Sean O’Connor starts an excellent article that gives a detailed account of history behind the Bayh-Dole Act (“Mistaken Assumptions: the Roots of Stanford v. Roche in Post-War Government Patent Policy“) this way:
The Bayh-Dole Act of 1980 was built on a mistaken assumption that “contractors”—recipients of federal funding—were securing assignments of inventions from their employees.
Despite O’Connor’s arguments, I don’t find evidence for any “mistaken assumption” in the run-up to Bayh-Dole. Drafting mistakes–yes, but not ones arising from mistaken assumptions–but just plain failure to execute a plan. Whatever the executive branch patent policy was, in 1968, when Norman Latker at the NIH and Howard Bremer from the Wisconsin Alumni Research Foundation collaborated to revive the lapsed NIH Institutional Patent Agreement program, the template IPA agreement for use with nonprofits–the future targets of Bayh-Dole–included a requirement that the nonprofit have a patent agreement with employees (Section VI(a)):
The Grantee shall require assignment to it of all right, title and interest in and to each subject invention on which it elects to file any patent application for administration by it in accordance with and subject to the terms and conditions herein set forth.
Latker claimed later that the IPA provided the basis for Bayh-Dole. If that was the case, then Latker, at least, had no mistaken assumptions packed in his bags. Furthermore, when the Nixon revisions (1971) of the Kennedy executive branch patent policy (1963) were codified at 41 CFR 9-1, the required patent rights clause included a requirement that contractors have a patent agreement with employees (9-1.107-5(a)(e)(3)):
The Contractor shall obtain patent agreements to effectuate the provisions of this clause from all persons in his employ who perform any part of the work under this contract except nontechnical personnel such as clerical employees and manual laborers.
Patent agreements are agreements under which employees promise to assign their inventions to their employer. In the case of the standard patent rights clause at 9-1.107-5(a), the contractor has an obligation to assign inventions to the federal government, so to “effectuate the provisions,” the contractor has to obtain title from its employee inventors. It’s pretty clear, isn’t it? There are no mistaken assumptions in play regarding invention ownership or patent agreements required by federal contractors. The standard patent rights clause in place five years before Bayh-Dole makes it clear that a contractor must have patent agreements with employees. A contractor, then, having complied with this requirement of the standard patent rights clause, would have at least equitable title in any subject invention–employees would have a legal obligation to assign inventions to “effectuate the provisions” of the patent rights clause.
The problem is, Bayh-Dole does not authorize such a clause. O’Connor can’t figure out why, and goes looking back to the Biddle Report of 1947 and works his way forward on the premise that Latker and Bremer–enmeshed in university patenting practices for a decade– were clueless about university patent agreement practices.
Whatever the problems with the Biddle Report–mistaken assumptions or whatever–they have nothing to do with Latker’s Bayh-Dole’s crock up. Rather than simply revising the existing federal procurement regulations, which would have been rather easy to do, Latker instead created a sewn-together statute to preempt all other statutes (but for Stevenson-Wydler). Latker wanted to preempt the various special statutes that required federal agencies to take ownership of inventions made in their contracting programs. Thus, he needed not only to preempt federal procurement regulations made under executive branch patent policies but also to deal with federal statutes. And he crocked things up doing so. There’s no patent agreement clause specified by Bayh-Dole that parallels the federal procurement regulation’s standard patent rights clause requirement.
One explanation is that such a clause couldn’t work as part of federal patent law–it would, in effect, make inventions made in federally funded work owned by contractors receiving money to supervise that work rather than the inventors that made the inventions. Bayh-Dole would have indeed been a vesting statute. But it would also appear that in doing so, Bayh-Dole would have fallen outside of the authority provided by the Constitution, which gives the federal government the power to grant to inventors for limited times exclusive rights to their inventions. There’s nothing there that gives the federal government to outright grant those exclusive rights to anyone else, let alone to organizations that it chooses to manage research funds. That would amount to picking favorite organizations. That might be done under a procurement heading–requiring contractors to have patent agreements in place–but not under federal patent law.
Instead, Bayh-Dole was passed into law incomplete. Latker then had to scurry over to the Office of Federal Procurement Policy to have a hand in drafting the implementing regulations.
During that process, the following was inserted into the standard patent rights clause at 37 CFR 401.14(a)(f)(2)–with no authority from Bayh-Dole to do so, and Bayh-Dole preempts all other statutes on the matter (35 USC 210). I quote it in the form recently amended by NIST, with the addition in bold:
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) of this clause, to assign to the contractor the entire right, title and interest in and to each subject invention made under contract, 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 written agreement requirement is supposed to take the place of the “patent agreement” requirement in the Federal Procurement Regulation that Bayh-Dole preempts. But it does not do what the Federal Procurement Regulation does. Although it sits at the same spot as 1-9.107-5(a)(e)(3), it does not do the same thing. First, in its original form, this written agreement requirement had no assignment requirement in favor of the contractor. Instead, contractors were to require inventors to execute all papers necessary to “establish the government’s rights in the subject inventions.” That is, the written agreement is a patent agreement with the federal government. The provision comes in a section headed “Contractor Action to Protect the Government’s Interest.” The written agreement, required by the contractor, binds potential inventors to each funding agreement’s patent rights clause.
Bayh-Dole does not require inventors to assign their inventions to federal contractors. I contend that it cannot do so. Congress lacks the Constitutional authority to do so. Similarly, the standard patent rights clause did not require inventors to assign their inventions to federal contractors. Instead, contractors were to require inventors to establish the government’s rights–ownership or a non-exclusive license. Inventors could not establish these rights if the inventors were forced by Bayh-Dole to assign their rights to federal contractors (or the rights vested outright with the contractors). No, the written agreement provision in Bayh-Dole is a requirement that contractors release any claim on inventions made in federally funded research. This is a complete reversal from both the IPA and the Federal Procurement Regulations. No wonder Latker, Bremer, and others made such ballyhoo about Bayh-Dole vesting ownership–they had failed to turn patent law into a vesting statute that favored institutional contractors.
The written agreement requirement does one more thing: it makes inventors parties to the funding agreement and by the definitions in Bayh-Dole, parties to funding agreements are contractors. When inventors invent, they own their inventions, and those inventions then become subject inventions, but those subject inventions are owned by the inventor-contractor, not by the institutional contractor. Universities don’t comply with the (f)(2) requirement, and NIST has recently muddied things further by inserting an assignment provision–but that provision can operate only on subject inventions.
If inventors are not parties to the funding agreement, then nothing they invent is ever a subject invention–Bayh-Dole applies only for subject inventions for its provisions regarding contracting–Bayh-Dole does not then apply and whatever vestiges of the federal procurement regulations remain would come into effect. If inventors are made parties to the funding agreement, then the very action by an institutional contractor of making them parties to the funding agreement also forces the institutional contractor to release any interest it may have in any inventions as a matter of federal contracting.
Any claim an institutional contractor may have has to arise because the institutional contractor has an equitable interest in the invention–hired the inventor to invent or assigned the inventor to the research or entered into a patent agreement with the inventor as a condition of employment entirely separate from any federal funding. In other words, university faculty simply don’t meet the conditions because universities as standard, basic policy assure faculty of freedom of research (the university cannot assign them to conduct any particular research) and freedom of publication (the university cannot require faculty to publish, nor to publish in the patent literature, nor control what faculty choose to publish, even using the patent system). For small businesses and university staff employees, there’s no big deal. Patent agreements take care of matters.
In 2011, the US Supreme Court in Stanford v Roche made clear that “of the contractor” means inventions owned by the contractor and that Bayh-Dole does not give contractors any special right in inventions made in federally funded research. The rights of the inventors are conveyed as they always have been, by conventional patent agreements, not by a stipulation of federal patent law. The great effort, then, of university administrators and their legal allies after Stanford v Roche is to supersede academic freedom policies without admitting that’s what they are doing. It sucks, really. The institutional ownership approach has not worked. Even if one does away with academic freedoms, there is nothing gained in the effectiveness of the resulting practices. Those practices sucked under the IPA program, which ran for a decade and was shut down as ineffective and contrary to public policy. Bayh-Dole, based somewhat on the IPA program, has done worse, yet is hailed as a success–but only by university administrators and their front organizations, biomedical lobbying groups, and various IP attorneys.
The Supreme Court was not asked to write a dissertation on Bayh-Dole, and they didn’t. But if we look at the implementing regulations, we can see that the Court got it right. Not only does the (f)(2) written agreement function as a requirement that contractors release their interest in inventions made with federal support to the extent that any claim is based on their status as federal contractors, but also the implementing regulations establish that inventors are to be treated as small business firms, and for that purpose have their own version of the standard patent rights clause–37 CFR 401.9. And there, interestingly enough, we find that inventors are not required to file patent applications, though nonprofit and other small business firms are required to do so for each invention made in federally funded work that they acquire. The implementing regulations neither require inventors to assign their patent rights nor compel inventors to use the patent system–both requirements are alien to federal patent law.
Look again then at the core provision of Bayh-Dole, 35 USC 202(a)–and for “small business firm” remember that the implementing regulations require inventors to be treated as small business firms:
Each nonprofit organization or small business firm may, within a reasonable time after disclosure as required by paragraph (c)(1) of this section, elect to retain title to any subject invention
The election to retain title is not qualified with regard to who might make a claim against inventors for ownership of their inventors. Remember, too, that this is federal patent law, no longer a procurement regulation. Inventors, as small business firms, are assured that they can elect to retain title–outright, against the claims not only of the federal government, but of any claims that might be made against their ownership position. An inventor must disclose (and Bayh-Dole requires disclosure to move through the institutional contractor) and must elect to retain title. That’s it. The inventor then has the benefit of the inventor patent rights clause at 37 CFR 401.9, and under that clause has no obligation to file any patent application. That’s as it should be. That makes sense of the law as it is written.
We can of course adopt the position that Bayh-Dole should mean what university administrators with rather large legal budgets claim it means–screw a close reading of the statute and the implementing procedures. That is, of course, how university administrators manage their own policies. In some cases, spending millions of dollars, they beat down a faculty member at a Yale or a Missouri, and in other cases, they lose badly, as repeatedly at California and at Stanford. When courts read the law rather than listen to the worm-tongues, then we get the rule of law, and that’s a good thing, as far as patent law goes as a tender of innovation (which these days isn’t all that far, actually).
If we choose to read the law carefully for reasonable interpretations that take into account the full statement of the law and its implementing regulations, we get a very different picture of Bayh-Dole than the one repeated incessantly in infographics, in university press releases and technology licensing office web sites, and in many academic articles and treatments by the popular press. We should opt for rule of law and not allow institutional administrators boondoggle inventors, the public, and the courts with extravagant, unsupportable “readings” of the law and extravagant, unsupportable accounts about how successful the law they don’t follow has been. If, following the rule of law, we find that things don’t work so well, then the neat thing about a democracy is that we can then adjust the law based on evidence.
Unfortunately, politics and basing policy on evidence and reasoning (and taking into account how much we don’t know and how much we disagree and how much we cannot even imagine is at work that we are not able to imagine) aren’t quite the same thing, and in politics, it seems so much better to lie or bluff about whatever sounds good or offers power, and since no one expects the truth anyway, lying or bluffing isn’t really unethical, or so the argument goes. If politicians were dairy farmers, we would get bullshit, not milk and cheese.