Psalm 23 includes the lines:
Yea, though I walk through the valley of the shadow of death, I will fear no evil: for thou art with me; thy rod and thy staff they comfort me.
Through some strange time warp, the “valley of death” has become a stock motif in explanations of university technology transfer practice. University patent administrators use the valley of death (leaving out the “shadow” bit, as perhaps too complicated for their audiences) motif to justify acquiring patents and dealing in exclusive licenses. In this schema, university patent administrators set themselves up as the benevolent deities shepherding university inventors through this evil valley, relieving their fears, and bringing them to accept and even to love both the protection and discipline they provide. Tech transfer officer as the Lord Shepherd. It is little wonder why the image is so popular. University inventors, however, sing a different song on the same theme. I’ll get to that.
There are plenty of depictions of this terrible place, the valley of death. Do a Google image search. Or use DuckDuckGo. It’s just that I’m not sure that this evil place exists, or if it does, that it is relevant to a discussion of technology transfer, or that even then that it has the properties attributed to it. Mostly the valley of death is an artifact of badly chosen methods of managing inventions and associated intellectual property. While the valley of death is set out as a material problem in the operation of the Linear Model of research innovation, it appears mostly to be a consequence of how universities seek to make the horribly limited and suspect Linear Model “work.”
When a university takes ownership of an invention, and applies for a patent, it creates a triple barrier to access. Before anything else can happen, the university first must overcome this barrier of its own making–and that often proves impossible. First, there is the patent itself. The patent is programmed to exclude. That’s its purpose in life. It takes real creativity to adapt an asset designed to prevent others from doing something so that it can be used to help them do that same something. Universities used to be up to this sort of creativity, and talked openly about dedicating inventions to the public domain, and licensing to everyone at little or no cost, but administrators have grown weary of such talk, or embarrassed by it, or think it is old fashioned and not trendy because no one of any standing talks that way anymore.
Exclusion, in turn, carries two kinds of value. One is the common one that university administrators are happy to chirp about: providing a wealthy investor with the opportunity to spend a lot of money in development in the hope of having a market monopoly for what’s left of the 20 year patent term by the time a product is on the market. For that, there are songs about how no one invests without the incentive of monopoly, and how evil free riders wait to copy what the wealthy investor has worked on so long and diligently, at great expense. Yes, all this sometimes does happen.
But there is another side to exclusion value. That value comes about when there are companies that want to practice the invention right now–without the need for a commercial product–or they are ready and able to make such a product, and really don’t care if others can make a similar product. They are ready to invest. Here, patent exclusion derives its value from preventing these companies from acting on what they have read in a research publication or heard at a conference. The value has nothing to do with attracting a grudging, risk-adverse wealthy investor and everything to do with what people who want the invention will have to pay to gain access to it.
For the wealthy investor version, typically the example used (and it is pretty much the only example) is that of turning a compound into a drug that can offered for sale. This costs a huge amount of money, and the only way to do it, so we are told, is for humongous companies to get monopoly positions, spend hundreds of millions of dollars of their own money, and then for some few of these compounds, reap billions of dollars a year in profits for the life of the patent, even though generic versions of the drug could be made available for a small fraction of the price. The problem is, so it goes–and this part makes some sense–the humongous company cannot recover everything that it has billed to the development of the drug if the generics step in, and therefore a company won’t develop drugs at all if it will waste its investment. And if that happens, we will have no more drugs.
It’s a nice story, and a lot of people accept it uncritically as the whole truth of the matter. But why do companies have to have monopolies on basic stuff? Why can’t they share development costs, as in a consortium? Other industries do this all the time. Sematech, say, for semiconductors. And perhaps a lot of that expense charged against the cost of developing a drug is not necessary. Perhaps if more work were shared in common during the screening and characterization of candidate compounds, less would have to be billed to the revenue stream of any particular compound.
Even if one accepts the premise that pharma has to do things the way it does, why should this approach come to dominate all other areas of university inventive life?
Patent exclusivity is the first barrier in this triple play. Next up is the overhead of the licensing apparatus. It is generally quite easy to grant a license to a patent. “I agree not to sue you for practicing my invention, and you agree to whatever we work out in exchange.” Yes, it complicates from there, but for organizations, and especially non-profit organizations, and most especially for state-run universities, it complicates in all sorts of crazy ways, so that what could be, really, just a one sentence exchange becomes 30 pages of dense (multiple meanings alert) legal text that everyone has to sort through, sort out, and fuss over. There are clauses on liability and insurance for liability, on dispute resolution, on governing law, and jurisdiction in which suits can be brought. The list is extensive. Are you going to use a mediation or an arbitration clause? If arbitration, under what rules? Will the procedural law governing the arbitration be different from the substantive law governing the license? Where will the arbitration be held? I like Hawaii, personally. There’s an epic in there, just on arbitration.
A license negotiation with a university can easily cost a company $10K in legal fees, just to work over an agreement, and can easily delay access to an invention by months. The negotiation also costs the university, though often this is buried in the salaries of the administrators who do the work, review the work, write the policies for negotiations, deal with exceptions to policy, deal with audit of policy, prepare the training for all university employees in the policy, and staff the committees that oversee the policy. Simply by claiming ownership and filing a patent application, an institution encumbers a research discovery with its own management practices and concerns–what the organization expects, and demands, in a license relationship is much, much more expensive than what would happen if the organization stayed out of it altogether, there were not any ownership claims made, or if they were made, the matter was with an inventor and his or her agent, not with the organization.
There is a third barrier in all of this, and that is the expectations of the business transaction. One could deal with licensing by granting a simple public license: “US Patent Number X is hereby dedicated to the public.” That would take care of the exclusion issue and the overhead of negotiating dense legal prose. But no, that’s not what happens. Universities take ownership for a reason, and that reason these days is to make money by “commercializing,” and by the time a university has rung up a bill for prosecuting a patent application, it may be $20K in the hole in legal expenses for the US patent alone, a lot more if they go national phase past the PCT filing, and even more if they count the time “processing” the invention report–reading it, getting the form filled out correctly, entering the information into the “database” (and the cost of developing the database, training folks on the database, backing up the database, fixing the database, updating the database, complaining that none of the reports from the database are actually of any use, dealing with errors in the data in the database…), getting assignments for the invention, choosing a patent attorney, getting authorization for a budget for spending on the attorney, setting up an account for expenditures (reality) and for royalties (mostly dreamin’), meeting with the inventors, getting supplementary information, reporting the invention to any sponsor, pulling the sponsor’s research contract to see what the IP terms are, check for background rights or related technology already under management, do a market evaluation (as if that’s possible with fundamental, new stuff). So $20K can easily be $30K or $100K or more.
Now the university has to try to get that money back. There goes any thought of dedicating the invention to the public. It’s simple. License for $20K or more, or don’t license at all. Given the workload on a typical university licensing office, this translates into “license exclusively, or you will have to license the technology again and again and again and will never get to all the other inventions the university has claimed.” The barrier posed by the choice of licensing strategy to make money represents the greatest barrier of all. Not only does the university want money, but because only a few licenses will pay for all the other work, the university wants lots of money. To get lots of money, a university has to find folks with lots of money–big companies, wealthy investors, venture capital companies–and it has to play to their needs. I like big companies, investors, and VCs. They have their place in the grand scheme of things. Big trees and little trees, it’s all good. The problem is how a university’s desire to make money from patent licensing shifts attention to a particular class of folks over all the other possibilities.
There is a logic to it all. Let’s see. We have taken ownership of inventions, and to keep that ownership we have to file patents. So we have spent a ton of money, and we have to make our money back. To do that we have to license patents for money. Given that we don’t do this very often, when we do, the money has to cover all of our other expenditures. So we will have to license to folks with lots of money, or to those who will make a lot of money. What’s the best way to do that? Commercialization! Make the licensees make products. The company we license to has to agree to make a product, sell lots of product, and pay us a royalty on every sale. Let’s put that in the license. Now, who will pay us under this model and make this “model” look successful? Thus, the hope is, license to big companies, and license to startups backed by venture investment. If those folks aren’t biting, though, then, why, it’s the valley of death. If only we had more money, we could develop the idea into product ourselves and then hand it to a company to be sold (and then get paid back even more for the extra expenditures we have had to make).
The valley of death is the refusal to accept that the application of a commercialization patent licensing model to all inventions made in university research is unworkable, dunderheaded, failed marketing. It is even more foolish to think that the lack of money is the problem. It’s the choice of business model. It’s the idea that there’s one business model, or that the university has to claim everything, and that everything claimed has to be owned by the university. That everything owned has to be “protected” by patents. That the patent costs have to be recovered. That money has to be made from “commercialization.” That everything has to be treated the same way (screw one inventor, have to screw the rest, to be fair). That no alternative can be considered until the effort to find an exclusive licensee with enough money to pay has been shown conclusively to be a failure, er, caused, yeah, by the valley of death. So then what? Revisit the dunderheaded architecture of the model? No. That would make sense. That would be, uh, reasonable. Instead the effort is to get the government to provide even more funds for research, and economic development organizations to chip in, too. In this way, abject failure, with the right wording, can appear to be a pitch for more money and at least making it appear the problem is lack of money and not management judgment.
A big part of the problem for universities is that the original, and spirited, debate on the merits and uses of patents by universities, by faculty, and by industry was rooted in a concern for university, and faculty, public service and benefit. Those were once pragmatic matters, where now they are seen as distant mostly irrelevant abstract ideals, to populate puffy preambles and thereafter to be ignored.
The architecture of university patent policies that has come to dominate is built on an agent model: selective, focused, public-spirited management of inventions to advance research and the public interest. All things considered, it is an interesting, and workable way to go at the problem. For each thing that anyone claims, get an agent to work on it. The agent accepts only what it can work, and declines anything else. Bayh-Dole itself is built on that agent architecture. The agent architecture, however, has been overridden by a financial portfolio model: work a range of assets, and some will generate a lot of income and others, less or none, but hold them anyway; abandon only what becomes worthless. And that architecture, questionable as it is from a research and public service perspective–and from an innovation perspective–has been conflated with a sales model: take in everything, and some bits are bound to be worth something, and those will be the bits of “success” that justify the taking of everything in the first place.
Along way, university administrators have done their best to twist and distort and patch the patent policy architecture to make it appear they are doing just what the policy mandates. Of course, the administrators are usually the ones revising the policy, so they end up with policy statements that require (on close reading of dense prose) folks to be dunderheads and work a market model that’s badly matched to university activity and policy architecture and academic strengths. All that’s needed is a lot more money and then the rich investors will see that every university idea is a great idea, that every great idea shall be a product to be sold, and that products shall be sold at premium prices, supported by patents, so that rich investors can get richer, and university administrators can eat at the crumbs that fall from their tables. All that’s needed is that the valley of death gets crossed, rather than it simply goes away with a change in policy and operating model.
Now, there is a “funding gap” in startup activities. It’s that period after a business idea is selected, and the principals work to raise the resources to get going–to build a prototype, to demonstrate their work, to field a product, to develop customers, to ramp up production to meet demand. Perhaps that gap is filled by sales, or by donations (like Kickstarter), or by money the principals put in because they want something to happen. Or they get a loan, or seed funding, or an angel investor. Or they get an SBIR or STTR grant. Or they get a contract from a company that will provide air cover. Or they pitch to VCs. That’s all real, and happens all the time. However, that funding gap is not the valley of death that university administrators talk about, but administrations would like to conflate a new venture need for funding with the consequences of a bureaucratic approach to patent licensing, to make it seem that they are talking about the same sort of fund raising, as if each invention they have claimed from the research community has been selected because it is a good idea, with real people behind it, rather than simply because it can be owned (or, claimed to be owned) by the institution.
There is also another funding “gap” and that is the “chasm” that Geoffrey Moore has identified. That’s where something new, appealing to visionaries and technologists, must be repositioned so that pragmatic adopters motivated by utility rather than newness will be willing to acquire. In this gap, early sales lapse until a new audience comes to accept a new product. That change, to cross the chasm, requires a change in marketing strategies. One of those strategies is to recognize that pragmatic buyers, especially of technology, do not like to deal with a sole source, which can lock them in to a single supplier, who then can control pricing, features, and availability. Pragmatic buyers like choice, and when they have choice, they choose more quickly, and when they choose, they choose the strongest option, and they like to confirm their choice by, collectively, making the strongest option even stronger. That’s the idea, anyway.
Now look at that in a university research environment. A good idea in a university setting will appeal to visionaries and technologists. That’s where a good research idea has life, and where it has purpose, which may be so much as to inspire and motivate, to illustrate or warn away. This is the native habitat. Some few of these ideas might have a life beyond this context and role. For those that do, the reason is not that they can be owned, but that those that know these ideas are motivated to do something with them–inventors, investigators, entrepreneurs, investors, company engineers–whomever.
To take these ideas, claim ownership of them, cut them off from circulation, and even from decisions that otherwise would be made by the research team, and create substantial barriers to their circulation–and then claim that it is the ownership of patents, to exclude others, to make a ton of money–that this is the magic sauce that will motivate people to take action with regard to these ideas–that’s just plain nuts. Just as it is also plain nuts to think that these ideas, if they were not patented by the university, would just “sit on the shelf.” That, too, is nonsense. And it makes no sense that universities would adopt a model of invention management that, after thirty years of being progressively debased and become sloppy, silly, and mostly unchallenged, would argue straight-faced that the highest calling of a university is to sell off its research discoveries to speculators, as they are the only ones who match the profile of folks taken with the idea of dealing in research patents to make money from them.
Ideas of the research sort do not go into some mythic shelf when they are not owned by the university–they are published and taught, they are discussed, they circulate, they motivate other ideas, they become part of the effort to figure out what is really important, what works, and what works better. They are the commons, the platforms, the libraries, the tools, the fabric of research. No patents needed.
The valley of death doesn’t exist for university licensing offices, as it is depicted, as something that happens after an invention is taken over by a university because there isn’t enough money to make it sufficiently attractive to speculators. The valley of death happens because an invention is taken over by a university. If you catch university licensing pundit using valley of death, you can bet they don’t know what they are talking about, but they will be very confident that the valley of death is well established, and exists sure as hell–and that they are the minor deity with rod and staff offering protection against shadows and evil. They will mock anyone who says this valley doesn’t exist or exists primarily because they are there. And, well, one will have to concede that there is a valley of death in university management of research inventions. There’s a reason why many inventions and ideas do sit on a mythic shelf: the university licensing office has become, where no amount of money can help it, that valley of death.
University inventors mostly pray to find some way to get through that valley, where the shadows of death are the bureaucrats.