Evaluating university claims of invention ownership

I am working through this idea that an invention is owned by a university merely as the result of work within the scope of employment or through the use of university facilities (and resources and funding and whatever–class 3 unknown).

The logic seems to go thus:

A master is entitled to the fruits of the servant.

An employer is a master, and an employee is a servant.

The university is an employer, and faculty (and others) are employees.

Thus, the university is entitled to faculty (and others’) work.

However, in some cases, it chooses not to take that work and releases it.

As is the case for “traditional academic works”.

The implication of this line of argument is that the university is the “master” and faculty are “servants”.  That is, their work is done under the control of the university.  While we might grant that for certain aspects of instruction and service activities, it might give pause when considering research, whether “departmental” or “extramural.”  Here, faculty propose the work, and while there is an approval step for submitting proposals to sponsors, that approval is about the paperwork, generally, and not about the specifics of where to spend one’s effort.  The faculty investigators choose what to study, how to study it, what data to collect, how to analyze it, what to report, where and when to report it, who to cite, who to acknowledge.  It’s hard to find the actions of a master-employer in this activity.

The fallacy in the argument from master to employer is that some employers are not masters.  For copyright, this is essential in ascertaining work for hire.  It is not sufficient that one be employed; it must also be the case that the employer acts as the author, as the master, and what matters then is whether the work is within that *scope* of employment.  The university is no doubt an employer of faculty, but it does not assert control over that employment having to do with research unless it expressly commissions work–say, to figure out how to provide automatic updates of which campus parking lots are full.

And I get it, I really do–a lot of administrators, led by patent officers, want universities to own inventions, lots of them, as expansively as one can get it.  They want this to be matter of fact, efficient, no bother, outright.  As clean as any corporate claim, as absolute, as uncontestable.  If someone wants their rights back, they come begging, but no lawyer worth the money is going to miss the chance to set up the most powerful policy claim possible, to stick it to the little people who might find a chink or a loophole or oversight.  If there’s going to be a negotiation, it will be from a position of university power.  I get this.  I have lived with it.  I’ve been among these believers.  They think ownership by university administrations is a really good thing.  It’s fuel for tech transfer offices.  It’s in the public interest, and for public universities, it’s in the state’s interest.  They believe Bayh-Dole gave them rights in federally funded inventions (only to be frustrated by the Supreme Court) and they have made it only natural to extend the same approach to all other sponsored research, and then to any use of facilities, and then to anything in any aspect of employment, and then to anything in one’s area of expertise.  All of that.

And I object to the mindset, oppose the approach, and think the underlying arguments for it are suspect.  And I know that no matter how much I show those underlying arguments are fallacious, contrary to law, erroneous, foolish, and wrong, that’s really not the point, because the urge is still there, and the urge-generator keeps coming up with more arguments.  They aren’t arguments, really–more like excuses, rationalizations, diversions.  The point of them is to make people stop questioning the urge, the program, the determination to make faculty into labor, and the tech transfer office into the citadel of economic development and discretionary revenue generation for the university.  For all of this, I’m perfectly fine with the idea of “technology transfer”, of the use of the patent system to encourage the development and use of research-originated inventions, and involvement of university offices in the process.   There are reasons to support creativity, and universities should have more than a handful.  But it’s a long way from support for creativity to a fundamental starting point in claiming ownership of inventions.

Thus, when we turn to IP policy statements such as “in the course of their University responsibilities or with more than incidental use of University resources” (Stanford), or “in the course and scope of employment” or “with the significant use of Board or university resources” (Arizona), or “in the course of or pursuant to a sponsored research agreement” or “with significant use of funds or facilities” (MIT), we have to ask, to what extent does this class of claim carry any force if the employer does not exercise control?  This goes two ways:  it could be that the policy, while trying to be expansive to include everything (class 1 unknowns), actually aspires to something much more modest, claiming only those things where the university does exercise master control–such as direct assignment and supervision of duties and expressly commissioned work in exchange for some additional benefit.  Or it could be that folks in the university are determined to extend university master control over all aspects of faculty work, turning them truly into corporate employees serving the desires of their employer.  The problem for these IP policies is that they leave this ambiguity unaddressed.

In practice, however, it is absolutely clear that the administrations intend to use the policy constructions to claim everything.   Some institutions go further, and stake a claim to anything made “with the direct or indirect support of funds administered by the University” “by any person, regardless of employment status” (Michigan); “in the field of expertise and/or within the scope of responsibilities covered by my employment/appointment/association with the University” (Penn State).  In these formulations, the university is not even staking the claim on employment, but rather on the area of expertise, or an association, or an “indirect” support of funds.

Some even extend the definition of “employee”–such as “non-employees who participate or intend to participate in research projects” (Stanford)–and others expand the definition other ways, such as using “University personnel” to include “volunteers” and “anyone working under University auspices”, from which “creators” are those University personnel who make and invention, and any invention “made in the field or discipline in which the creator is engaged by the University” “is the property of the University” (Florida).  The Florida claim is amazingly expansive.  If you are a professional electrical engineer, and you invent something while also volunteering your time at the university, this policy says that the invention is the property of the university–even if the invention does not have anything to do with whatever it is you are doing as a volunteer.

If you are in any way associated with the university, whether by employment, appointment, or association, and whether or not you are directly tasked, or work in a sponsored project, or just hang with folks as a visitor, and whether you use resources or just get a secondary benefit from them, the university claims your inventions (subject, of course, to what it doesn’t want, which it will graciously “release”).    Imagine a Western movie, in which a town has a law that the moment you ride into town, everything in your saddle bags is the mayor’s.  That sort of thing.

The idea that an employee owes the employer all inventions is something simmering in the back of all of this.  It’s not this way in patent law, employment law, or academic tradition.  In patent law, an invention is owned by its inventor.  If the employer wants ownership, the employer needs to bargain for rights, or have a contract providing for assignment of ownership, or rely on an implied duty to assign, based on the circumstances surrounding the invention.  Employment or use of facilities is not sufficient to establish a claim to invention ownership.  The idea of the shop right comes about in response to what happens when an employee uses facilities of the employer to invent, but has not entered into an invention agreement with the employer.

A survey of American universities in 1948 found that most did not have patent policies.  The academic tradition was consistent with the broader conventions–faculty owned their inventions.  Employment and use of facilities had nothing to do with it.  Just as the public mission is served by faculty owning their “scholarly works” so also it was served by faculty owning their “scholarly inventions”.  It was this private ownership that laid the foundations for Research Corporation, the Wisconsin Alumni Research Foundation, research foundations as academic invention brokers, the technology transfer office, and Bayh-Dole.

University IP policy statements that lay out a requirement to assign are working one of a variety of angles.  One is that it is a condition of employment to assign.  For that, we are back to what limits are placed on an employer.  A number of states have laws that restrict the scope of claims an employer can place on an employee as a condition of employment, including Minnesota, California, Washington, Utah, Delaware, Illinois, Kansas, and North Carolina.  Universities in these states routinely ignore these laws in their IP policies.

Another is that the invention agreement is a quid pro quo for sharing royalties.  The inventor assigns and the university shares royalties.  This is a vestige of the invention broker organization, such as a research foundation.

A third is that there is  simply an assertion of a duty to assign, and it is a conflict of interest to do anything but assign, and the policy just makes this clear.  This approach argues that it is a misuse of facilities or resources to own an invention privately, and that to avoid the charge of unethical behavior, the university makes a safe harbor by requiring assignment.

A fourth is that sponsored research agreements carry terms (often placed there by the university) that require university ownership, and thus anyone participating in that research is required to assign to ensure that the university does not breach its obligations under the contract.   This is an expansion of the employment argument, but now including volunteers and collaborators, regardless of employment.

In each of these, the policy requires assignment, or in more naive formulations, simply asserts that inventions are the property of, or belong to, the university.  In these more naive, or coarse, or brute-force approaches, reading Bayh-Dole as a vesting statute provides a tremendous advantage–not only is federal funding a large portion of the total extramural funding at most universities but also vesting is a ringing endorsement of the idea that it is national innovation policy that faculty don’t own inventions, universities do, to feed their technology transfer offices.   The problem for the policies then is to find some justification for the requirement.  Typical formulations include “public benefit” or “to be fair”.  Others are rather backward–we take ownership in order to provide an incentive to faculty to invent by sharing some portion (usually less half, often 1/3 after deduction of costs and an administrative fee) of licensing income with inventors, if there ever is any licensing income).   Still others argue that taking ownership is indicated to protect the public from the non-use of inventions that would take place if faculty were permitted to own.

None of these formulations, however, get at the standing of the university as an employer, the relationship of research (whether departmental or extramural) to the “business” of the university (other than that the “business” of the technology transfer office is to try to make money off of anything it can, and to prevent anyone else from making money without it).   If the university is not the master-employer but merely the enabler of activity by paying for academic services, and is certainly not the master-author dictating the shape of scholarly expression, then the “scope of employment” premise fails–it is either substantially restricted, despite the expansive claims, or it is simply untrue–that the university has imported language for a model of employment from industry that it in fact does not use in its own activities.

We are left with the claims, then, based on use of facilities.  These fail even worse than those based on employment.    Use of facilities may give rise to a shop right, a duty to permit the employer to practice the invention, but it does not change the ownership of the invention.  One cannot merely assert, because my university provided buildings and supplies, the university owns the invention, or is entitled (by what) to own the invention.  In fact, everything points the other way.  Use of facilities does not entitle anyone to own an invention.  Use of facilities has nothing to do with it.

So we push deeper into the argument.  The university sets up the claim that there is an agreement, that in exchange for use of facilities, everyone who uses the facilities agrees to assign inventions to the university.   I have not found a university that narrows it down, so that the facilities in which the invention was conceived or reduced to practice are identified on a disclosure form.  It’s much more nebulous than that:  if you have a lab in Building A, and you study rat fat, and you invent a lubricant for ventilation fans derived from rendered rat fat, then you must have used Building A to invent, even if all you did in Building A was learn a lot about rat fat, and made the application of some derived product of that fat in an entirely different building, such as in the shower at home (not a university facility) and reduced the invention to practice at a company (also not a university facility).  Somehow, because the area of research is connected to the information used to develop a particular inventive application, the university has a claim.

Consider, however, that universities (apart from certain issues pertaining to teaching for credit courses for other organizations and being the PI on grants that could have been sought through the university) have no non-compete clauses.  There are no trade secret requirements.  There are no requirements on what in particular to study.  There are no location restrictions, so that study has to be done in Building A.   Thus, it is essentially a choice of the investigator, not a duty of an employee, that determines where research work, and inventive work, might take place.

Imagine a public square.  It has a fountain in it, and a statue, and a big brick plaza, and an espresso cart operated by the city.  You go to the square every day, and you buy a latte, and you work on a remote controlled helicopter that can fly to the top the statue and wash pigeon goop from the head of some town hero.  You are using public resources, but there’s no way that the invention is claimable by the city for doing so.   University IP policies try to avoid this by excluding generally provided resources–offices, libraries, desktop computers.  But they negate these exclusions by instituting conflict of interest rules that say that you cannot run a business from your office, or use your computer for private purposes, so obviously you cannot use your computer to work on an invention that you would own personally, as that would be a private purpose, so even though it is not a “significant use of resources” and therefore the university will not claim the invention under its IP policy, it will charge you with misuse of facilities by inventing there and not assigning the invention to the university.

As one policy puts it:  “University personnel have the obligation to refrain from any act that would defeat the university’s rights in any works and inventions they make . . . ” (Florida).   As I’ve pointed out elsewhere, Washington’s approach is even blunter–if you consult without pay, it still stands to reason that your “expertise” may change in value as a result, and so your unpaid consulting (talking to someone at a conference, answering a phone call from an industry colleague, responding to someone in an on-line discussion forum) results in a personal economic benefit, and so is treated like paid consulting, and therefore must be approved, and as a condition of that approval, you must assign all inventions to the university.

The agreement, then, regarding inventions and university facilities, is really one of conflict of interest, and to set up that conflict, the university has to establish that it somehow is owed all inventions, and that it is unethical for individuals to own their inventions.  This is not the master-servant argument.  This is not a non-compete prudence.  This is not a vesting argument.  This is an argument based on the idea that although the university provides one with access to facilities and resources in the normal course of university activities–by virtue of employment, appointment, association, approval–the university also, in its IP policy makes it appear to be the case that one cannot use that access without also giving up ownership of inventions.

If the argument were that the university has put up money expressly for an invention to solve a problem, that alone would not be sufficient to create an obligation to assign ownership, though it might establish a shop right or an implied license so that the university can practice.  But this is not that argument.  It is an argument that apart from such express commissioning, in spaces that a university otherwise provides, any inventing is to be treated as if it were made in a master-servant relationship, or in an express commissioning of work.  The deal is:  we let you use your lab in Building A to do research if you assign all inventions to us, except that the deal is not made when you move into your lab space, it is made by assertion of a policy that (for many universities) can be changed unilaterally at any time.  The deal is actually not a material condition of using the space.  It is an assertion of an obligation based in institutional envy.

Consider a student working in a lab in Building A, as part of a course assignment, and she invents.   For a public university, this is just the public square situation again, but now there’s a gate and a limit for who can be in the square, and a sign that says you must conform to city rules, and somewhere off in city hall, there is a rule book that says, the square is public except that if you do anything but stroll around or buy an espresso–such as if you invent–then as a condition of access, the city owns the invention.  It’s not a reasonable proposition.  It’s no longer public.  And there’s no agreement made.  It’s just an assertion.  It is essentially a threat.  If you don’t assign, we will accuse you of being immoral, for using the public square for your own personal benefit.

Now vary:  it is a faculty member walking into the same lab in Building A, and faculty member and student co-invent.  Why should it be any different?  What does the use of the facility have to do with it?  The laws that lay out the outer bounds of an employer’s claims as a condition of employment include the use of facilities as those are provided to employees.  It’s not a separate thing, where one can employee someone but deny them access to their place of work unless they agree to something that’s otherwise not allowed to the employer.

The baseline in America is:  an inventor owns his or her inventions until it’s assigned.  Employment does not force such an assignment.  Use of facilities does not force such assignment.  Working on a sponsored research project does not force assignment.  Assertions in a policy statement that there have to be such assignments is not an assignment, and still has not found a basis for assignment.  Even statements in law that indicate that an employer could require a range of assignments do not make those assignments happen, and it’s not contrary to law if an employee does not assign, and it’s not contrary to law that the employer does not require assignment.  All the more so for universities, in which the employer does not control research or other scholarly work unless the university expressly contracts for it, where the university’s “business” or “foreseeable business” is not “anything anyone does that’s inventive that we could make money on.”

It takes more than statements.  It takes an agreement made with regard to inventions between the employer and the employee.   Not something after the fact, not something buried away in the planning office in Alpha Centauri.  If not an express agreement, then an implied contract built out of the circumstances and pattern of dealing–what one would expect where the employer directs the work, and the employee agrees to the employer’s involvement and expectations for the results.

The university IP policies have turned a common law expectation that an employer has a right to own where there is a written agreement and the invention pertains to the employer’s business or research into one in which the employer’s business is whatever the employees do, and the employer’s research is any research the employees may be doing, regardless of whether that research was assigned, ignored, or rejected.  It’s actually a more comprehensively grasping position than most companies would think to take, with less supporting it by way of agreements or rationale.  No trade secret.  No competitive position.  No line of business.  No post-employment non-compete.  No manufacturing or sales capability.  No prohibition on visitors or consulting or collaboration.  No control of research or publication.  And yet a standing, all-out, we demand to own everything that anyone involved in university activities cannot legally resist, and we will use employment, appointment, access to facilities, and association as a basis for setting up the assertion of ownership.

In some sense, it is a pure power play.  It is not about reasoning about inventions, and it certainly is not anything required by law.  It is about setting up a situation in which the inventor is at such a disadvantage that the inventor will capitulate and accept the outcome, regardless of what laws, equities, practices, or academic conventions might otherwise be at work.  It’s a kind of avarice, or envy, unchecked even by business conventions, set out in policy statements as if it is the most proper thing, that a university should own the inventions of anyone working at the university.  When it’s not proper at all.  The improper has been made to look normal.

The University of Southern California, for instance, makes a big deal in its policy about “academic tradition of faculty ownership” and then repeatedly explains why this can’t be–laws, public benefit, whatever.  No matter that the arguments are hopelessly flawed.  What matters most is left unsaid, that in the “academic tradition of faculty ownership” includes ownership of inventions.  Even in 1948, where Archie Palmer is arguing that universities should have patent policies, he’s assuming that faculty will own, and universities will encourage faculty to use the patent system, and do good things, whether that is protecting the research environment from simple variations that lock out everyone or whether it is providing an incentive for a company to invest where it otherwise would not.  It’s about faculty becoming inventors will access to a range of agents and collaborators, facilitated by the university, with a patent policy that authorizes that facilitation.  There were no offices of technology transfer at universities.  And the next 30 years were the time in which university inventions, managed by non-university agents, made such an impact that Congress was moved to pass Bayh-Dole and expand the range of inventions available to faculty inventors.

And all this is being quashed with a set of claims that ignore the past successes, misread laws, misconstrue the employment relationship, make owning inventions privately immoral, and claim everything possible without regard to what is equitable, presuming that faculty, staff, and students will shy from the fight and simply give over what they have made, whether patentable or not.  It’s a strong-armed tactic, cowardly in its implementation, deceptive in its representations of activity.

What should universities do about it?

  • 1) End the practice of inserting ownership claims in research agreements and then enforcing those claims as prior to statements of ownership in policy.  If a sponsor wants to insert an ownership clause, then it does so with the approval of the faculty investigator before it is accepted by the university.
  • 2) Drop the policy claim to invention ownership based on employment and use of facilities.  These are not the foundation for a claim but merely the outer scope of what in industry is regarded as a reasonable expectation.  For universities, that outer scope is much more limited–to expressly negotiated work to invent, directed by the university, for its benefit.
  • 3) Remove the pretense in policy that it’s immoral for individuals to own inventions, but that somehow it is fine for a university to own, and disrupt research use, interfere with industry collaborations, and work to distort academic work by offering a direct share of proceeds from money-making efforts, calling that a “culture change”.
  • 4) Deal with the money envy independently of ownership.  If an inventor associated with the university makes a lot of money, what is that to the university?  At the university medical center, there are people making more per year than others in areas like mathematics are going to make in a decade, just in salary, and more than 99% of all inventors will ever make, cumulatively, through their technology transfer office’s licensing program.   Use policy to provide ways by which inventors can dedicate income to university activities.
  • 5) Eliminate pre-set royalty schedules.   They don’t work for the division between university and inventor, and they don’t work for the allocation within the institution, and they don’t work for the technology licensing programs.  Make it a deal between the inventor and the tech transfer office.  If the inventor wants the university to own, then it’s also the case that the inventor also wants the university to benefit.  So why not see what you are playing for before deciding it’s worth it.  The aim of federal funding, at least, is for the university to work for the beneficiaries of the funding awards–not for the benefit of the university, as the pre-set royalty schedules assume.



This entry was posted in IP, Policy, Technology Transfer. Bookmark the permalink.