A dual monopoly approach to innovation management involves both a comprehensive institutional demand for ownership of inventive work and an institutional determination to convey monopolies in that work for private exploitation. The first monopoly is an institutional one. The second monopoly is a market one, created by transferring a monopoly right to a speculator or company for development as a commercial product. The combination of these two monopolies creates the impression of an institutional system to create commercial products from university research. In practice, the dual monopoly approach is among the worst possible choices for university technology transfer.
In the dual monopoly approach, a university creates monopolies (by acquiring inventions from its personnel and seeking patents) and then transacts these monopolies (typically, by exclusive licenses that are assignments). Thus, the approach to innovation is to move monopolies into the hands of investors in those monopolies, and in exchange taking a share of the investors’ returns from their exploitation of the monopoly right (typically, one or more patents). Thus, the dual monopoly approach diverts inventive work from broad circulation preferentially into the hands of those speculative investors who insist on monopoly control to justify their investments.
The Institutional Monopoly
The institutional monopoly demands that all institutional personnel assign ownership of their inventive and creative work to the institution for management. Typically the requirements for assignment are drawn more broadly than similar requirements in for-profit settings. A university, for instance, may require faculty to assign any inventions made in their area of expertise rather than just those inventions made within their scope of employment. Many university policies define “invention” to include a long list of assets, many of which are not a matter of patenting, and some, such as “know how” or “tangible research materials” or “technical information,” are not even statutory intellectual property.
The basic idea behind an institutional monopoly is to take all intangible assets that might be disposed of for money (sold, licensed, leased, rented, traded for research funding), evaluate them, and hold those that appear valuable. In theory, those that are not valuable may be “waived” or “released” or otherwise returned to their “inventors.” In practice, however, few things are returned. The logic goes like this: if the thing isn’t valuable, then it is not worth the effort to return it to its inventor. If the inventor wants a thing, then it must have some value and the inventor should pay that value if no one else will. It’s rather a catch-22 for an inventor. If you want your “invention” back, then it must be valuable so you can’t have it unless you pay. If you don’t want your “invention” back, then it is not worth the effort to return it to you.
The Institutional Monopoly Isn’t Necessary
In all of this, of course, we may miss the obvious: even if some investors require a patent monopoly in order to participate in developing a commercial product, why should an institution insist that investors must deal exclusively with administrators controlling patent rights rather than with inventors or principal investigators or with independent agents representing their interests? That is, what makes the institutional monopoly necessary to the transaction?
Various arguments have been made. A university must have a single, uniform policy. It is unfair that some inventions are handled by the university and others are not. The university has more resources to patent inventions and so should take all inventions whether inventors want the institution to do so or not. The university has licensing specialists trained to manage inventions, and so the university should require all personnel to use these specialists. The university must comply with all sorts of laws and contracts and therefore must own all inventions to limit its liability. The university always has the public interest at heart and inventors and principal investigators may not, and therefore the university should take ownership of all inventive work. The university makes its resources available only for public benefit and therefore it is inappropriate for any inventor to own or profit from his or her invention personally. The university has invested a great deal in supporting its research staff and ownership of inventions is consideration paid by these people for the university’s investment in their skills and careers. The university is better positioned to manage inventions than individual inventors and can avoid the pitfalls of unscrupulous invention management organizations, companies, and investors; therefore, the university should own all inventive work. Employees owe the university their inventions as a matter of employment. Employees have a fiduciary duty to the university to turn over to it all things of value that they create. Federal law mandates that universities take an active role in commercializing inventions, and therefore the university must own all inventions made by its personnel or using its resources or under its auspices. Institutional control of all inventions is the most effective way to realize the public benefit from university research.
Faced with such an inventory of arguments, the institutional monopoly might look justified, but each of these arguments is fallacious or unsupportable and can be rebutted. It’s just that it takes more effort to rebut fallacious arguments than it takes to assert them. For instance, a single policy is arbitrary when it fails to allow adaptation to unexpected circumstances. And invention is often unexpected. Universities traditionally have had at least two IP policies, one for patents and one for copyrights. In a number of universities, there was (for some years) a separate patent policy for medical inventions (forbidding patents, at the request of the medical faculty). These multiple policies worked perfectly well.
Further, even if one has a single policy, it does not have to be one of institutional ownership of all inventive work. Stanford, for instance, for a long time had a uniform patent policy that stipulated that inventors owned their inventions unless Stanford was legally obligated to take ownership. That was a single policy, but it did not implement an institutional monopoly on inventions.
We could work through the remaining arguments in this way. The number of arguments does not determine the validity of the claim–one has to consider the validity of the arguments before accepting a conclusion (unless, of course, one already believes the conclusion–in which case arguments are there to defend the belief, not to show how the belief arises).
In practice, the institutional monopoly has arisen because
- The Bayh-Dole Act was made to appear to mandate it
- University attorneys claimed everything they could to avoid liability
- Technology licensing officers wanted control of all inventions
- People feared inventors would misuse university resources
Thus, the institutional monopoly appears to come about as the result of urges and fears, not reasoning based on evidence of good practice or outcomes, or on principles of freedom in research, collaboration, and publication.
The Market Monopoly
On the market monopoly side, the dual monopoly approach aims to preserve the monopoly acquired from inventors (and others) in transactions that produce income for the institution. Income may be disclaimed as an institutional goal overall, but in each transaction, there is a demand for income–consideration for the license, payment of patenting costs, upfront fees, milestone fees, warrants or options for stock, indemnification, and the like. When income is disclaimed, it means that university administrators repudiate the idea that it acts as an agent for inventors–that is, the university has no obligation to make money from inventions for inventors, only that if it does make money, it will share that money with inventors, even if they don’t want it. Disclaiming income also signifies that university administrators expect many failed transactions. It’s like buying groceries and expecting maybe all of them to be inedible, and calling it a “high risk, high return” activity.
The market monopoly is preserved because an institutional monopoly–often a patent on an invention–is assigned to a for-profit venture, typically an existing company (usually large) or a startup created to take the license (and thus, in the interest of the investors in the company). The assignment typically takes the form of an instrument labeled “exclusive license.” In copyright law, an exclusive license is expressly identified as a means of conveying ownership in a copyright. For patents, courts hold that the transfer of all substantial rights in an invention–to make, use, and sell–constitutes an assignment of the invention, even if the licensor reserves a right to make and use an invention for its own non-commercial purposes. One of the hallmarks of an assignment of an invention is that the assignee has the right to enforce the patent on the invention. Universities routinely announce in their exclusive licenses that they permit the licensee to enforce licensed patents. These are assignments.
The irony in the market monopoly is that most universities declare that they will not assign inventions that they acquire to others (except to the federal government, as part of the apparatus of the standard patent rights clause authorized by Bayh-Dole). Bayh-Dole, for its part, requires federal agencies to prohibit the assignment by nonoprofits of subject inventions except to organizations that have as a primary function the management of inventions, unless they get federal agency approval, and for all such assignments, the nonprofit must also transfer the nonprofit’s own standard patent license clause obligations. But universities ignore this prohibition. Their argument is that they do not assign, formally, any patent, except to the U.S. government (if the university decides not to pay a maintenance fee, or has been found to have failed to report to the federal government the underlying invention as a subject invention). But to not assign title to a patent when one has assigned the invention (and thus the right to the patent) by means of an exclusive license is just playing word games. The transaction is what it does, not how it is called, and it’s an assignment.
The Claimed Need for Monopoly To Justify Investment
The overt claim is that investors require monopoly positions to justify their investments. In general, this claim is not true. While investors (speculators, company officials allocating funding) may expect a competitive advantage, they do not generally require that the advantage have the form of a patent monopoly. Some investors do indeed expect such a monopoly, especially with regard to startups and in some industries that rely on patents (such as the pharmaceutical industry and biotech). But many others focus on other elements of competitive advantage, such as features, price, availability, quality, service, and channel.
A second element of the overt claiming here is that a monopoly position is necessary to create commercial product. Again, this claim falls apart in practice. That competitors are willing to copy a product (and even risk infringement to do so) indicates that a monopoly is not necessary to create a commercial product, as a general case. So the overt claim here must be qualified–that the monopoly position is necessary for a company to exclude such competition and control the market for the product for the life of the patent, and thus derive maximum value from the monopoly position through pricing and market share. That is, the claim isn’t about necessity to create commercial product, but rather on the desirability to create that product with an expectation of having no competition for it.
The qualification of this argument goes a step further. Some invention-based products may be difficult or expensive to develop but relatively easy and cheap to copy. For these products, a patent permits the investor to recover development costs without the threat of competition to push down prices and take market share. Stated this way, the argument appears reasonable. But one might consider, by way of perspective, that a product that is expensive for one company to develop (say, a startup with no prior experience or existing customers) might be relatively inexpensive for another company to develop (say, an established company with a commercial position and capability responsive to the invention). Or, an invention-based product may be difficult or expensive because the product is directed at a necessarily expensive market. For example, a compound might be used for human health (an expensive market to enter) or for veterinary use (a less expensive market to enter). The choice of licensee and the anticipated product both influence whether a patent monopoly is necessary to justify the creation of a commercial product.
According to the Harbridge House report, the U.S. Department of Agriculture led the development of a number of inventions–such as new fertilizers–to the point of commercial use during the 1950s and 1960s–none of which required any company to have a patent monopoly.
But we are not done with the qualifications on the general argument that a patent monopoly is necessary for commercial product development. Many biotech research inventions such as the original gene-splicing invention, have been made available non-exclusively and immediately put to commercial use, without a commercial product version required at all. A similar thing happens for software–once the code is written, even if it embodies a patented algorithm or process, it is available for use in that form without requiring the attributes of a “commercial product” version. All sorts of additional features might be added–a different user interface, for instance, or better error checking, or more I/O formats–but these are not necessary for use. Some users might prefer these features, and be willing to pay for them, but that preference does not translate into necessity unless all potential users refuse to use the software until someone else adds the required features in the form of a commercial product.
Transferring Inventions or Patents for Value
There, then, is a competition for one’s imagination about how innovation happens between the overt claim that patent monopolies are necessary–which may be true in the particular but is not true in the general case–and a perspective that offering patent monopolies selects out certain recipients for the exploitation of inventions over others. Typically those selected are speculators in the future value of a patent rather than users seeking to put an invention to work.
Illustration: a university may license a patent to a startup company. The startup then raises a round of investment funding. The startup uses the investment to hire engineers and build a product. The product may have nothing to do with the licensed patent. The company may not ever complete the product. Instead, the startup is acquired for its engineering talent and technology. The university makes money on the deal by holding equity in the startup. No royalties are ever paid on sales of a product covered by the licensed patent. The monopoly in the patent was used to raise investment funding, but that funding then is diverted to a product that has a better market profile than ones indicated by the licensed patent. The patent is used to raise money to create a company with value and does not create a commercial product based on the patented invention. This scenario plays out all the time.
The value of the licensed patent, in such cases, turns out to be the potential to “monetize” the patent right if the startup fails–fails to develop any commercial product, fails to be acquired. In such a case, the startup can then become a platform for suing industry for using the invention subject to the monopoly license. This situation, too, happens relatively frequently. The patent is used to promote the use of the patent itself, as an intangible financial asset, not to justify the development of the underlying invention into a product that carries the financial value.
In these situations, one can see how the AUTM licensing statistics paint a misleading picture. We find inventions, patents, licenses, startups, investment money, and licensing money. We may even find products. It’s just that none of this activity necessarily pertains to the monopoly-licensed invention that is involved. That invention may not have been developed at all. Other economic activity clearly takes place–patent attorneys get paid, investors make money (sometimes), companies get started, and a university makes money (from investors, from stock holdings), but the invention that was the premise for the activity remains merely a patent asset licensed as a monopoly.
The Portfolio Model in Response to Activity Volume
The dual monopoly approach greatly limits the diversity of university outputs–everything must route through the university’s licensing office or risk being called out for violations of policy, misuse of university resources, charges of corruption, and the like. Thus, non-compliant outputs may go covert through informal exchanges or personal consulting. The dual monopoly approach also creates a huge increase in the volume of activity. Invention reports go up–because more things are claimed and must be reported at risk of penalty. With the increased volume comes an increased cost to process all such reports–each must be entered into a filing (data) system, each must be reviewed for commercial potential and intellectual property positions, and each must be secured through assignments. This volume typically exceeds the professional carrying capacity of the licensing office staff. A reasonable invention load for a professional invention agent may be five to ten inventions per year. Universities operating a dual monopoly model may require a licensing professional to handle 100 to 500 inventions each (as was the case at the University of Washington, when I was there). Few people can be effective under those conditions. Good technology gets lost in the noise of ungood technology; opportunities that would develop with diligence are lost; and exclusive licenses appeal because once one has been negotiated (so the superficial thinking goes), that technology is no longer a focus of licensing work.
As a result, universities adopting the dual monopoly approach also adopt a “portfolio” model–many inventions are acquired, but only a few are expected to be licensed, and of these, only a few will become commercial products. At Stanford, about 20% of inventions are licensed. Only about 1% of inventions make more than a million dollars cumulative–that is, at least $50,000 a year for 20 years. That’s hardly the income expected in royalties on a commercial product (of course, a university could license an invention royalty-free, or for a nominal one-time fee–but when it comes to patents, this is almost never the case, and certainly it is rare in the case of the dual monopoly approach–only in research consortia, which require non-exclusive licenses, or in the case of an industrial research sponsor that negotiates a royalty-free non-exclusive license to any inventions made in the research it sponsors).
What’s not evident in the portfolio model is the move from promoting the use of each invention claimed (the public claim) to the expectation of financial returns from the portfolio of assets (winners justify the model). What is also not evident that the portfolio model needs only one “big hit” deal every twenty years to be viewed as financially successful. Thus, while there may be a level of activity–inventions, patents, licenses, startups, investments, and income–for most university licensing offices adopting a dual monopoly approach, only a few instances account for the majority of the significant activity, the activity that accomplishes the public claim for the activity, that inventions are better used for the public good via institutional ownership than otherwise. The reality under the reported figures is not disclosed. Anyone who does not know to ask will come to a misleading conclusion.
Justifying Misleading Metrics
Licensing administrators believe such misleading conclusions are justified because they believe that their dual monopoly approach is the best approach possible. At best, the dual monopoly approach provides a livelihood for those who depend on it–extra staff to manage the volume of work produced by dual monopoly, especially on invention intake processing–and staff to manage the complexities and liability of monopoly licensing. If a university were to broaden its policies so that dual monopoly had to compete with other forms of technology transfer, the limitations of the approach would become quickly evident. So those involved prefer to mislead university officials, faculty, government officials, and the general public with regard to the activity.
The aspirations–that university inventions will become beneficial to the public–are treated as fact and ascribed to the workings of the dual monopoly approach. In this way, the dual monopoly approach perpetuates itself. It’s just that it is among the least effective ways–and therefore among the most crappiest–to transfer inventive work done at a university.