AUTM Summarizes Bayh-Dole

Here is an AUTM summary of Bayh-Dole:

Major provisions of the Act include:

• Non-profits, including universities, and small businesses may elect to retain title to innovations developed under federally-funded research programs
• Universities are encouraged to collaborate with commercial concerns to promote the utilization of inventions arising from federal funding
• Universities are expected to file patents on inventions they elect to own
• Universities are expected to give licensing preference to small businesses
• The government retains a non-exclusive license to practice the patent throughout the world
• The government retains march-in rights.

I have bolded a few passages that could be improved.  Then we can look at what’s been left out.

Bayh-Dole applies only to an invention “which is or may be patentable” (or otherwise protectable under US patent law) *and* is 1) owned by a contractor to the government and 2) was “conceived or first actually reduced to practice” 3) in the “performance of work under a funding agreement.”   Each of these elements matter.  None is there for fluff.  The phrase “of the contractor” in the definition of “subject invention” in Bayh-Dole means ownership, not agency.  This was made clear by the Supreme Court, following the carefully reasoned AIPLA amicus brief (and perhaps, in some way, raised by the discussion here).

“Innovations” are something almost but not entirely different.  An invention is something new (and useful and non-obvious, for patentability) that’s been made or discovered.  An innovation is something that is perceived as new to an adopting (or resisting) party, especially something that will produce a change in a political, social, or technological order.  Ah, I know, it’s just a couple of words that get switched around.  But there’s a huge difference when it comes to ownership claims.  Ownership of an invention in the context of Bayh-Dole focuses on the patent rights available, while ownership of an “innovation” extends claims to all sorts of things that may or may not involve intellectual property.   A number of universities are doing just this–claiming to own know-how, expertise, ideas, data, software, anything “tangible”.  The claim to ownership does not even have to have a theory of title associated with it–the universities are even claiming things that they can’t own, that no one can own.  Using “innovation” rather than “invention” is just part of this effort to grasp at everything, just in case there’s money in it.  But really, claiming to own what can’t be owned is rather of the nature of a threat rather than a title.

It may seem silly to pick at the word “under” in “developed under” but the wording used in the Act is “in the performance of work under a funding agreement”.   It could be that these things are merely equivalents for “whatever it is that the government might claim”.  But there is more going on with the AUTM usage than that.   Stanford argued (brief account here)  that if even one dollar of federal money was used, Bayh-Dole was triggered.  The Supreme Court rejected that argument.  The “even one dollar” argument is not original with Stanford.  It is endemic in the university technology transfer community.  (An instance cited by Peter Lee is here, at note 496). Faculty are told this all the time–if even $1 of federal money has touched your work, then Bayh-Dole applies.  University administrators love this concept.  Why is that?

Here’s how it works.  There are multiple steps, but when it comes together, you will see how remarkable the argument is.

If Bayh-Dole is a vesting statute, meaning that somehow under the operation of federal law the ordinary rules for invention ownership for inventions made with federal funds is suspended so that universities that host such research can claim ownership, then

If a university administration wants to own all such inventions, and especially if the university does not have adequate patent agreements with its personnel, or if the university has a policy that states the university only takes ownership when required by law or contract,

Then the best thing to do is to construe federal funding compliance as broadly as possible, so that “even one dollar” of federal funding invokes Bayh-Dole, vests ownership of any invention with the university, and creates “deal flow” for the technology transfer office.

It is little wonder, then, that some university patent administrators think that a challenge to Bayh-Dole is a challenge to their livelihood, even when that challenge is to a misinterpretation of the law.

Bayh-Dole is set up anticipating that contractors will push for limited federal interest in inventions made by a contractor’s employees.  If a contractor claims an invention is not within Bayh-Dole, an agency can challenge that claim.  But universities have gone the other way–expanding the interpretation of Bayh-Dole to gain the benefit, they thought, of the vesting action of the law.  If Bayh-Dole applies, then the institution owns the inventions. Simple as that.  So why not make Bayh-Dole apply everywhere possible, on the thinnest of arguments?  In essence, that is what administrators have done, seeking to expand the apparent claims of Bayh-Dole well beyond its stated scope, for institutional self-interest, as an exploit of university policies that otherwise would not have provided the administration with a claim anchored, apparently, by the force of federal law.

But as I have pointed out, there is a substantial difference between “under” and “in the performance of work under” a funding agreement.  The first instance means “even one dollar” while the second instance means “within the planned and committed activities” of a federally supported project (following the language at 37 CFR 401.1 (a)(1)).  That is, for inventive work to be “under” Bayh-Dole, it has to be an objective or deliverable of the statement of work setting out the project.  An individual can be paid with federal money to work on a project, and in doing so can make discoveries that are not part of the planned and committed work, and those discoveries are not within the definition of “subject invention” in the standard patent rights clause established by Bayh-Dole.  They are not made in performance of the work specified, even if they arise in the context of such work, and even if those making the discoveries are paid with government funds.  As 37 CFR 401.1 makes rather clear, if a federal project seeks a basic understanding, and as a consequence of that basic understanding efforts are made to apply that understanding, and such efforts to apply are not part of the statement of work–not what the government is paying to support–and those efforts do not “diminish or distract” from the work that is specified–then the invention is not within the definition of “subject invention”.

Whew–that’s a lot, isn’t it?  But it is really quite basic stuff.  The government contracts for certain deliverables.  That is the scope of its interest.  The government funds research within a particular, written scope, and claims an interest in the deliverables of that research as specified in the statement of work–often the proposal submitted by university faculty, as established by the budgets and division of responsibilities set up to manage the project–the planned and committed activities of the grant.  If what is discovered isn’t a deliverable, then it’s not of any interest to the government.  While such stuff may be of great interest to patent administrators, the means to take ownership does not run through Bayh-Dole or the standard patent rights clause of the federal funding agreement that controls the work.

From this, one can see how the logic works, however.  The idea of the “one dollar rule” as reflected in Stanford’s arguments aims to use Bayh-Dole to take ownership of everything possible.  Thus, it is an easy substitution to replace “in the performance of work under” with “under”.  In doing so, one throws out the key concept–deliverable–and substitutes a different one–anything.   Both are concepts that can be expressed in summary form.  For instance, AUTM could have written:

“Universities may elect to retain title to inventions assigned to them and made as deliverables under federal funding agreements”

If an invention is not a deliverable, then it is not subject to Bayh-Dole, and universities have no need for the deal under which they retain title relative to a federal agency request for title or a license.  If the invention is not assigned to the university, then the university has no benefit or obligation with regard to the invention under Bayh-Dole.

While “deliverables” is clearly an interpretation of “in the performance of the planned and committed activities, or other activities that diminish or distract from such work under a federal funding agreement” the “major provision” of the Act is preserved, while the substitution of the seemingly simple “under” allows the concept of the “One Dollar Rule” to persist in technology transfer usage.

Moving along, AUTM’s next “major provision” conflates a number of things.  First, the law states that it is Congress’s objective to use the patent system to promote the utilization of inventions, and to promote collaboration between universities and industry.  These objectives are restated by AUTM as if the law mandates universities to do these things, when the objective is for the *law* to do them by creating conditions under which such things will happen.  It’s not so much that AUTM has written universities in as if the law provided them with a mandate rather than an opportunity, but that AUTM then conflates this mandate with that of collaboration, as if the law’s objective of collaboration was met when industry “collaborated” with university technology licensing offices by paying for the licenses to patents on offer.  That isn’t in the law.  There is nothing in the law that says that universities must charge for patent licenses or that industry must pay.  The bit in the law about nonprofits sharing royalties doesn’t say there has to be royalties.  So here AUTM has written the universities into Bayh-Dole playing a role that Bayh-Dole was intended to play.  If one wanted to play at this game, the summary could have easily been written to read:

“Commercial concerns are encouraged to collaborate with university inventors to promote the utilization of federally supported inventions” or

“Commercial concerns are encouraged to collaborate with universities to promote free competition and enterprise where federally supported inventions are made publicly available.”

Either of these statements is also a conflation of Bayh-Dole’s stated objectives.  Both are a mash-up for a reason, and both bake in an agenda and recast it as if it were an objective of the law rather than of an interested party seeking advantages under the law.  The appropriate thing to do is to recite the objectives of the law as they stand, and offer a statement of how the interested party aims to meet those objectives in its own practice.  It is not appropriate, however, to turn a special interest into a mandate of federal law when it is only the private agenda of a group of university patent administrators eager to build up their operations.

Now, as for “elect to retain title” we know from Stanford v Roche that it means, in Bayh-Dole, to “decide to continue to own title acquired through private arrangements, relative to any federal agency request to assign that title to the government.”  It does not mean “choose to take title” as AUTM and many university administrators asserted in their amicus briefs to the Supreme Court.

However, AUTM has not given up their interpretation.  We find this in such usage as “Universities are expected to file patents on inventions they elect to own.”  Well now, technical points first.  Universities *agree* under the standard patent rights clause in their funding agreements with the government that they will file patent applications.  There is no “expectation” about it.  Also, it is indeed the case that they will file applications, not patents, as no one “files a patent” unless it’s in a filing cabinet after it has been issued.  But the big point here is the usage “they elect to own.”  In a pleasant way, this may mean merely that universities may choose to accept ownership of inventions made by faculty and students with federal support.  But this is a summary of Bayh-Dole, not a general account of university choices.  The implication of the “elect to own” is that somehow Bayh-Dole gives universities special powers to simply take ownership of inventions made with federal support.  This of course was the core issue of the appeal of Stanford v Roche to the Supreme Court, and the court ruled that there is no “elect to own” in the law.  There is only an “elect to retain ownership” got by some other means than the operation of Bayh-Dole.  AUTM, it appears, simply cannot give up a position that is simply wrong.  Here, on their web site, they continue to put out mis-information to the world.

Moving along, notice that AUTM describes the government as “retaining” a non-exclusive license when in fact Bayh-Dole sets up a multi-threaded set of conditions under which the government is granted a license, or title, to subject inventions, by whomever happens to own title, depending on the circumstances.  This use of “retaining” is actually quite at odds with what is going on in Bayh-Dole.  It suggests that somehow the government owns the patent rights, and cedes that ownership to the contractor, when the contractor “elects title”, leaving the government to “retain” only a non-exclusive right.

But that is not at all how Bayh-Dole operates.  US patent law vests ownership of inventions with inventors, and inventors may assign those rights to others.  Bayh-Dole is about creating a uniform policy for federal agencies when they contract with universities (and other nonprofits, and small companies) with regard to the disposition of rights as between the inventors (and anyone the inventors might assign rights to) and the expectations of the funding agencies.  The core of this relationship between the inventors, their agents, and the funding agencies is authorized by Bayh-Dole in the form of a standard patent rights clause, which is to be included in every funding agreement.  It is this funding agreement, and the particular patent rights clause it contains, that actually controls the disposition of any inventions made in a given federally funded project.

In addition to some technical adjustments to AUTM’s summary of Bayh-Dole, then, one might add a few additional points, omitted entirely by AUTM:

• The disposition of inventions is subject to a patent rights clause included in each federal funding agreement.
• Universities are required to require certain of their employees to make written agreements to protect the government’s interest, including a commitment to license or assign their inventions to the government as may be requested.
• University inventors own their inventions unless and until they assign that ownership.

It is understandable that AUTM would have its agenda.  One would hope that this agenda would be to promote professional practices pertaining to invention management in the context of university research and discovery.  Instead, it would appear that the agenda is rather to preserve a misreading of Bayh-Dole and maintain a kind of hegemony over discussions of how research inventions might influence innovation.  In particular, Bayh-Dole does not represent so much an opening up of opportunities for private efforts to promote the use of federally supported inventions so much as it stands as if it were a special deal by which the federal government gives ownership of such inventions to universities for the purpose of institutional profit-seeking.

AUTM really ought to correct its information, and even publish a retraction and correction to its claims with regard to Bayh-Dole.  Even in simple summaries, AUTM’s misconstruction of Bayh-Dole persists, misleading administrators, faculty, and the public alike.