AAU Fools with Words, 2

Let’s look then at the AAU statement produced by the task force charged with–if we read the preamble correctly–finding better words to declare that university technology transfer operations are to advance the public interest.

AAU starts with fake history:

Before Bayh-Dole was enacted, the federal government retained ownership of federally funded discoveries,

Before Bayh-Dole, there was the IPA program, under which scores of universities and nonprofits could secure ownership of inventions made with federal support. Both the NIH and the NSF had IPA programs. Other federal agencies, such as the Department of Defense, allowed contractors to own inventions made under federal contracts. It’s nonsense.

But it’s worser nonsense. Where the federal government sought ownership of inventions, it did so by requiring assignment, not by “retaining” ownership. The government did not have any ownership of patentable inventions to retain. Here, the AAU task force shows that it is clueless with regard to history–or relying on the same faux historians of Bayh-Dole that populate the country rather than doing their own research, which in the course of a half an hour would show how inexpert (or deliberately deceptive–take your pick) the faux historians are. The government obtained inventions the way anyone must–by assignment. For instance, the DoD could, if a contractor declined an interest in an invention, require (as a contract condition) that inventors assign their inventions to the government.

In two areas the federal government made claims to inventions that precluded private patent ownership–for atomic and nuclear energy and for space technologies. In these cases, special statutes gave the government ownership of inventions outright, and the government could file patent applications without the need for an inventor’s assignment by including an affidavit citing the funding, the area of the invention, and the law. Even now, the federal government can suppress patent applications for national security purposes. No, I can’t talk about it.

What the AAU task force might be referring to without having the courage (or is it awareness) to do so was the Public Health Service policy, consistent with the Kennedy patent policy, of requiring assignment of inventions in areas of public health such as medicinal chemistry where the invention should (in the judgment of the Public Health Service) be made available to the public and not turned into a monopoly. That’s a debatable position, but it was not a government-wide position. It was a special, special case–special: inventions that can only be developed for use with private investment; special special: special inventions directed at matters of public health. And even in the special special case, until 1978, universities could acquire invention rights ahead of the government’s claim and attempt to license special special case inventions to the pharmaceutical industry for “development.” We might say there was a huge policy dispute between the PHS and the pharmaceutical industry over the area of research called medicinal chemistry (and now all sorts of things–genomics, proteomics, biochemistry, nanotechnology, bioinformatics–anything but “medicinal chemistry”).

but the government often neglected to license those discoveries to the private sector for further development.

More made up stuff, starting with “often neglected to license.” The Kennedy patent policy made clear that when the government obtained ownership of inventions, it should “dedicate or license” the inventions for broad public use. Dedicate–don’t patent at all. License–license non exclusively, royalty free, but perhaps control for quality and safety and accurate claims.

Most of the 28,000 patents held by the U.S. government by 1978 were of this sort–ones that a defense contractor did not want to pursue and which the DoD might want to hold to 1) recognize technical merit; 2) introduce the invention into the patent literature to raise the bar to further patenting; 3) control to level the field for competitive bidding; 4) to regulate foreign competition with U.S. vendors; 5) to permit broad use without the bother of formal licensing. None of such purposes requires “commercialization” by means of a monopoly. Where the government did hold patents in medical inventions, its licensing record for commercial development was comparable with the best rates at universities–and nearly 5x better than the university patent agent rates for inventions handled under the IPA program.

Bayh-Dole sparked technology transfer by creating an incentive for universities to secure patent protection for inventions resulting from federally funded research;

Technology transfer was sparked long before Bayh-Dole. It was sparked by Cottrell’s creation of the Research Corporation in 1912, and in a different way by the formation of the Wisconsin Alumni Research Foundation and scores of similar university-affiliated research foundations. It was sparked by the University of California’s decision in 1960 to operate its own patent licensing program, followed by MIT and Stanford. It was sparked by the Public Health Service IPA program, which from the mid 1950s to 1978 allowed universities to selectively divert patentable inventions for agent or university management. So more fake history.

Where, too, is “the incentive” for universities to “secure patent protection”? There is nothing about an incentive for universities in Bayh-Dole’s statement of policy and objective. There is nothing about “incentive” in the provisions that must be included in the standard patent rights clause. The closest one might come to an incentive is that if a university manages to obtain assignment of an invention made with federal support, the university has the means to “retain” that ownership, provided it jump through a bunch of administrative hoops. Even in the use of royalty and other income from exploiting an invention, Bayh-Dole creates restrictions for universities and other nonprofits. They can use such income only to pay inventors, to recover costs of managing subject inventions (not other inventions, not non-inventions, not administrative slush), and to use any balance for “scientific research or education”–not for, say, public service or humanities research. No incentive there–encumbered money.

Perhaps the AAU means by “incentive” an “incentive to make money by dealing in patents.” If so, then we have a dilemma–how can the federal government by law give universities “an incentive” to make money from patents on federally supported inventions and then have critics complain about this incentive? Otherwise, just what “incentive” is there that the AAU argues is in Bayh-Dole? Or, perhaps here the AAU is just making up something fake that sounds good–there is no incentive in Bayh-Dole, but it sounds good to say there is an incentive to explain why university administrators are so enraptured with being in the patenting business. If the incentive offered by Bayh-Dole is that universities can be in the patent business to make money, then surely (so the AAU reasoning must go) it advances the public interest when universities make money.

this, in turn, allowed businesses to gain the necessary rights to develop and commercialize those research discoveries.

The “this” points backwards indifferently–it could point to “Bayh-Dole” or to “incentive” or to “technology transfer” or to “patent protection” or to the whole complex of creating, sparking, and securing, or to the assertion that the depiction of the whole complex is accurate. Perhaps at this point the AAU task force thinks its work is done and it can write indifferently. But whatever the “this” refers to, somehow it was this “this” that “allowed businesses to gain the necessary rights.” Let’s parse that claim. It gets interesting.

The assumption AAU starts with is that universities gain ownership of discoveries made with federal support. According to the AAU, it’s Bayh-Dole that gives universities a mysterious incentive to take ownership of inventions, for which they then come, also mysteriously, to have a responsibility to be “good stewards.” Now, however, we find out that universities gain ownership of inventions so that they can allow businesses to “gain the necessary rights.”

Since previously, according to the AAU’s fake history, the federal government owned inventions made with federal support, we might think that the “necessary rights” were any form of permission to use these inventions. But no. These are “necessary rights” not “sufficient rights.” A sufficient right would be a non-exclusive license, or an invention that entered the public domain. In the AAU’s fantasy regarding innovation, no company is willing to use an invention that isn’t subject to a monopoly claim. But also in the AAU’s fantasy–and it is fantasy–no company thinks of merely using an invention. An invention must be “developed” and “commercialized”; that is, become a commercial product. For this activity to take place, the license must be, apparently, exclusive.

What does it mean to “develop” an invention? What is the distinction between “research” and “development” in play here? We might observe that it’s quite possible there is no meaningful distinction between research and development in terms of behaviors–people study, observe, document, test, change, and repeat in both activities. Perhaps they don’t even differ in their intentions between research and development. It may be that the only differences are in the expectations of people who watch or hire or direct people involved in research or development.

R&D Magazine a couple of years ago ran an article by Bradford L. Goldense about research and development (whoa!) and lays out the typical linear model scenario of basic research, applied research, advanced development, and product development–setting aside “skunk works” (and thus, the Vannevar Bush innovation engine and alternatives to the linear model). Goldense’s point is that this conventional approach is being replaced by “any organization can launch a product.” According to Goldense, basic research enables possibilities, applied research tosses what’s impossible and impractical, and then we get “advanced development”:

Advanced Development generally takes these possible and likely feasible solutions and further reduces the risk in hopes of culling out the best alternatives to deliver to an expressed target for a capability or feature to incorporate into products. Downstream manufactured cost considerations start to come into play as a culling consideration.

In this usage, “development” means “ruling out” with an eye toward market and production. An invention gets developed by ruling out a bunch of variations of the invention. In terms of invention and claimed invention, “development” means deciding not to use or allow the use of swaths of what has been claimed in a patent as “the” invention. In this sense, “development” means deciding not to use versions of an invention. A patent allows the patent owner to enforce this decision on the rest of the country, so that only what “advanced development” has chosen may be considered for use and commercial product.

This gets interesting, then. When a university takes over an invention made with federal support, it must be that a “necessary” right in play is that of preventing any others from making, using, or selling those parts of a claimed invention that a company decides must be suppressed. “Develop” must mean, here, “suppress.” Contrast this usage to Bayh-Dole’s happy statement of policy: “use the patent system to promote the utilization of inventions.” A university, in licensing necessary rights, agrees to collaborate with the licensee to prevent the use of all portions of a claimed invention that the company chooses to suppress. Such suppression prevents competition with the company’s technical and marketing decisions. It’s not as if a university says, “Company, you get first dibs–define your product from the claimed invention and we will give you an exclusive license to exactly those portions of the claimed invention that you choose. Everything else we’ll offer to your competition.” Thus, for the compound sequence licensed by UCLA to Medivation, the company might choose compound 169′–and that would be the scope of Medivation’s license, with UCLA offering all other compounds within the claimed invention to others. But that of course is not how things work. University exclusive licenses–usually made to function as assignments–form an agreement to permit the licensee/assignee to enforce the patent rights without reference to defending a commercial product from copying–the defense is general, extending to all those elements of the licensed invention that the company has suppressed.

It takes some work to get past euphemisms like “develop” to begin to see what is in play.

There’s more. Let’s look at Goldense has to say about product development:

Finally, Product Development invents what’s needed and necessary “now” and packages both the form and function of all feasible and risk-reduced features and/or capabilities into products planned for release to the marketplace.

It’s that word “invents” that should stop you in your reading tracks. Product development “invents what’s needed and necessary ‘now’.” In this usage, development starts with the narrowed possibilities and invents the implementations to create products. That is, whatever the purpose of the original patent licensed by a university, when a company comes to “develop” a product, it will (in general) do more inventing. Consider–the new inventing may lie within the scope of the university’s patent, or the new inventing may be outside the scope of the university’s patent. The product itself may even lie outside the university’s patent. It’s entirely possible that a company will be motivated to “develop” a licensed invention so that the resulting product is outside the scope of the license.

Here’s an illustration.

Imagine a university patenting office as a UPO hovering over a strange planet, such as Earth, shining a patent beam down on an area of technology. In this patent beam, nothing can grow but for the grace of the UPO operator. The patent beam is the scope of the claimed invention–all the claims map out what the patent owner can exclude or suppress. But with development, a company might use only the little bit labeled “A.” That little bit is still within the scope of the UPO’s patent beam, so the license matters, and if a product is created, the company will pay royalties.

But the company might also develop the product around a different little bit, labeled “B.” For that part, the company might have its own patent positions–and actually B could be part of a big area of claimed invention. Furthermore, B could represent improvements to A that anyone hoping to practice A would have to also use to make something that works for a given application, with satisfactory function, reliability, safety, and cost. The company could rely on the licensed patent and merely select bit A. Or the company could invent a new bit B and practice both A and B. Or the company could drop A and rely only on B.

University licensing folks actually fear the B. They write their exclusive licenses to require diligence in “developing” the licensed invention. But they cannot prevent a company from also developing its own inventions that lie outside of the scope of the patent claims in the licensed invention. A company might, then, take an exclusive license from the UPO while developing its own bit B. The UPO license prevents the UPO from finding anyone else to develop the invention. That is, no research competition. Even with a diligence clause, a university will find it difficult to punish a company for developing a product outside the scope of the license–as long as the company makes a show of also attempting to create a product within the scope of the license, it will be next to impossible to prove short of lengthy discovery in litigation, that the company had no intention of ever building a product within the scope of the licensed patent.

No company I know willingly builds product within the scope of a licensed patent when they can build equivalent or better product outside that scope. Universities protect themselves against this outcome in various ways–one way is to charge a big up front fee, perhaps payable in stages. Even if the company fails to build a product within the scope of the license, the university makes some money. Another way is to take an equity position in the company, so that what matters is not whether the company builds a product using the licensed invention but rather whether the company’s value increases over time. In taking an equity position, a university all but announces that the use of the licensed invention doesn’t much matter.

Part 3 is next. 


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