While working on a new Guide to Bayh-Dole (one that COGR has yet to thank me for), I spent some time looking at recent patent infringement litigation initiated by universities. It’s not a pretty picture. In “Sue U.,” Jacob H. Rooksby found “more than 60 universities” participation in more than 280 patent-infringement lawsuits since 1973. A dissertation by Maria Teresita Barker at the University of Iowa found “171 patent-related lawsuits involving a college or university” between 1980 and 2009:
The 85 universities identified in this study were most often public research universities suing corporations for infringement in order to protect their rights in a university-owned patent. These corporations were most frequently competitors of a corporation with whom the university had an existing licensing agreement.
Among the flagship cases have been Stanford, which sued Roche, a company that had acquired technology assets from a company that had collaborated with Stanford, sharing technology and lab space. Stanford took its case all the way to Supreme Court, and lost. Perhaps it was only coincidence that Roche closed its 900 person lab in Palo Alto, laying off 500. Part of a restructuring and consolidation after its acquisition of Genentech, or something like that.
Cornell through its research foundation sued HP for infringement, after HP had supported research at Cornell. The suit has dragged on, with Cornell asking for $900m, getting awarded $184m by a jury, an amount that was reduced by the judge to $53.5m, and now nearly a decade on, after yet another is still on appeal, was settled in secret in 2010. As a direct consequence of the suit, HP terminated its research relationships with multiple universities. I was working with senior HP personnel during this time, and it was clear that as the result of Cornell’s action they had lost a political fight within the company to develop collaborations with research universities. In the document linked above, Cornell tried to spin the case as a hero’s effort to make evil corporations pay:
According to University Counsel James Mingle, “This case has had a powerful impact on university licensing efforts, as other companies also recognize the university’s resolve to protect valuable intellectual property produced by its faculty.
“If you knit the Cornell case together with several other successful suits marshaled by other major research universities — such as the University of California’s patent cases against Microsoft and Monsanto, the University of Michigan’s suit against Bristol-Myers-Squibb and the University of Wisconsin’s suit against Intel — the impact on tech transfer is nationwide,” said Mingle. “These cases, all filed as last resorts after licensing efforts failed, tend to bring balance to the bargaining table between universities that produce intellectual property and companies that want to use it.”
The idea that litigation is a “last resort” utterly ignores the idea that research inventions are intended to be used, not litigated. Use equals success. *Exclusion* is the “last resort,” not litigation for infringement. If there is use, then, why, there is no further need for exclusion, as far as federal policy on subject inventions is concerned. If some company holding an exclusive license is damaged by others’ use, even then one has to look at whether an exclusive license was warranted in the first place, whether the company holding that license has acted diligently to develop or use the invention, and whether a sublicense or cross-license would resolve the dispute and provide appropriate consideration. There are lots of possibilities, before one reaches a “last resort”–unless the last resort is simply how to make as much money as possible, rather than, say, transferring technology for public use.
The University of Illinois in 2011 sued Micron Technology, a company that has provided endowments for engineering professors in its association with the University. The suit is on-going, but the damage has already started. Micron informed UIUC’s dean of engineering that it was no longer going to hire UIUC graduates, was pulling out of job fairs, and like that. Well, UIUC filed a motion demanding that the court prevent Micron from doing this, and that got denied, too.
There is a whole list of others. The University of California with its exclusive licensee Eolas, a company without products, is out suing a lot of companies, including Facebook, Wal-Mart, and Samsung–even after its patents were apparently declared invalid. The University of Rochester with its exclusive licensee My Health is suing ZeOmega, Boston University continues a line of litigation against Samsung and Amazon by adding Apple as an additional target. The University of Washington is suing GE. WARF settled with Intel (after Intel had funded $90,000 in research leading to the invention), Penn is suing St. Jude Hospital, NYU with an exclusive licensee sued Abbot Labs got an award of $1.67b that was overturned on appeal, and now what? The University of Pittsburgh sued Varian Medical Systems (awarded $85.8m–appears Pitt may have had a point). CMU sued Marvell Technology Group (awarded $1.17b, under appeal, with CMU arguing for triple damages and Marvell for something closer to $9.9m). It’s like a bar room brawl in an old west movie.
Of course, we also have the University of Utah (along with Penn and others) and Myriad suing Ambry Genetics and Gene by Gene, two companies that propose to offer assays for BRCA 1 and 2 at one quarter the price Myriad charges. For that one, we have a US Supreme Court decision that invalidates some of Myriad’s patents, but not others. Now Senator Leahy has asked the NIH to consider march-in proceedings against Myriad on the basis that Utah and Myriad are not providing the benefits of federally supported inventions to the public on reasonable terms. The Senator might have also added, that the federal government is apparently also having to pay four times more than it otherwise would have to in order to subsidize the cost of these tests. It makes sense for the government to consider whether this sort of brawling is what Congress intended when it aimed to have Bayh-Dole “promote collaboration between commercial concerns and nonprofit organizations, including universities” and that “inventions made by nonprofit organizations and small business firms are used in a manner to promote free competition and enterprise” (35 USC 200).
There are some arguments that get put forward for patent infringement actions. One is that this is the right that comes with patent ownership, the right to exclude, so folks should not be shocked by infringement actions. A second is that if a university does not enforce its rights, then no one will respect those rights and then what’s the point of holding a patent. I have even heard some folks talk about how it is good to kick up a fuss now and then just to make sure everyone knows the university is serious about its rights. A third is that such litigation is a necessary adjunct of exclusively licensing valuable rights into competitive situations. The fact of life is that competing companies will try to get away with infringement, and they will do so until someone stops them, so you have to “call your own fouls” so to speak. Folks argue that infringement litigation shows just how valuable patents can be, with judgments against university infringers topping $1b. What follows on from such an observation is that if infringement litigation is how a university has to get industry to recognize the value of its research discoveries, amidst all the profits, then that’s how it is going to be. If companies would just take the licenses when offered, things wouldn’t end up in court.
There is something to each of these sorts of arguments, and there is also something that goes, why, if there are so many good arguments, then doesn’t that seal the deal? I’m not going to say that these are not all good arguments, but consider a few responses.
1. It’s a right to exclude, so don’t be shocked. True, but the objective of Bayh-Dole is to use the patent system to promote use, not to use the patent system to exclude use. So something is wrong. Furthermore, the federal funding agreement for grants to universities requires universities to be trustees looking out for the beneficiaries of the research. That is, the folks for whom benefits of inventions made with federal support should be available to on reasonable terms. Those folks are not just the end user consumers, but the whole chain of organizations that might benefit from use, from universities to government agencies to companies to all of us. Universities are held to a higher standard than any mere owner of patents, and ought to consider that in their decisions to take ownership of inventions, their decisions to license rather than assign ownership of inventions, and their decisions to license exclusively.
2. If we don’t enforce, we won’t be respected. Or, “litigation brings balance to the bargaining table.” There are variations that carry more weight: if we don’t enforce a standard, then the public will be hurt by inferior implementations that don’t work properly; if we don’t enforce, those that are working with us are put at a disadvantage relative to those that simply ignore the licensing effort. Those have more bite. But take a look at it another way–is the enforcement about money or about public welfare or free competition? If these latter things, then fine, it almost sounds like something a university should stick up for. But if it’s just about money, then maybe something is more deeply wrong with the mindset that got the university into the situation in the first place. You know, being on the wrong side of the tracks means one has to deal with it, but why are universities on that side of the tracks to begin with? Part of the “culture change” to put Porsches in university parking lots?
3. Litigation comes with exclusive licenses. The truism floats around university technology licensing shops that “a non-exclusive license is just a tax.” One of the arguments at the passage of Bayh-Dole, and heard frequently, is that the advantage universities (and other agents) had over the federal government is their willingness to license exclusively. It’s the exclusive license that is essential for “commercialization.” Without exclusivity, no one will invest in university inventions and they will sit on the shelves, wasting taxpayer money. Yes, well, like the digital computer and the internet and the web, which didn’t have exclusive positions and have done rather well. Sure, there are some inventions that may require exclusivity.
I can think of three classes. One class is those inventions that only one company might develop, and having done so, others could readily make copies without sharing in the recovery of development expenses. The free riders and all. The folks that want a share of the profits but don’t want to be into it for the risk of development. You know, like patent licensors (such as universities) who don’t invest along with the developer. Er.
The second class is those inventions that might not require exclusive licensing, but a company is willing to pay a premium to keep others from using or developing the invention in parallel, and competitively. This class would benefit from open competition, and might even result in more income over time (think Cohen-Boyer research tools), but a university might be attracted by a big, sure payout. In this class, the licensor is paying to prevent competition in use and development. Folks are willing to pay their own way. This is not a free-rider problem, but preventing any other transit on the route. Oddly, exclusive licenses here may slow down development rather than advance it, especially if the exclusive license is for the life of the patent rather than, say, no more than eight years.
The third class is inventions that probably have no purpose being licensed for product development exclusively and derive their value from preventing use in anticipate that the value of the patents will increase over time as the importance of the underlying inventions becomes evident. This is something of a gamble, because a new technology can take many possible paths, or none at all. But if the cost to acquire is low, then across a number of such assets, the likelihood of at least one of them becoming important enough to trip up a company downstream becomes pretty good, with an upside that’s way more than the cost to acquire. This is a portfolio approach, as distinct from an agent approach. In an agent approach, a university represents, for each invention, the interests of the inventors and public. In a portfolio approach you only need one winner every few years. Rather than rage against such an outcome, as one might do in an agent approach, in a portfolio, one makes a virtue of a few winners at the expense of the rest. While the portfolio approach is a time-honored way to make money, it does so at the expense of the actors in a market place, with the tut-tut that the actors should have been more diligent about avoiding the patent traps that have been set ahead of them. Is this sort of thing really part of what Congress intended when they passed Bayh-Dole? That universities should be partnered up with trap-setting speculators? It’s difficult to believe so.
While there may be arguments, then, for exclusivity, they are not nearly so general as they are made out to be. Universities rarely contribute the patents they own to a standard, and rarely think to release inventions at no charge, or at a charge that aims to recover the cost to obtain the patent or support the continuing research project that created the invention. There is little discussion at university TLOs these days of how it might be that some products arise because there is prior use, common platforms, freedom to explore and vary basic technology, and reason for dialog among competitors over interoperability and the like. That is, commercialization is not merely the process of granting an exclusive license so that a company can build a product in isolation at its own great expense, but rather that commercialization may take many other forms, including companies sharing a common platform long enough to find the most favorable implementations from which to build products, competing on price, availability, quality, service, and brand, among other things.
All things being equal, a non-exclusive license is not a tax if the university does not charge for it, or does not base the charge merely on a monopoly position. Consider: a university might hold inventions so that everyone can use them, and patents on them can be cross-licensed to those making improvements so that those improvements, too, are available to a user community. For that service, it may be that competitors in a market might see value in paying a non-exclusive license, as it represents a service to industry: providing a platform to encourage both collaboration and competition. Similarly, a non-exclusive license might be bundled with support services and improvements, so that companies pay for the relationship with the laboratory and its experts, not merely for an isolated patent right, which may be part of the relationship, but not what carries the primary value.
The point of these comments is that one does not have to set up a licensing program to take advantage of companies ready and able to use an invention, simply to extract additional value by recruiting one company, perhaps fronted by speculative interests, to disrupt the interests of the rest.
4. University inventions really are valuable and companies should recognize that. This is so true. There a lot of inventions arising in university work that truly are valuable. That’s why the public accepts so much of their tax dollars will go to support that research in the first place. But that value is not necessarily financial, and even when the financial value is high, it does not necessarily accrue to a patent holder, whether the inventors, the university, or anyone else. University faculty in seeking federal funding have chosen this outcome. It would appear there are times when the overall value of an invention (take the sum of everyone who can use it, and everyone who can benefit from its use, times the value to each over time) is diminished by a single organization willing to take pennies on the dollar of the value, simply for the pennies.
Yeah, I know. Companies don’t have altruism baked into their corporate genes. But why is it that universities have worked to eliminate altruism from their invention management? Again, is that the culture change Congress was hoping for in passing Bayh-Dole? The secret agenda: “Oh gosh we hope that as a result of this bill, universities will give over their load of carp about serving the public interest and just get in there and grub for every dollar they can squeeze from the patent positions we have given them the chance to take, their old fluffy value statements be hanged.”
In a world where inventors invent stuff that’s valuable, it stands to reason that they would like to see that value recognized. That makes all sorts of sense. The sense varies from inventor to inventor, from research director to bottle washer. A refined sense might also recognize, though, that the first invention may not be the one that carries value. It may be the third or the thirty-third. It may be that a user community must form first, or some development efforts have to try and fail before the right mix is found. If every invention is treated for its exclusive value, and cannot be allowed to serve as a secondary asset for something else, and cannot be included in a mess of such inventions that support the formation of a functional product, like, say a 3d printer, then the exclusive practice fragments the opportunity, and encourages a white-knuckled effort to run up value over a readiness to encourage, in the expectation of later returns.
More to the point: while an inventor may seek whatever value he or she desires, when a university takes ownership, the situation changes. The deal is not that the university can outgreed any faculty inventor and therefore serves everyone’s interest by demanding ownership to prove the point. The university serves entirely different purposes. If money is the driving thing, and especially as much money as a monopoly position in a research invention can produce, then the university should refer the inventors to other places for management. If there’s going to be heat for greed, focus it on the inventor and agent who make that choice, not the university and the rest of its faculty, not as a matter of policy, and certainly not as a great virtue of a so vastly improved academic culture.
Finally, it is worth noting the adage from the military: don’t give an order you know will not be obeyed. If one is holding a monopoly right, and proposes what most organizations find to be unreasonable terms, what do you think those organizations are going to do? Present them with a reasonable set of terms, and yeah, they should take them and not screw around, and if things work out, inventors ought to get plenty rich, or whatever it is they dream of. For all that, it’s good when inventors offer to share proceeds with a university, or better, with their community, rather than university administrators trying to make it out to be institutional generosity to share some piddling amount back having demanded ownership of the invention.
* * * * *
These comments don’t overturn the arguments for university litigation for infringement, but they ought to give pause for thought about appropriateness. There are times when it just comes to it that a company violates good faith and aims to screw everyone else, and fine, there is poetic justice, and sometimes it falls to a university to serve up a whooping. But also infringement litigation can be used to create a false value, such as by demanding use or sale of product be blocked rather than simply asking for fair and reasonable terms. I know, it’s not a false value for just any invention holder, but it is a false value for a university: it is false to a university’s position in a community, it undermines the public trust, and works against the university’s mission and support.
Bayh-Dole should be amended to reduce university use of infringement litigation.Rather than eliminating prior user rights when a university owns the invention, as was done for some fool reason in the AIA, it should be the other way: if a university owns a federally supported invention, then everyone has a royalty-free make/use right to practice the invention, and the university’s opportunity is to manage the sell right only. If that’s not good enough, then leave the inventors to make their own choices, don’t take ownership, and don’t propagate litigation choices throughout the institution.
As it is, companies hit by university infringement litigation, after taking inventory of their own moral compasses, should look to whether the patent in dispute covers an invention made with federal support. If so, then the university should have to prove that it has fully complied with the patent rights clause of the funding agreement that controls disposition of the invention before the university can sue for infringement. This is not about march in, but about standing to sue in the first place. The court should also look to the objectives of the Bayh-Dole Act at 35 USC 200 and require that the university show that their licensing practice for the patent in dispute has met those objectives, and that the suit, if the university prevails, is likely as well to meet those objectives. Show that. Then it’s all good, rip away, if that’s what university administrators believe is the new public purpose of a university.