Dear Vice Provost for Research,
I said I’d write again and explain how your technology transfer office is still hugely important. If you remember, in my first letter I discussed how the prevailing approach to university research innovation has failed, and the problem is not lack of funds or a need to change the culture, or just the wrong organizational structure or office name–but rather a fundamental misconception of how innovation happens and how the institutions that host and help research benefit when innovation does happen. In my second letter, I outlined four steps to make the change to a freedom to innovation environment:
- renounce university ownership and central control of research inventions
- restrict the scope of policy claims, first by interpretation and then by revisions to policy
- develop working relationships with invention management agents
- rebuild the infrastructure to support decisions to pursue commercial development.
If your technology transfer office is not having to manage hundreds of inventions per year (“and every day the paper boy brings more”), it can focus on doing a few things really, really well–and at much lower cost.
If your university receives federal funds for research–almost all American universities do these days–then your technology transfer office needs to provide education to research personnel regarding the importance of timely reporting of patentable inventions made within the planned and committed activities of federally supported projects. The technology transfer office may also be the personnel designated for patent matters at your institution, so that’s where invention reports will go, too.
You can even get the technology transfer office to implement the (f)(2) agreement at the heart of the standard patent rights clause authorized by the Bayh-Dole Act. The (f)(2) agreement is essential–it is a requirement of federal funding agreements–and it is the federal gesture to respect universities that choose to continue the path of freedom to innovate rather than adopting an ill-suited corporate model of central control. The (f)(2) agreement is the key provision to preserve academic freedom in the context of patentable inventions made with federal support. Use it or lose it!
Historically, technology transfer offices were created with the encouragement of the Research Corporation to provide campus-level help with understanding what makes inventions patentable, documenting those inventions that are or may be patentable, and transferring those inventions to a qualified invention management agent for review and possible management. That is, the technology transfer was not from the lab to industry, but from the lab to agent. If the agent chose to accept an invention for management, it would take assignment of the patent rights, file one or more patent applications at its own expense, and work with the inventor(s) to develop a strategy for deployment. If inventors felt strongly about a particular matter, they could negotiate that at the time they considered the agent’s offer to manage the invention. The agent, after all, works for the inventors.
This situation is unlike what happens now at most universities, which have made it clear that while they may share royalties with the inventors, and may ask inventors for help, they work for the university, not for inventors as private clients. If inventors try to negotiate an arrangement, it’s construed as a conflict of interest, and in the case of state universities, as an attempt to influence officers of the state in their official duties. Heart of badness. But in the agent model, this is the most natural and responsible thing for inventors to do–get their concerns addressed upfront in establishing the relationship.
The technology transfer office also can maintain a list of qualified invention management agents–law firms that provide quality patent work in areas of active university research, specialist invention brokers that work with a focused portfolio, and research foundations that support particular lines of research or work closely with specific industries or regions. The technology transfer office can develop royalty-sharing agreements with the invention management agents–in exchange for the referral, the university gets a share of any income. This is great money–the university is not on the hook for patenting costs, does not have to docket deadlines, does not have contract with patent attorneys, does not even have to worry whether the technology is any good–all it has to do is identify inventions, help the inventors do a good job with documentation, and get it to an agreeable agent for management.
Under the standard patent rights clause, any net income after invention administration expenses and royalties paid to inventors must go to “scientific research or education”–so an invention management firm receiving assignment of a Subject Invention also necessarily gets in the package of the “entire right, title, and interest” the requirements of the standard patent rights clause authorized by Bayh-Dole. So the agent will have to use any net after its expenses for scientific research or education–and the technology transfer office may as well work out how to be the recipient of a good share of the net.
A technology transfer office may also support invention development using its own funds. An invention policy at a university has an important role–to authorize university officials to provide resources to assist university inventors with their inventions. The policy can specify what resources, and with reasons, and for what purposes, and for what in return. For instance, a technology transfer office might provide financial support to permit an inventor to conduct more tests in the lab, and in exchange receive a financial interest in any commercial income the inventor might receive from licensing the patent. It’s a question of whether the inventor wants the help, and whether the university is willing to provide that help. You will see that there is the matter of equitable treatment and a judgment about how what tests are properly done in a university lab–but these are good things to work out. University presses do much the same thing–choosing some manuscripts to edit and publish, and others to pass over.
A technology transfer office can also play an important role in managing inventions and other forms of intellectual property directly. Faculty and other university inventors may want the university to be involved. Invention management agents typically aim to commercialize inventions by licensing patent rights to companies that will make and sell products. But that’s not the only valuable role for patents in a research setting. There’s another role, even more important to the university, and that is the effort to create technology commons–platforms, libraries, ad hoc standards, and testbeds in which new technology can be tried out, studied, used for research, improved, extended, and used publicly–even by companies engaged in profit-making activities. Research discoveries do not have to become monopoly-controlled products. They can become industry standards, shared among all companies. Perhaps the most important technology developed at UC Berkeley is the circuit design software package Spice–available at no charge, but for years a key industry tool, putting Berkeley at the forefront of development. Many good things come back to those that lead–better research questions, contributions of technology and funding, talent wanting to be involved, notes of appreciation, big government grants, industry consortia. Nice stuff.
Wherever the developers of an invention want to use patent rights to help build a commons or platform of technology or tools, it’s perfectly appropriate for a university to agree to manage the patenting and take ownership of patent rights on behalf of the project. In this capacity, the university acts as a trustee for the patent rights. Historically, this was one of the roles most people endorsed as one appropriate for a university. Even industry gets it–if the university holds the patent rights to assure everyone gets access, this is a good thing, limiting the fragmentation of an area of research into little bits of IP held with white knuckles by speculators and trolls. There’s money in standards, too. It is lost on some university patent administrators that many early university biotech “big hits” such as the Cohen-Boyer gene-splicing patents and the Hall patents on expression of polypeptides in yeast (leading to a Hepatitis B vaccine) were non-exclusively licensed to industry.
The emphasis throughout the Bayh-Dole Act is on “practical application” of inventions made with federal support. There’s no requirement that to be used, an invention must first become a monopoly, and then a product, and then get used. An invention–even patented–can also be “practically applied” directly, without first becoming reduced to a product format. Consider internet communication standards–these never become products themselves, but they enable many valuable products to be created. It would be a huge loss if these standards were to fragment into rent-seeking speculation.
If one university offers technology non-exclusively, consider the value to industry and the public if multiple universities provide technology on common terms–creating non-provincial research pools. Basic science is not helpful if patent rights in new discoveries are held back in the hope that someone will be willing to pay for exclusive rights. Of course, there may well be someone willing to pay for exclusive rights, but often that payment reflects the value of denying access to others. Denying access runs against the values society expects of university research, creates cognitive dissonance, and leads the public to withdraw its support for universities. Don’t wonder why universities have lost public support–a key contributor is the perception that universities are floating in research money and patent profits. You know that’s not true–most universities lose money on their extramural research programs–as much as 20%. And most universities lose money on their patent licensing programs, too. You can do something about the latter–your technology transfer program can break even or do even better.
When you return your university to an open program of invention management and use external agents to manage inventions that faculty (and other) inventors desire to commercially exploit, you transform the policy situation and the practice objectives for your technology transfer office. To see what is going on, perhaps an illustration will help:
Here is an illustration of the orbits of the planets seen from two different perspectives–one if the sun is at the center (it isn’t really, it’s at a focus of an ellipse), and the other if the earth is at the center. As you can see, the orbital paths taken by the various celestial objects are very different. Yet both diagrams capture the same orbital behaviors.
(See the page with full animation here (h/t to Kottke.org)).
Think of these diagrams as two ways of dealing with intellectual property policy. In the orbits on the right, with their wild and loopy retrograde motion, we have an emblem for the aggregate-patent-license model. It puts patenting at the center, and makes university research, public mission, and collaboration orbit around it. The result is poor alignment of the technology transfer function with university mission, complicated policy, and complicated, dissonant licensing practices.
In the orbits on the left, we see instead the result of aligning technology transfer practice with mission, creating simple policy statements and greatly reducing the dissonance of licensing practice. Why should any university attempt to augment the value of research findings by threatening to prevent anyone from practicing those findings? That’s the fundamental value proposition for the policy represented by the loopy orbits on the right side of the diagram. If one were to write out in plain language the loopy technology transfer policy, it would go something like this:
Innovation requires a university administrator’s thumb in every possible transaction involving scholarly research findings. The university rejects academic freedom for all scholarly findings from which administrators decide to attempt to make a profit.
The university administration will own outright all such findings and will decide who gets to use them and who is excluded, based on what someone offers to pay to obtain monopoly rights.
The university adds value to these scholarly findings by using the patent system to threaten with infringement anyone who does anything that might fall under any claim of a university patent.
If no one is willing to pay for monopoly rights, it is better that everyone be excluded from practicing these scholarly findings for two decades to teach industry and investors a lesson about refusing to cooperate with the university’s scheme for making money.
To induce faculty to be complicit in this effort to shakedown industry, the university shares a portion of any licensing income for a given patent with the inventors named on that patent and with the inventors’ schools and departments.
From there, it’s a matter of writing sketchy policy statements about how the public mission of the university is more important than academic freedom, how that public mission is mostly about making money from patents on faculty inventions, and how the university will obtain ownership of most anything it wants by expanding the definitions of invention, employee, and duties so that invention includes non-inventions, and employee includes non-employees, and duties includes non-duties. It can–and has–been done. To renounce all this means going back to the left side orbits, and writing policy that’s clear, that tracks university values, and reduces both the dissonance and complexity in using intellectual property to advance university interests.
In short, intellectual property manages relationships between organizations. The fundamental premise of intellectual property is that of exclusion. This is the competitive assumption at the root of patents, copyrights, and trademarks. For a university, however, exclusion for patents and copyrights, at least, runs against the fundamental public premise of university research–that of open, reliable, honest public availability of results. And not just the announcement of results, but also the freedom to use those results. You can see how a university policy taking ownership of all inventions undermines publication. What once was both an announcement of a finding and the freedom to use that finding becomes, under the aggregate-patent-license model, simply advertising for a monopoly patent right on offer to the highest bidder (which is often the only bidder, and often there actually is no bidding).
An honest university adopting the aggregate-patent-licensing model (I know, that’s an oxymoron) would require as a matter of policy that all publications from the university concerned with an invention to which the university has claimed ownership must carry the following disclaimer: “The University of X has asserted patent rights in inventive material presented in this article. No use of any such inventive material may be made, even for research purposes, without the express written permission of authorized representatives of University of X. Any unauthorized use of such inventive materials may constitute infringement under the patent or copyright laws of the United States and will be prosecuted to the full extent of the law.” Is it any wonder that astute companies advise their research personnel to ignore university publications?
In stepping away from this approach and returning to the policy framework on the left side of the diagram, of simple orbits that make sense of the situation, you do some remarkable things for management within your technology transfer office.
First, and this is important beyond the obvious, your technology transfer people will be working only with the inventors who want to work with them. This self-selection principle is the most important indicator that your technology transfer office will accomplish what they and their inventors seek. Commitment rather than compulsion lays the groundwork for productive working relationships.
Second, by making management of inventions discretionary, you reduce the management overhead for inventions. While a technology office might receive invention disclosures for a number of reasons (such as reporting to a sponsor), it only considers for management those that (a) they are requested to manage and (b) are in their sweet spot of expertise for managing. Everything else is declined. This is the second great indicator of success–not only has the inventor chosen them, but they too have chosen the inventor. Reciprocity is built into the policy to create a strong working relationship from the start. You cannot create such a relationship in a compulsory aggregation model any more than you could turn your plantation slaves into your best friends by trying to be nicer to them or educating them about how nice you already are. Sorry for the difficult analogy, but these are difficult times, and you need to understand how deep and dark runs the problem you face–and why you need to act decisively.
So what inventions will be in your technology transfer office’s sweet spot? Ones that
- are tools that others can use in research or professional practices
- build a platform or library of tools for others to use
- offer protection against speculators who might block research and access
- develop a consortium or commons for shared development and use
- are directly in line with your office’s expertise and industry relationships
Refer everything else to external agents. You don’t want it cluttering up your work areas, costing you time and money, creating relationships based on distrust, bitterness, and conflicting values. In business, one of the variations on power law phenomena is that 80% of your profits come from 20% of your customers. A smart company knows where their profits come from and is ready to refer the other 80% to others. A smart company does not hold on to the mis-matches, the problems, the schemers “just in case”–get them over to the competition, where either they are not mis-matches, or if they are, they take up someone else’s time and resources, not yours. If you want to be successful, then start by being selective in how you create successful relationships. That is a more important part of policy than worrying about expediting how you gain ownership of other people’s work.
When you move to a freedom to innovate approach, you give your technology transfer office the gift of strong working relationships. But you do more than that: you also offload the management burden of dealing with balky inventors, reduce the whisper campaigns about purportedly cruddy service, and–here’s the financial kicker–greatly reduce what you spend on patenting, databasing, marketing, and reporting. A technology transfer office can be more successful managing 40 inventions a year as it is managing 240 inventions a year. The metric is not how many inventions are reported, but how selective the office is in choosing what to manage, and the ratio of inventions selected to those that are used by others. In the aggregate-patent-license model, the percentage of inventions used to those claimed is on the order of 0.5% to 5%, and closer to the 0.5%. Prior to the passage of Bayh-Dole, when universities used the double selection method, with external invention management agents and a preference for an equitable financial interest rather than an ownership claim, that ratio was more like 20% to 30%–6x the productivity, at lower costs, with lower exposure to risk, with less disruption of collaborations, with much less dissonance with the university’s public mission.
More importantly for technology transfer practice, when a university adopts as its premise that it will use its intellectual property positions to encourage inclusion rather than exclusion, it changes the value proposition related to the scholarship of its faculty, students, and staff. Rather than leading with a patent right, technology transfer now leads with other things of value–the availability to teach, access to experimental setups, raw data sets and data analysis tools, talented people willing to help, ideas for improvements or new approaches, prototypes, technical documentation, user documentation, what doesn’t work. Many of these things can’t be owned–they are non-IP intangible assets (NIPIA). But they still carry value, and very much can form solid, long-lasting working relationships, and even paying relationships if money figures in your way of thinking.
There are also network effects to working non-exclusively, starting with NIPIA and working patent rights into the relationship as an added benefit. In projects I have managed, the first few organizations that join want to assess the new findings and see how they work. They are tentative (even if stuff is free), because they don’t want to waste their time if something is not so good as was claimed. But once five or six companies have joined, the next companies join in part because they want to see what the other companies are seeing. Now the value proposition changes. Each company may pay a share but gets access to the results paid for by the companies in aggregate. With even ten companies in the mix, each one says–wow, I am getting a 10x return on my payment to this university project! Even if the university lab is only half as good as my own team, this is still a great value!
We had one project that over time came to have 200 company research and development sites paying $4K per year. For $4K, each site got the research output (in the form of software and assistance) from about $600K of spending (net after administration and some royalties to the development team). That’s better than 100x return on investment. That’s a great value proposition created by IP management, and not based on or enhanced by the threat to sue for infringement.
The next round of organizations that join the project don’t need to do their own assessment of whether the technology works, and aren’t interested so much in seeing what others are seeing–they just want to use what they perceive to be a reliable platform that gets done what needs to be gotten done. And it is these companies, the later arrivals, the ones that show up with the uncertainty has been reduced–for the technology, for the use, for the ability to work with the university–that make the research technology an ad hoc industry standard. All this happens without having to interpose a “commercialization” step or exclude all others in favor of a monopoly licensee. Commercial products and startups with specialty focus come *after* a project has persisted and grown for a while. The commercialization is a natural consequence of effective university IP management, not a pre-condition.
And where we come to the most important realization that happens when you restore your university to this left-side, ellipse-based, freedom to innovate approach to technology transfer: People pay you because they want you to succeed. They don’t pay you to persuade you to withdraw your threat to sue them for using your research results. When you go into your favorite local restaurant, you have already decided you will pay for the meal, and you want to the payment to ensure that the restaurant will be there for you again the next time you eat out. Sure, you might wish for a free meal every time you come in, but you know that the restaurant needs to charge to provide the menu and service and ambiance you so much enjoy. You pay because you want to pay. This is the core insight for your entire technology transfer program–the university receives payments from companies that actually want to send money to the university. Company officials get it–they know that a business profit is the difference between income and expenses. Expenses come with the territory, but there are some expenses that are more fun (a meal out!) than others (paying lawyers to back you out of a mess).
The current, prevalent approach to university technology transfer claims the freedom to innovate approach is silly, ineffective. According to the conventional wisdom, companies are nasty exploiters of university good faith and will steal anything they can get away with. What’s needed is total control–over inventions, over patents, over what gets licensed, over the terms and conditions of each license, over auditing, over litigation. Dominate those bastard corporations and take from their management and shareholders the piece of action you deserve (and can extract) for your patent positions, or sue their socks off. And you know, once you have adopted this approach, it becomes in its own sad way true. Companies do try to prevent you from succeeding; they do work around you; they don’t want to pay; they do wish for your destruction; you will have to play down to their level–but keep in mind that you chose that level first. That’s where the aggregate-patent-license model starts and where it ends up.
You can run industry’s favorite restaurant, or you can have your hired thugs out creating road blocks on high mountain passes. You can dress your policy up to look nice for the clueless, but people are smarter than bureaucratic wordsmithers give them credit for. Make the change–put your university back on a path toward building collaborations with industry, with the public, with non-profit research foundations. Find your value in your NIPIA services, and use patents and copyrights sparingly to build commons, platforms, libraries, open software and architectures–these are the things that will establish your reputation as a great place to work and a great place to find new technology and talent. For everything else, support it as you can for an equitable share of any commercial upside, but refer it all to specialist agents willing to do the patent work and present options to industry and investors. Stay out of that side and save a ton of money, worry, time, and distraction.
There. That’s what happens when you return your program of technology transfer to a path that’s aligned with your university mission. You add focus and reduce costs, you build great projects, you transfer research findings for use, you play the role of trustee and advocate, you create working relationships with many companies, and you make money because people want you to have the money. So change your interpretation of policy–you can do that immediately–wherever there’s ambiguity (and most university IP policies are a stinking mess of ambiguity), interpret your policy in favor of freedom rather than control; if that’s not enough, then open up a dialog with the faculty (and with your deans) to shift the policy statements from control of everything to control of only those things that the university must control by law or contract, and make everything else voluntary. Open up a dialog with invention management agents–law firms, venture firms, invention management firms, non-profit foundations with targeted project goals. See how they are ready to play if you adopt a freedom to innovate strategy.
You might start by opening up a few departments at a time. Engineering, business, and the humanities first, then the social sciences and the sciences, then bioinformatics, and last medical devices, anything in cages or refrigerators, and pharmaceuticals. There are plenty of other changes you can make in your technology transfer program to bring it along. Freedom does have its consequences. There will be problems. But your role is to govern, not manage. There’s a big difference. Governance involves mediating among competing interests. Management aims to own those interests and control them.
Both approaches achieve a form of order, but governance respects creative culture and recognizes that the initiative and the intelligence is at the periphery of the organization, while management assumes that creative culture works best when it is told what to do and it has no option but to obey. One of these approaches has a long history of working for innovation; the other doesn’t. If you want to connect your research program with public recognition, then you will return to the freedom to innovate path. The snake-oil model has got a thirty-year run and demonstrated its incapacity to deliver anything but glossy appearances of potential success. Get over the marketing hype. Get real. Look at what things have become–bloated, ineffective, monstrous in their way–even when run by sincere, diligent, talented folks.
It may take a conversion experience. You may have already endorsed policy and a vision that presumes university central control. It’s easy to slip over to that way of thinking and feel you have to be consistent with it. But you don’t have to do that.
“We are going to take a different direction to these matters from the line taken by other universities. We are going to align our technology transfer policies and practices with our public mission, not with rentier-speculative investment culture.
The entrepreneurship we desire in our faculty, staff, and students comes from their initiative, and their sense of responsibility to their colleagues and to the public. They chose the university with these values in mind, and we expect them to stay true to those values even as they find or create things that may have commercial value. Those things also have social value, and it is from social value that innovation arises.
With a few or even no changes to formal policy, we can reduce the scope of ownership claims our policy makes on our creative culture. It will take some effort, and a few years to adjust, and we will make some mistakes, no doubt. The price of freedom is ethical responsibility. That’s a price we are ready to pay.
It feels so good to get started now. We will become a destination of choice for companies doing difficult, valuable research and development. The public will see us once again as a beacon, an advocate. We can be that place, and if other universities adopt a similar approach, together we can restore American university research to the powerhouse it was forty years ago. It all starts someplace. Why not here? Why not now?”
I hope that gets you going on a productive path. Let me know how it goes.
With hope for the future,
Gerald