There is a continuing effort to misconstrue the Supreme Court decision in Stanford v Roche. Here is an excerpt from a newspaper article on the reasons why the University of Utah changed their policy to make even more claims to ownership of faculty scholarship:
The changes, approved by the school’s Academic Senate earlier this month, came about after a 2011 U.S. Supreme Court decision that “dramatically changed the landscape of patent ownership,” according to a U. memo. The case centered on a professor who created a test to measure the amount of HIV in a patient’s blood while conducting research at a drug company, according to a New York Times report. His idea was used to create a testing kit now widely used in hospitals and clinics. Stanford sued, saying the professor’s ideas belonged to the school by contract, but the Supreme Court sided with the drug company because its contract was more precise.
How wrong can folks get this? And why? First off, the case at the Supreme Court was about whether the Bayh-Dole law voided otherwise valid assignments of inventions. The Supreme Court said it didn’t. Second, Stanford had a patent policy that was perfectly clear: inventors owned their inventions “if possible.” That is, Stanford’s policy only made a claim to ownership when a contract or law required Stanford to own. It was a great policy, one that supported a creative faculty with a string of big hit licenses. A present assignment might have changed how Stanford obtained title to inventions, but it would not have changed Stanford’s scope of claim.
Utah’s general counsel argued roughly the same thing put out in the newspaper article:
Invention Assignment Language. The Policy’s invention assignment provision incorporates language required by the U.S. Supreme Court’s recent decision in Stanford v. Roche. (Section III.B.1)
The Supreme Court does not require any such language. The whole point of the Supreme Court decision was that the Bayh-Dole Act does not vest ownership of inventions made with federal support in universities that host the research.
Perhaps the Utah folks meant the CAFC decision. Well, that didn’t turn on one policy being more precise than the other, either. The CAFC looked at the assignments that had been made. There was a promise to assign in the Stanford policy, but the invention in dispute was out of scope of that promise, Stanford had approved the consulting (for nine months) with the commitment to assign inventions to the company as a condition of the consulting. Stanford’s own request for assignment came three years after a patent application had been filed. Here is the chronology, according to the CAFC. I have highlighted a few things and added the Supreme Court decision at the end. H is the co-inventor whose invention was the subject of the Stanford litigation.
1988 H employed at Stanford as post-doc
Signs Stanford patent agreement
Scope: inventors own “if possible”
Form: promise to assign
1989 Feb H signs patent agreement with Cetus
Scope: inventions arising “as a consequence of” the work at Cetus
Form: promise to assign and present assignment
1990 H returns to Stanford
1991 Dec Roche buys Cetus’s PCR business and makes HIV detection kits
1992 May Stanford files a patent application
1992 June Stanford reports invention to NIH
1994 Nov Stanford grants Government a non-exclusive license
1995 Apr Stanford notifies Government it elects to retain title (beyond 2 year period)
1995 May H signs Stanford assignment document
1999 Oct first patent issues
2000 Apr Stanford asserts ownership to Roche
2003 Jan second patent issues
2005 Oct Stanford sues Roche
2006 Oct third patent issues
2007 District Court declares patents invalid (among other things)
2009 CAFC rules that Roche has an ownership interest so Stanford lacks standing to sue
2011 Supreme Court rejects Stanford claim that Bayh-Dole vests title with Stanford
The CAFC reviewed the assignment trail. The only valid assignment in place at the time Stanford filed a patent application was that of H to Cetus. That assignment happened to take the form of a present assignment. If it did not take that form and was merely a promise to assign, then it is reasonable to expect that Cetus (then Roche) would have on notice of the invention obtained an assignment per the terms of the visitor’s agreement.
The present assignment to Cetus did not “trump” an earlier promise to assign made by H to Stanford. The invention was outside the scope of that earlier promise, and even if it had not been, by Stanford approving (and encouraging) the consulting, whatever claim it might have would be subject to an argument that Cetus (then Roche) relied on Stanford’s approval to undertake the relationship. When Stanford asked for assignment from H, it was not relying on its policy to demand assignment–all Stanford had was the opportunity to make a request.
The CAFC found that when Stanford attempted to obtain assignment from H and his co-inventors years later, H had no rights to assign, as he had assigned his rights already to Cetus. Stanford then decides that because federal funds were used in the research related to the invention, that Bayh-Dole vests in Stanford ownership of the invention, or mandates Stanford ownership of the invention, regardless of any other assignments inventors may have made. It is only by having Bayh-Dole require assignment to the university that Stanford’s own patent policy operates to bring the invention within its scope and makes the promise to assign operate.
One might then recognize an exploit used when a university patent policy retains the old language under which the terms of any approved research contract trump the defaults of the patent policy. One can be as fluffy as one wishes in the patent policy–just make sure that research agreements all claim ownership of everything for the university, and force faculty to accept this claim or they are not released from their official duties to participate in the research. Essentially, one creates a second patent policy better than the first. There, one might find indeed that the second policy is “more precise” than the first, since it is embodied in research contracts rather than in formal patent policy. What is the policy foundation that permits a university to introduce into a contract for research terms on inventions that are not consistent with its own patent policy? Essentially, such a practice means that the patent policy means nothing for a great deal of university-based activity if administrators can impose contrary terms as part of research contracting. “The patent policy applies to you unless we decide differently in a research contract–then it is whatever we decide applies to you.” It is understandable that one might agree to terms requested by a sponsor and approved by the faculty involved. But not that one can interpose one’s own terms and force the sponsor and faculty to accept it. That disrespects the authority for the policy. Try to explain this, however, to the university patent counsel who believes clever is better than moral.
There is nothing in the court decisions that indicates that the assignment to Cetus was “more precise” than anything Stanford tried to do. The raw facts of the case are that Stanford accepted a deal with a company as a basis to gain access to the company’s technology, obtained federal funding in the same area as the company’s own work, filed a patent application on an invention that was determined to be within scope of the company’s expected claim, and then regretted its deal with the company and sued 14 years after the company had product on the market. And from these facts, Utah’s general counsel comes away with the lesson that one has to be “more precise” and for that, one has to expand the claim of ownership to more stuff? Is it incompetence, malpractice, or mendacity? Perhaps it is that lemming desire to “race to the bottom.” Dunno.
Even if Stanford had put a present assignment in its patent agreement, the outcome would have been the same, but by means of a different argument. Then, when the invention is made, there would have been two competing assignments–one to Stanford and one to Cetus. Still, Stanford loses, because its own claims are out of scope at the time that the invention is made, since Stanford is not required to take title per its own patent policy.
Even if Stanford had revised its policy so that it was not attractive to creative folks and a gem among the nation’s patent policies, and had adopted a crappy policy requiring assignment of everything some terrier could dig up–like, say Utah has done–Stanford would still lose. Even if the scope of the present assignment made a claim on the invention, Stanford’s approval of the consulting arrangement with Cetus, and the assignment obligation there, would have given ownership to Cetus. Stanford agreed to the conditions of the consulting. Even if Stanford could have asserted a claim, the claim would have failed because Stanford had waived the scope of that claim with regard to inventions made as a consequence of the work at Cetus. Stanford would not be able to simply change its institutional mind later and decide that it could enforce a policy claim that Stanford had six years previously made clear to all was out of its future scope of interest.
The practice lesson that there needs to be more “precision” in drafting patent policy claims is just plain bunk. It’s bullshit, really, in that technical sense of not having any regard whatsoever for the truth.
Here are some more equally stupid practice lessons from Stanford v Roche:
- Universities must have present assignments. Meaningless in 35 USC 261 issues.
- Universities must own everything. Antagonistic to academic freedom and innovation.
- Universities must impose what they claimed Bayh-Dole imposed. Perpetuate a wrong.
- Universities can create clever schemes so they can later sue research collaborators.
- Inventions in use mean “infringement” not “success.” Forget Bayh-Dole’s objectives.
IF Stanford had had a crappy policy antagonistic to creative faculty and refused to release employees from claims under that policy in extended consulting with companies to learn their cutting-edge technology which Stanford wanted for its own money-making interests, THEN, uh, no company worth its assets would let Stanford employees anywhere near its cutting-edge technology. That’s a decent practice lesson.
GIVEN that Stanford had a really insightful policy when a lot of universities didn’t, and had allowed an employee under that policy to learn cutting-edge stuff that other universities would have chafed at because of their crappy patent policies, THEN Stanford’s practice choices included (1) run with co-ownership of the inventions and develop other licensing opportunities in competition with Roche; or (2) work out a deal with Cetus on H’s return to Stanford for any subsequent inventing H might do at Stanford. Stanford could also have excluded H from work in areas that might lead to inventions Stanford wanted to be able to assert against Roche, but then that would have frustrated the very purpose for sending H to Cetus to learn PCR. This is also a decent practice lesson.
For most everyone, the best practice lesson is that a liberal patent policy, as Stanford had, allowed its faculty to get into all sorts of interesting, valuable relationships that kept Stanford at the forefront of research. When you are liberal, and at the forefront, opportunities find you. You don’t have to carry a gun, you don’t have to be the nasty, you don’t have to run a patent protection racket. Look at you: you are a university, not a half-witted, patent-crazed, speculator-serving zombie organization flush with government research dollars which you mistake for your own.
I wish I had a neurolyzer. <<FLASH>>
You are a university. Act like a university. Adopt a liberal patent policy that promotes freedom to innovate. And while you are at it, get a general counsel who respects you.