A few years ago you hired a brilliant researcher to work on new technology your institution is developing. In the stack of normal employment documents he executed, he signed a “Copyright and Patent Agreement” where he “agree[d] to assign” all inventions resulting from his employment.
His brilliance not withstanding, your researcher needed to develop skills in a particular analytical area that was being pioneered at another entity down the road. You decided to let your researcher work there for a few months, to get hands-on experience.
The other entity had your researcher execute a “Visitor's Confidentiality Agreement,” which stated that he “will assign and do[es] hereby assign” inventions made “as a consequence of [the] access” he was granted.
While conducting research at the other entity, your researcher developed new methodologies relevant to your technology. When he returned to your institution, he continued to refine the methodologies. You eventually filed patent applications, and obtained written assignments of the patent rights from the researcher. Because you had obtained NIH grants for some of the research, the patents were governed by the Bayh-Dole Act, which gives the U.S. government certain rights in inventions supported by government grants.
A few years later, you find out that another entity is commercializing your patented methods. You sue them for patent infringement, but they claim to have obtained rights to the methods from the entity where your researcher worked for a few months. How could this be?
The U.S. Court of Appeals for the Federal Circuit addressed this basic fact pattern in Board of Trustees of the Leland Stanford Junior University v. Roche Molecular Systems, Inc., which it decided in September 2009. Now the Supreme Court has affirmed the result in a decision that emphasizes the U.S. patent law principle that the rights to an invention belong to the inventor in the first instance, even if the invention is made by an employee in the course of his employment and is supported by government funding.
Promising to Assign Inventions in the Future vs. Assigning Future Inventions
Of course, inventors can assign their rights, and employers can make such assignments a condition of employment. The problem Stanford (the researcher's employer in the above fact pattern) faced is that the language of its Copyright and Patent Agreement was deemed to be ineffective to transfer rights in and of itself, and the later patent assignments were trumped by the Visitor's Confidentiality Agreement required by Cetus (the commercial entity).
Under Federal Circuit law, which governs this particular patent ownership issue, “the contract language ‘agree to assign’ reflects a mere promise to assign rights in the future, not an immediate transfer of expectant interests.” On the other hand, Federal Circuit jurisprudence holds that an agreement that purports to “hereby assign” inventions yet to be invented does effect “a present assignment of . . . future inventions.”
Thus, while the Stanford Copyright and Patent Agreement did not itself transfer the researcher's rights to Stanford, the Cetus Visitor's Confidentiality Agreement did transfer his rights to Cetus (which later transferred those rights to Roche, the patent litigation defendant), such that Cetus “immediately gained equitable title to [the researcher's] inventions.” Because of this, the subsequent patent assignments to Stanford were ineffective.
Once the invention was made, the legal rights automatically vested in Cetus, so the researcher no longer had any rights that he could assign to Stanford. As the court summarized the situation:
The researcher “signed away his individual rights as an inventor . . . while performing work for Stanford after promising to assign his rights to the university.”
The Bayh-Dole Act Does Not Divest an Inventor's Original Rights
Stanford argued that the Bayh-Dole Act required a different result, but both the Federal Circuit and the Supreme Court disagreed. As summarized by the Supreme Court, the Bayh-Dole Act (35 U.S.C. § 200 et seq.) was enacted in 1980 in order to “promote the utilization of inventions arising from federally supported research,” “promote collaboration between commercial concerns and nonprofit organizations,” and “ensure that the Government obtains sufficient rights in federally supported inventions.”
The Bayh-Dole Act permits a grant recipient to “retain title to any subject invention” as long as certain conditions are met. These include (i) disclosing the invention to the granting Federal agency “within a reasonable time”; making a “written election [to retain title] within two years after disclosure”; and filing “a patent application prior to any statutory bar date.”
Even when the grant recipient retains rights, the granting Federal agency has “a nonexclusive, nontransferrable, irrevocable, paid-up license to practice . . . [the] invention.” Additionally, the agency has “[m]arch-in rights” to grant a license to a third party under certain circumstances, such as if the grant recipient does not develop the invention. Moreover, if the grant recipient does not elect to retain title to the invention, the agency can entertain requests from the inventor to retain rights to the invention.
Stanford focused on language of the Bayh-Dole Act that defines governed inventions as “any invention of the contractor conceived or first actually reduced to practice in the performance of work under a funding agreement,” and language that refers to a grant recipient's right to “retain” rights to an invention. Stanford urged that this statutory language vested federally funded inventions in the grant recipient in the first instance, bypassing the inventor.
The Supreme Court disagreed, noting that where Congress has intended to depart from the default rule of inventor ownership, it has done so more directly. The Court cited three specific examples: (i) inventions made under “certain contracts dealing with nuclear material and atomic energy,” where the relevant statute (42 U.S.C. § 2182) expressly states that “title to such inventions ‘shall be vested in, and be the property of, the [Atomic Energy] Commission’”; (ii) “certain inventions made pursuant to contracts with the National Aeronautics and Space Administration,” which the relevant statute (51 U. S. C. §20135(b)(1)) provides “shall be the exclusive property of the United States”; and (iii) “certain inventions under contracts with the Department of Energy,” which the relevant statute (42 U.S.C. § 5908) provides “shall vest in the United States.”
Finding no such language in the Bayh-Dole Act, the Supreme Court determined that it did not upset “the fundamental precepts of patent law and deprive inventors of rights in their own inventions.” The Court stated:
The Bayh-Dole Act does not confer title to federally funded inventions on contractors or authorize contractors to unilaterally take title to those inventions; it simply assures contractors that they may keep title to whatever it is they already have.
What Is an Employer To Do?
The Supreme Court decision reminds us that patent ownership is the exception to the rule that “whatever an employee produces in the course of his employment belongs to his employer.”
Instead, “unless there is an agreement to the contrary, an employer does not have rights in an invention” until the inventor “expressly grant[s] his rights . . . to his employer.” Most employers are aware of this, and require their employees to sign such agreements, but the result in this case highlights the criticality of the exact language used in such agreements.
It is important to keep in mind that this issue was not reviewed by the Supreme Court. Indeed both Justice Sotomayor (in a concurring opinion) and Justice Breyer (in a dissenting opinion joined by Justice Ginsberg) expressed doubts regarding the Federal Circuit's temporal line-drawing between “agree to assign” and “hereby assign” contract language. Justice Breyer went so far as to say that the Federal Circuit's “reasoning seems to make too much of too little.” Still, unless and until the Supreme Court faces the issue head-on and reaches a different conclusion, the Federal Circuit's “linguistic” rule will prevail. This means that employers should consider having their employees execute carefully drafted agreements as soon as possible, and well before any potentially inventive activity is undertaken.
This case also highlights the potential pitfalls of collaborations, particularly where conflicting agreements may arise regarding ownership of inventions. While research institutions often want to promote the free exchange of ideas and collaborative advancement of research, they should be aware of the potential impact on their intellectual property rights when an employee walks out the door.