the university has assigned its Bayh-Dole rights (not patent ownership), or
• it could be an invention management agent to which the university has assigned ownership of the invention and also its Bayh-Dole rights and obligations (once the university has acquired ownership of the invention by assignment from the inventors).

The university serves as the public registrar for the (f)(2) agreement. Bayh-Dole is a law that authorizes the use of uniform invention management agreements by federal agencies. To understand the law, you have to follow the agreements. You cannot read only the statute, as legal commentators–who should know better–generally do. To do so is to fail to comprehend the statute. If there were a client involved, it would be negligent, but in the forum of academic papers and blog articles (now, about the same thing, except the papers are stuffier), folks can skip steps if it suits them. I aim not to skip steps. Thus, this is longer, and runs against the “Three Amigos” version of Bayh-Dole, to get at what is actually required by the law.

The great challenge in designing Bayh-Dole is to constrain federal agencies to a uniform approach while permitting all the various inventors, universities, and invention management agents to be flexible in how they manage things at their end. Do you see the challenge for the drafting? Bayh-Dole is nothing like the old Institutional Patent Agreements. Despite what Joe Allen claims, it could not possibly be so. While an IPA was a federal contract that did require universities to obtain ownership of federally supported inventions, Bayh-Dole is a statute placed with federal patent law: it is a statute about federal agreements. Bayh-Dole not only is not of the same nature as an IPA, it also does away with IPA-style requirements. An IPA was negotiated between a federal agency and a single university or research foundation. An IPA then could be specific to the requirements of that organization.

But Bayh-Dole demands a uniform agreement–and thus cannot know what the requirements of any particular contractor might be. If Bayh-Dole had been drafted to demand that all universities change their patent policies to take ownership of all federally supported inventions, the law would never have been approved by Congress. It would have upended the patent management environment that faculty and universities and agents had developed. It would have undone the very approach to invention management that it was seeking to exploit by making federally supported inventions readily available. If Bayh-Dole had been drafted to require a uniform IPA, it would have been laughed out of Congress because it would have merely been a ploy by a few research foundations to force all universities to adopt their approach to invention management, as a condition of federal law. A bad dream–and one that amigos continue to dream–but not what Bayh-Dole does.

On the federal agency side, agreements are to be uniform. On the research side–the inventors’ side–the agreements must be flexible, to accommodate whatever happens to be the policy and practice at an organization receiving federal support on behalf of its research personnel.

Anything else beyond this (f)(2) agreement made by the inventors  is deal play between the university and its employees over something *other than compliance with Bayh-Dole*. That could be money, power, notoriety, faculty rights, officiousness, fear, sense of public purpose, or whatever, but it ain’t Bayh-Dole. Bayh-Dole does *not require a university demand assignment to itself of patent rights from its federally funded inventors*. That’s because, at a minimum, Bayh-Dole is drafted to recognize that some universities want that assignment to go to a research foundation, not to them directly, and others may not want to direct assignment at all. Moreover, and more importantly, it’s because assignment specifically to the university hosting the research has little to do with the broad objective of Bayh-Dole, which is to use patent rights to promote the use of federally supported inventions. How this makes money for universities or patent agents or inventors or companies is not the concern of Bayh-Dole. How, even, this makes money for scientific research or education is of no concern. Same for how patenting results in benefits available to the public (not necessarily inventions or products), better university-industry relationships (where are the metrics for *that*?), support for small business (not just your equity-granting venture backed start ups sucking up all the local investment capital), and support for US labor (in the forms of manufacturing jobs, especially). Bayh-Dole is about the minimum that must happen if anyone other than a federal agency ends up owning a patentable invention made with federal support. It does not specify how patents are used for the public good, but it does state expressly as objectives that this is to happen. It is odd to see university officials tie themselves up in a nit worrying about complying with timely disclosure requirements in the standard patent rights clause and ignoring entirely compliance with Bayh-Dole’s statutory statement of objectives.

Absent a university claim on invention rights in federal research, agencies deal directly with university inventors. The university is a service intermediary. It serves as a contract aggregator for federal agencies working otherwise directly with faculty investigators. A university provides logistics support–resources, budget management, reporting. These services allow faculty investigators to focus on their projects rather than on paperwork. The investigators propose research; the investigators control the performance of that research, the investigators may invent; and the investigators report and publish. Absent the university’s interest in patents, it would also be the investigators reporting their inventions and discoveries directly to the funding agencies, as they do with their final reports.

Bayh-Dole is about managing the role of universities as stewards of these interactions relative to differential federal agency requirements pertaining to inventions and patents. The federal funding agreement used with universities is found at 2 CFR 215 [now 2 CFR 200]. Bayh-Dole’s implementing regulations and standard patent rights clause are incorporated by reference at 2 CFR 215.36(b)  [2 CFR 200.315(c)]. Most commentators on Bayh-Dole fail to follow the agreements and not only fail to read the standard patent rights clause but also they fail to consider the entire federal funding agreement set forth in 2 CFR 215. If they did, they would see that the standard patent rights clause is not the only instrument that places conditions on the university disposition of patents. Bayh-Dole concerns the tug of war over patent ownership between a federal agency and the inventors (and, especially between an agency and anyone that obtains ownership of inventions from the inventors). Bayh-Dole’s authority in the patent rights clause extends to patent matters, not how patents should be used to meet Bayh-Dole’s objectives. But 2 CFR 215.37 takes up university management of intangible property, including patents and patent applications acquired or improved with the use of federal funds.

According to 2 CFR 215.37, universities as recipients of federal awards serve as trustees on behalf of the intended beneficiaries of those awards with regard to any intangible property acquired or improved with the use of federal funds. When the university-contractor-recipient requires assignment of its employee’s patent rights as a condition of employment, or of using federal funds, or of being involved in a federally funded project, it is using the federal funds to acquire intangible property. (See 37 CFR 401.14(a)(g) for an express prohibition against such acquisition in the case of subcontracting: “The subcontractor will retain all rights provided for the contractor in this clause, and the contractor will not, as part of the consideration for awarding the subcontract, obtain rights in the subcontractor’s subject inventions.”)

The university acquiring patent rights in this manner becomes a trustee, not merely a corporate owner of patent rights free to anything at all. Its rights and interests in that intangible property–here, patent rights–are conditioned on looking out for the beneficiaries, not for itself. It is both a breach of its federal obligations–and deeply unethical–for a trustee to use assets entrusted to it to look out primarily for its own concerns. In the context of Bayh-Dole, the extra apparatus in the federal funding agreement with universities shows this difference in treatment between true owners of patent rights (the inventors) and those that choose to obtain that ownership to support federal objectives with regard to patent rights.

The lesson in each of these places–in Bayh-Dole, in the standard patent rights clause, in the federal funding agreement with universities–is that if a university is going to obtain an inventor’s patent rights ahead of the public as represented by the federal government, then that university had better be sure that it is taking up a public interest and not merely its own self-interest or worse, allowing itself to be used–for the money or out of weakness–as an agent of corporations, speculators, and monopolists.

Bayh-Dole is not God’s gift to the country, or to universities, or even to university patenting professionals. It surely is not federal innovation policy that universities have free reign to make money from federally supported patents any way they wish. Bayh-Dole is not even a gift to universities from the Three Amigos. It is more like dynamite tied up with a ribbon, and it is the ribbon, not the spark at the end of the fuse, that’s the gift from the Three Amigos. Folks like Joe Allen don’t want you to work through the law, the regulations, the agreements. They don’t want you to read this. They want you to think that the work that it takes to see through their false claims is not worth the effort. They want you to think that bad people who want to understand the law and accept their proper obligations under each federal funding agreement.

We are deep into it now, and if you have read this far and have not flamed out with a note about long posts, I am deeply appreciative. Some things do not work with quips and pithy two-sentence repartee. Too much is written blaming Bayh-Dole for university behaviors that ignore its provisions. Too much is written ascribing to Bayh-Dole requirements that are simply not there. Consider:

1) A university does not have to elect to retain title. Bayh-Dole is not a vesting statute. It does not force invention ownership to universities. Universities may choose never to “elect to retain title” to subject inventions. That means: they never intervene by policy or contract in inventor decisions. University obligations under Bayh-Dole then would be merely to secure written agreements to protect the government’s interest, educate potential inventors on the timely disclosure of subject inventions, and report disclosed inventions to the government. That’s durn simple stuff. That’s all Bayh-Dole has to be, until a someone wants to be ahead of the public in obtaining a patent on publicly funded research.

2) A university may therefore choose to permit their employee-inventors to decide whether the university should obtain title to a subject invention (and then it’s up to the university to decide whether the conditions proposed by the inventors are acceptable). That is, put a check box on the disclosure form. If the inventors want the university involved, then they check the box. Otherwise, it’s between the agency and the inventors via 37 CFR 401.9–that is, *outside the standard patent rights clause* at 37 CFR 401.14(a). Everything that follows for university involvement has to do with what the inventors expect when they check that box. That’s what university IP policy under Bayh-Dole should address. If a university has a bureaukleptic policy that requires everyone to check the box without making any commitment as to how the inventors’ work will be managed by the university as a trustee working on behalf of the beneficiaries of the publicly funded project, that says something right there about the university impulse, but says next to nothing about compliance with Bayh-Dole, innovation, or public service.

3) A universities may designate a range of invention management organizations–not just themselves. There’s no requirement in Bayh-Dole for a university to demand assignment to itself. A university could designate one, two or many other organizations–foundations, universities, federal labs, or invention management companies. That is, the university policy may say, “If you invent, then you must choose an invention management agent acceptable to the university to manage your patent work, or you must otherwise take steps acceptable to the university to separate your commercial interest in patenting research funded by the federal government with your continuing work at the university. You may choose from this list of agents that are pre-approved, or propose another agent.”

In such a case, the university manages the potential conflict of interest between a researcher holding patent rights for commercial purposes and continuing to develop research at the university. So long as any commercial work moves outside the university, and stays within the appropriate time constraints for such involvement, great. A university could even have a financial arrangement with agents on its pre-approved list, just as universities did when they mostly worked with Research Corporation. The inventors chose to work with Research Corporation, assigned their inventions to Research Corporation, and if Research Corporation received royalties, it shared a portion of those royalties with the inventors under its deal with the inventors, and shared a portion of those royalties with the university under its deal with the university. The university share, in essence, is consideration for the referral relationship. The inventor share is consideration for the assignment of the invention.

This approach would open up the invention management landscape and give full play to what Bayh-Dole was designed for. It may also prove better for outcomes–if a university already has an established licensing practice with expertise and industry connections in the area of discovery, it might be the ideal management partner for what is invented by faculty at another, otherwise unrelated university. If universities were serious about national innovation, they would exploit this approach. If universities were serious about making money, for that matter, through deployment of patent rights, they would exploit this approach. Instead, it’s all me, me, me provincialism. And to what effect?

4) A university may permit its employee-inventors to designate an invention management organization *after* assigning ownership to the university or related foundation. As with 3), there’s nothing in Bayh-Dole that requires a university to ignore its inventors or separate them somehow from further disposition of patent rights. The university is a trustee. When the university acquires ownership of patent rights, the inventors become beneficiaries of its actions under Bayh-Dole (which requires sharing of royalties with inventors) and under the federal funding agreement (which requires the university to act as a trustee if federal funds were used to acquire or improve intangible property). One might argue that the inventors *have a moral right* (in the copyright sense) to be involved in the university disposition of the inventions they assign under Bayh-Dole until they freely waive that right.

5) A university could for that matter allow their inventors to continue to own their inventions and leave the disposition of inventions to 37 CFR 401.9. Given the requirements on that ownership (from the federal agency that funded the work), allow the inventors to continue to use university facilities to work on their inventions, treating them as 37 CFR 401.9 directs, as small business contractors. The conflict of interest matters for federal funding purposes are covered by the federal agency obligations accepted by the inventors. Further, the university could reach a financial arrangement with the inventors for contracted facilities use in those circumstances–again, just as is done in many university copyright policies.

6) There is nothing “corporate” about Bayh-Dole patent ownership. University ownership of inventions is not simply “the entire right, title, and interest” because the law (through an express statement objectives and conditions of utilization) and the funding agreement (through the trustee requirement) places limits on what the owner of a subject invention may do. A university cannot freely dispose of a subject invention. If the university does not file, or does not choose to maintain, a patent right, the rights must be offered to the agency or go to the public domain. That’s not quite “ownership,” is it? If the university or its licensees are not diligent, the government can “march in” and force compulsory licensing. Pretty big lien on ownership. If the university makes money, it has to share that income and use the remainder for specific things. If the university wants to assign, it must have agency approval except in certain cases. That’s not rightly “ownership” either. It’s all “patent title, with limitations.”

7) This is your Buzz Lightyear moment. You are not a flying toy. University “title” in subject inventions is a matter of “falling with style.” If you check the sole of your university patenting shoe, you will see a govt agency has signed its name. Live with it. More importantly, recognize what you are (a steward) and change your policy and practice behaviors accordingly. This applies equally to university inventors and administrators.

Federally supported inventions are to be deployed in support of public purposes. There are many ways to do it. We have explored a tiny bit of the space. Changes in world research and investment economies mean it’s not business as usual in the US. Changes need to be made in practice to diversify it, not to fix it or replace one autocratic system with another. Who will step up?