Reagan’s Executive Order 12591-1

[Some slight elaborations here and there. 1/6/2020.]

In 1987, President Reagan issued Executive Order 12591, which extended, sort of, the Bayh-Dole Act to all federal contracting. Let’s work through EO 12591 and try to sort out this out. It’s rather a mess. In doing so, we end up with something remarkably simple and direct to say about Bayh-Dole. It is worth working through the details to show how it comes about.

Here’s the relevant bit of Reagan’s EO 12591, with some parts in bold to mark points of interest:

(a) The head of each Executive department and agency, to the extent permitted by law, shall encourage and facilitate collaboration among Federal laboratories, State and local governments, universities, and the private sector, particularly small business, in order to assist in the transfer of technology to the marketplace.

(b) The head of each Executive department and agency shall, within overall funding allocations and to the extent permitted by law:

(4) promote the commercialization, in accord with my Memorandum to the Heads of Executive Departments and Agencies of February 18, 1983, of patentable results of federally funded research by granting to all contractors, regardless of size, the title to patents made in whole or in part with Federal funds, in exchange for royalty-free use by or on behalf of the government;

Let’s go down the rabbit hole. Reagan’s February 18, 1983 Memorandum, however, does not mention commercialization. Here:

To the extent permitted by law, agency policy with respect to the disposition of any invention made in the performance of a federally-funded research and development contract, grant or cooperative agreement award shall be the same or substantially the same as applied to small business firms and nonprofit organizations under Chapter 38 of Title 35 of the United States Code.

Other than that Bayh-Dole is now Chapter 18, not Chapter 38–there is no chapter 38of federal patent law, the memorandum extends at least some Bayh-Dole treatment to any contractor not otherwise specified by Bayh-Dole–any contractor other than a small business or nonprofit. That means foreign contractors and large company contractors, if they acquire title to an invention made under federal contract, can also choose to keep that title, subject to the requirement that they disclose the invention to the federal government (and the rest of the patent rights clause requirements in 35 USC 202(c) and the rest of Chapter 18), unless not permitted to do so by law (i.e., any of the Acts that Bayh-Dole asserts precedence over but presidential executive orders cannot).

But there are big differences between Bayh-Dole and Reagan’s Memorandum. Whereas Bayh-Dole expressly preempts various specialty laws (“Acts”) dealing with the ownership of federally supported inventions, Reagan’s Memorandum is limited to treatment of contractors “to the extent permitted by law”–that is, to the extent permitted by those specialty laws that Bayh-Dole otherwise preempts for small businesses and nonprofits. While Bayh-Dole preempts most specialty laws, those same specialty laws preempt Reagan’s Memorandum.

The effect of Reagan’s Memorandum then is to amend the Nixon executive patent policy, which in turn amended Kennedy’s policy. Nixon’s patent policy, however, required codification.  One would expect that when Reagan amended the Nixon policy, the amendments would be subject to the same codification requirement. Reagan’s change skips the new codification step because it expects the codification done for Bayh-Dole to work for it, too–which is just what NIST tried to accomplish by conflating Bayh-Dole’s “nonprofit organizations and small businesses” with everyone else and referring then to just “businesses.” But doing so is nonsense, because Bayh-Dole preempts most other federal statutes and Reagan’s Memorandum expressly does not. The upshot: if an invention is *not* a subject invention under Bayh-Dole, and no federal statute applies to it, then executive branch patent policy applies–back to the non-preempted version of Bayh-Dole’s regulatory implementation–that is, 37 CFR 401–*not* Bayh-Dole the statute. Yes, two legal regimes–Bayh-Dole and Reagan–both use the same regulatory apparatus but with different import but made to look like they are exactly the same. Strange? Yes.

Bayh-Dole and Reagan also differ on their statements of policy and objective. They are not identical. This makes for conflation, non-compliance, and overreach. For instance, Bayh-Dole says nothing about promoting commercialization or transferring inventions to “the marketplace”–that’s something specific to the executive order. Even the preliminary statement of the executive order mentions only “the development of new products and processes”–obliquely commercialization, though neither products nor processes have to be commercial in nature. A look at 35 USC 200, Bayh-Dole’s statement of policy and objective, shows that the primary objective is to promote the utilization of inventions arising from federally supported research or development. The threshold for march-in compulsory licensing is practical application–use of inventions with benefits available to the public on reasonable terms, not commercialization. There is in Bayh-Dole no mandate for “commercialization” and no provision for dealing with contractors that fail to commercialize. True, there is a reference to commercialization in 35 USC 200, but it pertains to the promoting American industry and labor in the commercialization and public availability of inventions–if there is commercialization or public availability, then either should involve American manufacturing. (This requirement finds its expression in the unbearably weak provision of Bayh-Dole (35 USC 204) that only in cases of exclusive licenses to use or sell in the United States, the exclusive licensees must agree to source their product substantially from American manufacturers unless a federal agency waives the requirement.)

But Reagan’s executive order mandates the promotion of  “commercialization” by means of contractor ownership of, um, patents (not inventions):

by granting to all contractors, regardless of size, the title to patents made in whole or in part with Federal funds, in exchange for royalty-free use by or on behalf of the government;

I have highlighted oddities.

title to patents

Bayh-Dole does not grant anyone title to anything. The Supreme Court in Stanford v Roche (2011) made that clear enough even to scores of university lawyers who insisted that the law meant otherwise, even if it didn’t read otherwise. Bayh-Dole applies only after a small business or nonprofit contractor has acquired an invention made with federal support. Neither Bayh-Dole nor Reagan’s Executive Order grants title to inventions.

granting to all contractors

Furthermore, federal agencies don’t have any title in such inventions to grant. Inventors own their inventions. If a contractor doesn’t own an invention, then at least initially the inventor does, not some federal agency that handed out money. A federal agency might release a claim on title to an invention as a contract deliverable, were a contractor to obtain ownership or even if not, but how could any federal agency come to own inventions (or patents) made under contract in the first place, to be able to grant title to anyone? To be able to grant title, a federal agency would have to first obtain assignment of the invention (or there would have to be a law that made express Congress’s intention to make a “sea change in intellectual property rights” (as the Supreme Court in Stanford v Roche put it).

If Reagan had wanted an operative Executive Order here, he would have insisted that government-owned, government-operated federal laboratories to take ownership of all inventions made under cooperative research or development agreements by contractors other than nonprofits and small businesses, and then require the federal agencies to assign the title they have obtained back to the contractors. That would almost work, except that Bayh-Dole stipulates how federal agencies may deal in inventions owned by the federal government, and assigning title outright is not there (though granting an exclusive license with the rights of ownership is).

Bayh-Dole’s fundamental gesture is that small businesses and nonprofits can elect to retain title in the inventions that they acquire and which were made with federal support. They receive no grant of title from the federal government, nor from Bayh-Dole. They may notify the federal government that they will not give up title, and the government is conditionally required to acquiesce in the contractor’s choice. Nothing more.

On the other side of Bayh-Dole–the part having to do with the disposition of federally owned inventions–ownership of inventions made by federal employees is dealt with under a different regime, based on President Truman’s executive order 10096 and codified by 37 CFR 501. Bayh-Dole, by contrast, does not deal with how contractors obtain ownership, but only what federal agencies must do if a contractor or the federal government does obtain ownership, and what a federal agency can do if a contractor gains ownership but finds it doesn’t want that ownership, or screws up in the administration of that ownership. Similarly, Bayh-Dole on the federal side does not stipulate how a federal agency acquires ownership of inventions made by federal employees, but rather deals with how federal agencies may license the inventions that they do acquire.

But here, in the passage quoted above, Reagan’s executive order appears to require federal agencies to grant title to all contractors in patents that federal agencies do own. That would be a strange requirement as written, as there would be many co-owners, then, of each patent–all contractors. Or, more specifically, what is intended when multiple contractors work on the same project–in a consortium, or under a prime contractor that distributes multiple subcontracts? And how strange to restrict the grant of title to contractors and not everyone–why should contractors have any special priority for commercialization? Contractors are chosen based on their capacity to conduct research or develop products to government specifications, not to develop commercial products for a generic “marketplace.” The companies that might excel in developing commercial products may not be at all the ones that excel at bidding for federal research contracts.

We are left with the idea that the executive order does not intend what the words say, but there’s little to establish that intent but to say the executive order has made a hash of things and a plausible thing to expect is something rather different than anything put forward by the words.


The Reagan executive order focuses on patents, not inventions. The subject matter of Bayh-Dole, however, is patentable (or plant variety act protectable) inventions. While Bayh-Dole expects the patent system to be used in a relatively narrow way–to promote the utilization of inventions rather than, say, exploiting the patent monopoly right to maximize income–the Reagan executive order deals in patents directly. That’s an odd switcheroo. Under Reagan’s executive order, contractors do not hold title to patents until federal agencies grant them that title. That makes little sense, even if technically the Patent Office is in the Department of Commerce, which then might be thought of as in a way granting title to patents when it issues patents. But there’s no need for an executive order for such a thing.

made in whole or in part with Federal funds

Reagan’s executive order provides its own scope for the patents that are affected:

patents made in whole or in part with Federal funds

This scope is narrow compared to Bayh-Dole. Reagan’s policy applies to *patents* not *inventions*, and further applies only when the patents are “made” using at least some federal funds. Literally, this would mean using federal funds to pay attorneys to draft patent applications and to pay USPTO fees. Ranging further afield, making “patents” mean “inventions,” we are still left with a requirement to establish that federal money was used to *make* the invention. Either someone is hired expressly to invent what has been invented or money was used to actually reduce the invention to practice–building a prototype or conduct testing and the like. That’s *nothing* like Bayh-Dole, where the issue is first what the full project was, and whether federal money supported any part of the project in which an invention has been made.

The Nixon patent policy gives a somewhat different scope:

inventions or discoveries made in the course of or under any contract of any Government agency

At least the Nixon patent policy specifies “inventions or discoveries”–tracking the statement of scope in federal patent law. “in the course of” makes clear that the scope is not merely what was specified to be done but also what has come about as a result of the actions to accomplish what was specified. Broader than Reagan. It’s like the Reagan policy was drafted by incompetents.

“Made” has this meaning in the Nixon patent policy:

the conception or first actual reduction to practice of such invention in the course of or under the contract

There’s a tendency to attempt to harmonize the two texts, to settle on a meaning that might be approximately shared by both the Nixon policy and the Reagan policy. But if there were the case, there would be no particular reason for Reagan to state a scope. It would be enough to write that federal agencies should adopt the same protocols for large companies that they use for small companies, or that they should use Bayh-Dole’s standard patent rights clause and procedures for administrating it in all funding agreements with contractors where no other federal law requires otherwise. But harmonizing the readings requires that we already know that Reagan intended that the executive order merely recite Bayh-Dole. Otherwise, we are left to consider that Regan may well have intended something different, for clearly no executive order can countermand federal statutes–so while Bayh-Dole may preempt both statutes and the Nixon patent policy (and therefore even the Reagan revision of Nixon), the Reagan executive order cannot preempt those other statutes, nor can it preempt Bayh-Dole.

In the Nixon patent policy, there’s no “in whole or in part” language to the scope of interest. Reagan’s addition of “in part” changes things considerably. If an invention is made “in part” with federal funds, it must have been also made “in part” with support from non-federal sources, if only one’s personal time and effort. Reagan therefore substantially expands the scope of federal interest in inventions made with federal funding while at the same time Reagan restricts that scope to only those inventions for which some federal money was used in their making. It is all very strange, then, because the overall effect of the executive order is to concede that federal interest in inventions to contractors, but only for those funding agreements outside Bayh-Dole (not with small companies or nonprofits) and for patents (meaning, perhaps inventions) otherwise under the Nixon patent policy (because they meet the scope of interest set forth by the Nixon patent policy as modified by Reagan’s executive order) and not under Bayh-Dole (because not owned by a party to the funding agreement).

What about Bayh-Dole’s treatment of the scope of interest in inventions? Bayh-Dole has a different statement of scope than that of the Nixon patent policy or Reagan’s revision. There’s a whole book to be written about the matter, but the short of it is this:

35 USC 201(e):

The term “subject invention” means any invention of the contractor  conceived or first actually reduced to practice in the performance of work under a funding agreement:

That is, in Bayh-Dole, an invention has to be owned by a contractor before an invention is a subject invention. In Reagan’s policy, like that of Nixon, an invention merely has to be made with federal funds. Now consider “the performance of work.” Here’s Bayh-Dole:

The term “funding agreement” means any contract, grant, or cooperative agreement entered into between any Federal agency,  other than the Tennessee Valley Authority, and any contractor for the performance of experimental, developmental, or research work funded in whole or in part by the Federal Government.

Shortened we have “funding agreement means any agreement for the performance of work funded in whole or in part by the federal government.” Put the definitions of funding agreement and subject invention together, and one sees that subject inventions are those arising in a given “work”–regardless of whether one can show that federal funds were specifically used to make the invention. That’s the necessary implication of “in whole or in part” used to modify “performance of work” rather than “invention made.” In Bayh-Dole, it is not that an invention is funded in whole or in part, or even as Reagan has it, a patent (which makes no sense, or silly sense)–rather, it is the work that is funded in whole or in part. “Under a funding agreement” then means for Bayh-Dole “in work funded at least in part by the federal government.” The invention could arise anywhere in that work–so long as some part of the work is funded by the federal government. The invention does not have to be funded at all with federal funding–it is enough that the work in which the invention has arisen is funded at least in part by the federal government.

This is a huge difference. In the Reagan policy, one looks to see whether federal funds were expended in making a given patent, er, perhaps invention. In Bayh-Dole, one looks to see whether an invention has been made in a project for which the federal government has provided at least some funding to the project, even if the invention was not made with the direct use of any federal funds at all. Instance: a federal contractor recognizes that an area of federally funded research is ripe for invention and moves that work to a non-federal budget. An invention arising in that shifted work, therefore, would not be made with federal funding, but would be made within the project that has received federal support. Bayh-Dole protects against such practices (see 37 CFR 401.1 for the convoluted guidance). Reagan’s policy does not. But perhaps there is no matter in the difference, as Reagan’s policy stipulates that contractors may keep what they want to keep unless federal statute says they can’t, and if Reagan’s policy doesn’t claim something only to release it to contractors, then contractors can keep that unclaimed stuff, too.

in exchange

Reagan’s executive order sets out an exchange: federal agencies grant title to patents to contractors in exchange for royalty-free right to use.

This is deeply strange. Get past the problems with federal agencies gaining title to patents, or to inventions, in the first place. Imagine magic happens and federal agencies own inventions that inventors would otherwise own. The executive order stipulates that agencies are to grant title to contractors (that is, pass a thing of value to contractors) and the contractors as consideration for this thing of value grant back to the federal government a tiny portion of that thing of value.

The consideration for granting the whole thing is a tiny part of the thing. It cannot be. I give you a hundred dollars in exchange for one of those dollars back. You are paying me with what I gave you. There’s no exchange of one thing for another. It’s just a division of what I already have.

It’s much worse than this in Regan’s executive order, however. Under Bayh-Dole, there is no such “exchange.” Contractors that acquire title to inventions made under federal contract are permitted to keep that title, on the condition that they disclose each such invention. No exchange–just a condition upon which a federal agency is prevented from requesting title. If a contractor fails to disclose, then the federal agency may request title.

Bayh-Dole specifies that in the provisions for a patent rights clause, for each invention that a contractor elects rights (i.e., elects to retain title), the federal government “shall have a nonexclusive, nontransferrable, irrevocable, paid-up license to practice and have practiced for or on behalf of the United States.” The scope is “practice and have practiced”–that is, “make, use, and sell,” as the term “practice” was defined in the Kennedy executive branch patent policy and replaced with “make, use, and sell” directly in the Nixon revision to Kennedy–and the policy and its codification on which Bayh-Dole is built. The Bayh-Dole government license is much broader than that of the Reagan executive order.

And there’s no exchange. It is not that the government lets the contractor to retain title in exchange for permission to use the invention–it is that the government waives its claim to receive title on the condition that the contractor agree it has no standing to file a claim in the Court of Federal Claims for compensation for the use of the invention by the federal government (for which see 28 USC 1498) or for any other practice of the invention. Again, no exchange. Just stipulations, conditions placed on contractor’s continued ownership of inventions that otherwise might be considered contract deliverables.

This point is difficult. We are into the technical innards of federal patent policy. You’re welcome. Enjoy the stink.

This entry was posted in Bayh-Dole and tagged , , , . Bookmark the permalink.