Does Bayh-Dole require a patent policy?

Here is a bit (see paragraph 9) from a major US university policy on treatment of inventions under Bayh-Dole:

Incumbent upon members of the University community who apply for and receive federal funding to support research or who use federal monies in the conduct of their research is the requirement for written agreement that they will promptly disclose patentable inventions to the University and will execute all instruments necessary to protect the rights of the government and/or the University.

Note: the (f)(2) requirement is actually: “The contractor agrees to require, by written agreement, its employees, other than clerical and nontechnical employees, to disclose promptly in writing to personnel identified as responsible for the administration of patent matters and in a format suggested by the contractor each subject invention made under contract in order that the contractor can comply with the disclosure provisions of paragraph (c), above, and to execute all papers necessary to file patent applications on subject inventions and to establish the government’s rights in the subject inventions.”

There is nothing in 37 CFR 401.14(a)(f) about requirements on people requesting funds (as the university policy text quoted above would have it) and nothing about protecting the rights of the university (it just isn’t there!).   The university is just making up that part of the law.  It sounds good.  It may even be a desire of university administrators.  It for that matter may even be a desire of some faculty members.  But it is not a requirement of Bayh-Dole.  Is it simply a drafting mistake?  Just sloppiness?   If so, it is at the heart of the law, the university’s involvement in research, and the rationale for using the patent system.

Here is how it works:  the (f)(2) agreement is an agreement made within a federal agreement.  It is an agreement by the employee to report inventions timely and fully [to the university], to sign paperwork to allow patent applications to be filed, and to sign paperwork to establish the government’s interest in inventions.  Is the (f)(2) agreement also a federal agreement?  Is it made between the research employees and the government?  Or is it made between the employees and the university?

[The next couple of paragraphs are a bit sloppy, and I’ve edited them to tighten the argument, along with the text below.  Deletions are strike through, additions in square brackets].

The issue is whether a university can rely on an (f)(2) agreement in lieu of having its own patent policy.  Clearly, (f)(2) requires disclosure to the university, so that part is covered.  If the university obtains title to the invention, then (f)(2) shows that the inventors have agreed to sign paperwork for patent application filings, so that part is covered, too.  The part that’s left is whether the inventor’s commitment to establish the rights of the government in a subject invention means that the inventor has also committed to establish the university’s rights, when the university has “elected to retain title” before it has title.

That is, if the university receives a disclosure of invention, forwards it to the government, and elects to retain title in the invention, but hasn’t actually obtained assignment of the invention nor has a policy that requires assignment of such inventions, then what?  If the inventors have committed under (f)(2) to establish the government’s rights in an invention, and the arrangement between the government and the university after the university’s election to retain title is that the government will defer to the university as a pre-approved invention management agent, subject to the various conditions of the SPRC, then does that election to retain title also allow the university to request assignment of the invention from the inventors relying on their (f)(2) commitment to establish the government’s interest in the invention–now that the government’s interest would, if the university obtains title, come about through the license granted to the government by the university under the SPRC.

Put it another way:  does the election to retain title mean anything more than that the federal agency allows the university quiet possession of title when the university obtains title?   Does it also mean that the university has the right to request title in reliance on the inventors’ (f)(2) commitments with regard to the government.  Clearly the disclosure obligation in an (f)(2) agreement is directed at the contractor/university–one could say the contractor is a beneficiary of the agreement, since the purpose is to permit the contractor to comply with its reporting obligations.  The commitment with regard to signing paperwork for patent applications is open-ended:  there is no indication of what invention management agent might be involved.  It could be the government, the contractor, or a pre-approved invention management organization.   Whoever obtains title would be able to rely on (f)(2) to obtain signatures.

So then what about establishing the government’s rights, the third part of (f)(2) agreements?  Clearly, the government can rely on this commitment to request from the inventors assignment or a license, whatever Bayh-Dole permits.  If, however, a contractor has “elected to retain title” under the SPRC, then the government has agreed it will not request assignment or a license from the inventors directly, unless the contractor fails in its performance under the SPRC.

Is the “elect to retain title” then equivalent to the government allowing the contractor (or an assignee of the contractor’s SPRC rights and obligations) to rely on the (f)(2) agreement to request assignment of title?  Clearly, if there is no assignment of title, then the contractor (or assignee) cannot comply with the SPRC requirements, such as to grant a license to the government, and things end up in 37 CFR 401.9, which pertains to situations in which the inventors have not assigned title to a pre-approved agent (the contractor or an organization having invention management as a primary function).  So if the contractor cannot rely on (f)(2) to request assignment–so that the contractor can establish the government’s rights in the invention–then there is no point in electing to retain title prior to having an agreement with the inventors regarding assignment (and that agreement does not have to be compulsory, and does not have to be done in advance of the research, or at employment).  Without assignment, the contractor cannot comply with the SPRC.

There are three scenarios.  Play them out in your thought-experiment bin.  Starting conditions (in Piscopo):  Federal funding!  SPRC!  (f)(2) agreement made!  Invention!  Election to retain title!   Contractor requests title!

Scenario 1:  Inventors refuse!  Election to retain title meaningless!  SPRC finished!  Agency requests rights under (f)(2) with reference to 37 CFR 401.9.

Scenario 2:  Inventors agree!  Voluntarily!  SPRC controls!

Scenario 3:  Inventors refuse!  Contractor invokes (f)(2)!  Inventors agree under (f)(2)!  Fulfills obligation to government!

A patent policy, and pre-existing patent agreement to assign inventions, is not needed by a university unless (a) administrators are adamant that they must own all inventions made by research personnel with federal funding; (b) scenario 3 fails.  If a contractor having elected to retain title can rely on the (f)(2) agreement to then obtain that title, then no private agreements are necessary–no patent policy required to implement Bayh-Dole’s SPRC.  Of course, it is obvious that no patent policy is required to implement, anyway.  The first two scenarios are fine, so long as administrators have not fixated on having to own everything.  That is, a patent policy and private compulsory agreements are only required when the university refuses to allow the inventors to decide the disposition of inventions.

The sole purpose of a policy and private compulsory agreements is to prevent inventors from deciding that disposition.  Bayh-Dole makes no such demand on contractors.  They do not have to have policy that prevents inventors from deciding the disposition of their inventions.  Indeed, the whole point of the (f)(2) arrangement is a recognition that university inventors may–and perhaps should–control the disposition of their inventions, in keeping with academic freedom and the separation of science from government controls.  If one doesn’t care about such things, then it is easy to see the (f)(2) agreement as useless–why give university inventors an inch of hope about anything involving their scholarship when it happens also to be inventive?  But that’s a different sort of question than the one we are working through here.

The federal policy, as established in the Bayh-Dole Act and implemented by means of 37 CFR 401 and in particular the Standard Patent Rights Clause at 37 CFR 401.14(a), is that the policy on disposition of rights should start with the inventors and make no assumptions or mandates with regard to disposition of rights.  Federal invention policy does not assume contractors will have a patent policy, and it does not attempt to dictate or construct what such a policy might involve.   Everything that is needed to comply with Bayh-Dole is contained in the SPRC, and the core of it is the (f)(2) agreement, one of the few bits that is in the implementing regulations and not in Bayh-Dole proper.

There is, of course, a fourth scenario:  Inventors refuse!  Agency requests title under (f)(2)!  University requests assignment of title from agency!  Election to retain title not meaningless!

Under this fourth scenario, even if the inventors refuse a request for assignment from the university, the university could arrange with the federal agency for title to flow via the (f)(2) agreement to the government, and then be assigned by the government to the university, subject to whatever conditions the government wished to impose (since the title would no longer be subject to the SPRC).  Of course, the federal agency could require the SPRC terms and conditions, and keep everything compatible with Bayh-Dole.   The key difference is that such a transfer of title is at the federal agency’s discretion, whereas a title transfer by the inventors to a pre-approved agent is not.  Furthermore, given that a federal agency, in obtaining title to a subject invention, is not obligated to share any licensing income with the inventors, it would appear that a university obtaining title would not have to do so, either, unless the federal agency required it.  Something, then, for inventors refusing to assign to a pre-approved agent might consider.

The effect, then, of this fourth scenario is to prod everyone back to the second and third scenarios–assignment is voluntary, because it is the best alternative, or is made under an (f)(2) claim brought by the university as the intended beneficiary (once it has elected to retain title) of the (f)(2) agreement made by the inventors.  In practice, then, no university patent policy is necessary to comply with Bayh-Dole.]

Here’s the thing.  It sort of doesn’t matter.  The SPRC *is* a federal agreement, and the university in agreeing to it is constrained in what it can do with its employees, because the university agrees to those constraints.  The university has to require the (f)(2) agreement.  If the university fails to get [does not seek, in addition to (f)(2),] a private agreement with its employees [regarding ownership of federally supported inventions], [or]if it has [merely] a research compliance policy that [generically] requires employees to comply with the terms of extramural contracts, then the (f)(2) agreement is just what the CFR says it is, an agreement to protect the government’s interest that the government can rely upon to obtain what it requires by way of title or license.  Note that an employee can comply fully with (f)(2) obligations without assigning any title to the university. [Note also that a university policy that employees must comply with the terms of extramural contracts does nothing for (f)(2) compliance, because (f)(2) is an obligation of the university to undertake, not the university’s research employees.  The policy would require employees to require themselves to make the (f)(2) agreement.]

With an (f)(2) agreement in place, when a university elects to retain title, then (f)(2) can be used by the university to require its research employees to comply with its own Bayh-Dole requests obligations under the SPRC], since these requests are within the scope of the government’s interests, as represented by the university[(f)(2) agreement].   When a university elects to retain title, it gains standing with regard to (f)(2) to obtain that title.  The government in essence assigns to [allows] the university, upon university’s written notice, the right to direct the disposition of title under (f)(2), subject to the SPRC’s conditions.  [The right to direct disposition of title is not quite the same thing as having a federal mandate to obtain title, or the vesting of title outright.  It is simply the deferral by a federal agency of the agency’s own interest in the disposition of title in favor of university actions.]

Of course, the university may have whatever other claims it wants to have on its employees.  The question is, when the university agrees to the SPRC, and it requires the (f)(2) agreement, does the (f)(2) agreement take precedence over any other agreements that the university may have with its employees?  From the employees’ point of view, the answer has to be yes, it does.  And we find this confirmed in university policy statements.   For the policy quoted above, we find this:

When Arrangements with Outside Organizations Override This Policy. Arrangements with outside organizations that propose terms which are exceptions to this Policy must be submitted to the President or Provost for review by the University with the advice of the Committee on Cooperative Research, Patents, and Licensing. If approved by the University the terms shall be binding upon all members of the faculty, staff, and employees of the University conducting such research or utilizing such facilities, and will supersede the provisions of the patent policy to the extent that the terms are inconsistent therewith.

Here is a simpler statement from another university patent policy:

Rights to inventions arising in the course of government or other externally sponsored research are controlled by the terms of the agreement between the University and the sponsor and/or applicable federal regulations.

These sorts of statements would make it policy to knock out university policy that conflicts with the federal agreements—both the SPRC and the (f)(2) agreement made by research employees supersede the university’s own policy and agreements.  It’s good that university policies do this, since it has to be done.  Federal agreements have precedence anyway.

What’s really odd is that the terms of the SPRC are binding on the university, not on employees of the university.  Even the (f)(2) agreement is required only of certain employees.  The government has no interest in the inventions of clerical and non-technical employees, and no interest in the inventions of non-employees.   Such inventions are not “of the contractor”.

A university accepting a federal patent rights clause is in an unusual position.  It cannot accept the SPRC and turn around and demand something else from its employees.  It is constrained by the SPRC, and it is constrained to require the (f)(2).   In the case of federal grants, it is also constrained by 2 CFR 215 (Circular A-110).  Those are the controlling agreements for federal research IP.  What a university can do is have agreements with its employees that implement various pathways permitted under the SPRC.

Thus, to be plain about it, (f)(2) controls for as much as (f)(2)’s required scope.  A university doesn’t have to have a policy about it.  It just has to do what the SPRC requires.  That includes requiring the reporting of subject inventions  and signing paperwork to file patent applications and establish the government’s rights.

Protecting the university’s interests comes later, after all this, and is not parallel with it.  That is, when the university elects to retain title, then there is a time to determine what actions research employees are to take with regard to the university’s interest.  Even then, the university’s interests are often constrained by 2 CFR 215.37 (requiring it to act as a trustee for IP it acquires in the course of federal research).

Again, it appears that everything that happens by way of employee agreements on title to inventions is covered by (f)(2).  There are various ways to comply with the (f)(2) obligations.  But it’s really simple stuff.  Where a university has a policy that requires assignment of inventions based on university interest, that interest is only possible as a result of the university’s actions under the SPRC to elect to retain title, and then the authority for obtaining title (or causing it to be assigned to a research foundation, or whomever) arises from the SPRC and the (f)(2) agreement then comes into play  If the university does not elect to retain title or fails in its obligations under the SPRC, then it has no standing to control title to subject inventions and the government can require assignment or march in and compel a license, following either the SPRC or the (f)(2) agreements as necessary.

In all of this, a university’s own patent policy is essentially set aside everywhere it conflicts with either the SPRC or the (f)(2) agreements.  The controlling document for employee commitments is the SPRC. Further, many university patent or research compliance policies expressly provide for this by making extramural contracts supersede the patent policy requirements.  And the SPRC certainly qualifies in this way.  What folks writing the policies don’t appear to figure on is that (f)(2) also qualifies.  And what the (f)(2) agreement requires is that employees protect the government’s interest in certain ways.

The focus of university subject invention practice is properly on implementing (f)(2) as required–with the obligations as stated, in writing.  Why is that so hard?  I don’t know a university that simply does this for each federal research agreement, even though there is a flood of paper generated for each grant.  (I have to sign off on grant budget reports monthly, confirming that it is proper for me to be paid on the grant I direct–let’s see, that will be on the order of 36 confirmations that I should be paid for my efforts as set out in the grant budget before I’m done with the grant.  Somewhere there will be a file drawer with these.  Is there any benefit to the public?  No.)  There is ample opportunity in the university administration of grants to require an (f)(2) agreement in writing by research personnel.

The upshot of all this is that university patent and research policies are superseded by the SPRC and (f)(2) when it comes to federal funding agreements, not because of federal law, but because a university agrees to the SPRC and is obligated to require (f)(2).   [In a sense, then, Bayh-Dole not only does not require a patent policy, but also sets aside competing aspects of university policies when universities accept federal funding agreements.]

This entry was posted in Bayh-Dole, IP, Technology Transfer. Bookmark the permalink.