The miracle of Bayh-Dole came about, so the story is told, because Senator Long, the arch-critic of Bayh-Dole (“the worst bill I’ve seen in my life”), suddenly flipped his position to give Senator Bayh a consolation gift for losing his bid for re-election, on the condition that Senator Dole promised not to amend the bill later to extend it to large companies. “Sure, sure, absolutely not, Senator Long.” Senator Long must have got something mighty tasty in return to flip his vote on a lame-duck session bill that gutted a previous bill and added language that had already failed in the Senate and now required a unanimous vote.
But what? What was worth blowing away the community of science commons that had paced the interface between science and technology development from the 1950s to the 1970s? Given that President Carter flipped his position, too (he had backed a competing bill that covered all federal contractors, sponsored by Senator Schmitt). So had Senator Nelson from Wisconsin (home of the Wisconsin Alumni Research Foundation, which was closely aligned with Norman Latker, the former patent examiner turned NIH patent counsel who drafted Bayh-Dole). Very strange, very tasty something.
So Bayh-Dole passed in 1980, got signed by President Carter, and came into effect in the summer of 1981. Now Senator Dole had promised no amendments to the law to let big company contractors have the same right to retain ownership of inventions made in federally supported work that Bayh-Dole gave to small companies and nonprofits. What then to do to complete the agenda and not go back on the promise? Ah, play dirty politics.
Norman Latker drafted for President Reagan a “Presidential Memorandum,” picking up the language of the Kennedy executive branch patent policy published in 1963. In this new Memorandum, President Reagan turned executive branch patent policy on its head. Reagan required federal agencies to use Bayh-Dole practices for large companies. Bayh-Dole wasn’t extended to large companies. Executive branch patent policy was flipped instead to require Bayh-Dole for large companies (except where executive branch policy could not preempt federal statute, and except when federal agencies decided to give large companies a better deal than they would get following Bayh-Dole practices).
The 1983 Memorandum is a short bit of work. Here is the first paragraph:
To the extent permitted by law, agency policy with respect to the disposition of any invention made in the performance of a federally-funded research and development contract, grant or cooperative agreement award shall be the same or substantially the same as applied to small business firms and nonprofit organizations under Chapter 38 of Title 35 of the United States Code.
“To the extent permitted by law” recognizes that Bayh-Dole claims precedent over other federal statutes, and presidential memorandum cannot. “Shall be the same or substantially the same” indicates that federal agencies have some room to vary from Bayh-Dole. “Research and development contract” is just sloppy drafting, since the conjunction should have been “or,” not “and” if Latker aimed to follow Bayh-Dole. Chapter 38 is now Chapter 18, if anyone cares. And the use of “any invention” rather than “any subject invention” is perhaps just a happy lapse.
The gist of Reagan’s memorandum is that Bayh-Dole is now extended to large company contractors, but with flexibility, but also as a matter of executive branch policy, not the policy of Congress (35 USC 200), even as it stipulates following the policy of Congress.
There’s more, though, to the screw over of Senator Long:
In awards not subject to Chapter 38 of Title 35 of the United States Code, any of the rights of the Government or obligations of the performer described in 35 U.S.C. 202 – 204 may be waived or omitted
Whoa. We are talking here about federal agencies giving large companies the same deal as executive branch policy that Bayh-Dole gives small companies and nonprofits. But now there’s authorization for federal agencies to vary from that deal–they can waive anything in that deal–sections 202 to 204 are the standard patent rights clause, march-in for nonuse or unreasonable use, and US manufacturing requirements–if the conditions are right. Reagan’s memorandum then becomes “federal agencies must use Bayh-Dole for large companies except they can do whatever they want under the proper conditions.” [Under Reagan’s Memorandum, large companies can get a better deal than small companies, and with fewer procedural barriers to do it–all the federal agencies have to do is find some way to provide funding that’s not called a contract or doesn’t specify research or development as its primary purpose. And that’s what federal agencies do. See, for instance, Other Transaction Authority. Or see the treatment of Bayh-Dole here. ]
One might think, if cursed with that capability in these policy discussions, that if a president aimed to extend Bayh-Dole practice to large companies in 1983, s/he might draft it to require federal agencies to use existing patent policy (Nixon’s, a variation on Kennedy’s, codified with a whole infrastructure of implementation, with a fuss in 1973 and a reissue in 1975), and if that didn’t work, try Bayh-Dole-like practices, or even default it the other way. Instead we get a wholesale authorization to do whatever. Odd–and certainly not the makings for a “uniform” policy, that grail of federal bureaucrats. Instead, federal agencies are authorized to waive any government (public) rights and any contractor obligations. That’s a big asymmetry. Agencies cannot waive government obligations, and they cannot omit contractor rights.
Reagan’s Memorandum then is an outright attack on the public protection apparatus of Bayh-Dole. Large company federal contractors can expect, then, fewer obligations than Bayh-Dole requires of small companies and nonprofits, and can expect that the government will have fewer rights.
Now for the conditions under which large companies should get a more favorable deal than Bayh-Dole offers to small companies and nonprofits. Federal agencies can waive their rights and company obligations:
if the agency determines (1) that the interests of the United States and the general public will be better served thereby as, for example, where this is necessary to obtain a uniquely or highly qualified performer;
Rephrased–if the agency thinks it is better, such as if the agency wants a particular company to do the work and that company won’t do the work under conditions in which the public would be protected from being exploited by patent monopolies–such as price gouging, planned scarcity, and suppression of adaptations and applications. Reasoning–better to have this company “participating” even if the public gets screwed over than not.
And if that’s not enough, a second condition:
or (2) that the award involves co-sponsored, cost sharing, or joint venture research and development, and the performer, cosponsor or joint venturer is making substantial contribution of funds, facilities or equipment to the work performed under the award.
If the big company provides “a substantial contribution” of its own. No guidance on what makes a contribution “substantial.” Is $100,000 substantial–one employee for a year? Is $100,000 still substantial if the federal agency provides $1,000,000? Again, a federal agency has the discretion to waive rights or omit obligations whenever a large company make a contribution. Again, this is a way better deal than Bayh-Dole offers. Bayh-Dole goes entirely the other way: if the government contributes anything under a funding agreement to a project in which an invention is made, then the government has rights in that invention as soon as a contractor acquires it.
Notice that Bayh-Dole says nothing on spending money “on” an invention. The government spends money on a project in which an invention “arises.” Others may also spend on that project. Their expenditures, as it were, become contributions to a public project, accepting the Bayh-Dole terms whenever an invention arises and is claimed by a party to the funding agreement. You will see nonsense around trying to argue that only money spent “on” inventing should count toward whether the federal government has any rights. But that’s goofy fantasy thinking, up there with unicorns and flying turtles. How would anyone track how the conceiving of an invention receives funding? Conception, despite popular usage, takes place in the brain. It’s difficult to book thoughts–especially realization thoughts–to a particular time sheet or place. An invention may run against one’s work assignment, may not have anything to do with one’s work assignment, and generally won’t have been anticipated by one’s work assignment. If one charged out that thinking time–maybe a matter of seconds–to a federal grant, it might look at first like a kind of fraud, charging for something that wasn’t otherwise specified.
This distinction between “paying for an invention” and “an invention made in federally funded work” is important. In the case of Stanford v Roche, the Supreme Court took up the university argument that Bayh-Dole forced ownership of an invention to the university if “only one dollar of federal funding was applied toward the invention’s conception or reduction to practice.” This, the Court noted, would be “truly surprising” and could not have been what Congress intended or they would have “said so clearly.” And that’s the difference between thinking that money can be “applied” to conceiving an invention or even reducing that invention to practice as a matter of drafting a patent application. The test is not how federal money is spent, but rather whether the invention that arises in the context of a federal funding agreement is within the scope of that agreement–is responsive to the purposes that the federal agency seeks to achieve or support.
Bayh-Dole’s implementing regulations start out with an extended statement of scope (37 CFR 401.1) that NIST is now trying to get rid of, because that statement of scope argues against the idea that a contractor can keep separate accounting for grant work and invention work, and by doing so avoid Bayh-Dole obligations and government rights. NIST wants to make it easy for small businesses and nonprofits to avoid Bayh-Dole–and that, apparently, will “unleash” innovation. This, after all the blather that Bayh-Dole already has unleashed innovation, which it of course hasn’t, and neither will NIST’s changes, for all that–as a general rule bureaucracy suppresses innovation, no matter how it wordsmiths its virtue signaling.
Back in the 1940s, the University of Southern California tried keeping double accounting in a Navy contract to help to develop better helmets for pilots. The USC part of the effort was that personnel in its aviation medicine program would look at the physiological effects of “acceleration, decompression, anoxia, and cold, and methods to combat these effects.” On the side, they came up with a helmet design, got patents, and licensed exclusively to a company, which, in 1966, sued the federal government compensation for federal use of the invention made in the USC-hosted work. At trial, the company argued that all the work on the invention had been done with non-federal money, nothing billed to the Navy. The Navy argued in effect that once USC became a contractor and agreed to help the Navy with its helmet problem, then the Navy had a royalty-free right to use whatever USC came up with responsive to the Navy’s project, whether or not USC billed the Navy for thinking up an invention, or for working to develop it, or licensing it out to a company to develop. The court sided with the Navy. Once you agree to help a federal agency, anything within the scope of that help is a deliverable, no matter that you don’t bill the federal agency for it. For a good read, see Mine Safety Appliances v. United States.
It would take someone especially incompetent to allow a research contractor to keep separate books so all the really good stuff could be swept outside of the contract deliverables and so screw over the party that commissioned and provided funding for the work. But that’s what NIST wants to have happen, in the interest of “innovation” of course.
The Reagan Memorandum, then, allows Senator Dole, the patron saint of Norman Latker, to screw over Senator Long. Yes, Bayh-Dole has not been modified (yet) to be extended to large companies. So nice. Instead, executive branch patent policy has been totally flipped so that large companies can get a “better” deal–fewer obligations, fewer government rights– than Bayh-Dole permits for small companies and nonprofits. Sweet.
In 1987, President Reagan followed up his 1983 Memorandum with Executive Order 12591, which tasks the heads of federal departments and agencies to, among other things,
(4) promote the commercialization, in accord with my Memorandum to the Heads of Executive Departments and Agencies of February 18, 1983, 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 EO is deeply messed up. Inventors hold title to their inventions until they assign their rights away. A court might construct such an assignment for them if they were expressly hired to invent. Otherwise, they have title until they choose otherwise. A federal agency then cannot grant title of any invention to a contractor until the federal agency has acquired that title from an inventor. The 1983 Memorandum does not grant title to contractors–it specifies use of Bayh-Dole provisions unless an agency waives or omits those provisions. Bayh-Dole does not take title from inventors or vest title with contractors, and so the Supreme Court ruled in Stanford v Roche. The apparent effort of the EO, if this is not just shoddy drafting, was something more nefarious, to make it appear as federal law that Bayh-Dole does vest title to inventions with federal contractors and that in all other cases, somehow title to inventions made in federal work vests with federal agencies (or vests no where until a federal agency decides where it should vest), who are then directed to grant that title to any contractor, not just to any large company contractor–no action by the inventor necessary. Such a deal is not only a big screw-over of Senator Long but it turns federal patent law topsy turvy, which is the reading that the Supreme Court objected to in Stanford v Roche.
Senator Long got his requirement–Bayh-Dole has not been extended to large companies. Instead, he got doubly screwed over. Executive branch patent policy has been changed to give large companies better deals than Bayh-Dole gives small companies and nonprofits. And the attempt was made–successfully for 25 years–to screw over inventors, who were told that they had no rights to their inventions, per Bayh-Dole, when they did. Seems like a felony offense to defraud inventors under the color of law. But that’s what Norman Latker apparently aimed to do, and university administrators have got away with doing. When Bayh-Dole didn’t do enough, he moved on to presidential memoranda, executive orders, and implementing regulations to make it appear that Bayh-Dole did more than it does, and that large companies get a better deal than do small companies and nonprofits.
The present fuss over large pharma companies profiting extraordinarily well from inventions made in federally contracted work is a direct consequence of how Senator Long got screwed over.
What’s not so much recognized is that when Latker displaced the whole of executive branch patent policy with Reagan’s 1983 Memorandum, he also undid the authority for federal agencies to have a default general claim on inventions arising from funding agreements with contractors. That default claim was in those prior patent policies, as codified by the Nixon patent policy in 1971. Bayh-Dole changed that default claim for small companies and nonprofits when an invention became a subject invention–that is, when a contractor acquired the invention, and it was patentable, and it was made under a federal funding agreement. But if a contractor did not acquire the invention, then it was not a subject invention and Bayh-Dole did not apply. Before Latker’s piece of work with the 1983 Memorandum, federal agencies would have had a claim on those inventions that inventors did not assign to contractors under the parts of executive branch patent policy that did not get preempted by Bayh-Dole. But after 1983, the only authorized means by which a federal agency could acquire title to any invention made in federally funded work were (1) a contractor acquired title and failed to disclose, or disclosed and did not elect to retain title, or failed to timely file a patent application; or (2) the agency determined there were exceptional circumstances and wrote a requirement into the funding agreement before it was accepted that all inventions must be assigned to the federal government. Exceptional circumstances have hardly been used.
The outcome then of Latker’s exploit to screw over Senator Long has been that as a formal action, the federal government no longer has a standing expectation that inventions are funding deliverables unless some other authorized (by statute, by regulation) arrangement controls. For inventors, this should mean freedom–but their rights have been suppressed all sorts of ways. For federal agencies, the Latker exploit means they can do pretty much they want, except in dealing with small companies and nonprofits, which must continue to get the raw deal (from their perspective) that Bayh-Dole as a default requires.
Sure, it’s politics, and politics can be played dirty, but we don’t always have to live this way do we?