Migitating Private Business Use
In a recent on-line article, Peter H. Serreze at Ropes & Gray, discusses private business use. The article offers strategies to mitigate the effect of the 5% limit on private business use (PBU) for the proceeds of a bond issue. Serreze points out an approach that universities could take that would prevent private research arrangements from ever reaching private use limits:
A nonprofit research organization can promote simplicity by including, in addition to tax-exempt debt, some taxable debt or other equity in the plan of finance for any construction project. Typically, only a relatively small portion of the organization’s research funding is derived from commercially sponsored arrangements. A relatively small equity financing component— say, 20 percent of the plan of finance — may be sufficient to enable all industry-sponsored and quasi-commercial activity to be allocable to equity. If so, the organization may entirely avoid the need to structure those activities in a manner that avoids PBU, which would reduce its compliance burden and potentially expand the range of business opportunities it could pursue. (734)
In other words, if the administrators responsible for building research facilities actually wanted to attract industry-supported research on flexible terms not dictated by strange IRS safe harbor guidelines, they could readily do so. The reality, though, is that administrators apparently either don’t want to or don’t care.
The final portion of Serreze’s article looks at Rev Proc 2007-47. Serreze points out that “little authority exists on the types of facts and circumstances indicative of PBU in the area of commercially sponsored research” (739). How true. Serreze works through a possible interpretation of Rev Proc 2007-47 and finds that if one focuses on a company sponsor’s intentions, then even the intention to obtain access to research results might be construed as a “specific commercial objective” and cause the safe harbor to fail. Serreze recommends looking instead to the university’s motives for conducting the research. Here is where things get dicey. First let’s work through Serreze’s ideas–which are very good–and then let’s look at the problems.
Navigating Rev Proc 2007-47 with a “Fair Market Value Savings Clause”
Serreze lays out the elements of Rev Proc 2007-47–that the sponsored research is a) basic research; and b) any results acquired by the sponsor are priced at fair market value. Serreze then looks at two variations on an upfront agreement that would likely fail Rev Proc 2007-47’s safe harbor: 1) a royalty rate in a set range; and 2) a royalty rate adjusted for the sponsor’s contribution. The first still pre-sets the value; the second suggests the rate will be below fair market value. Serreze, however, suggests a third alternative that he suggests should comply with the safe harbor:
‘‘The royalty rate will be 2 percent per year,
provided that if either party believes that such rate
does not represent a fair market royalty at the time
the option is exercised, the parties will negotiate a
fair market royalty rate.’’ (740)
This version includes a “fair market savings clause”–stating clearly that a condition of the deal is that a fair market rate is paid. The risk is that this clause still presets the consideration for the technology. In essence, a fair market value savings clause approach allows the parties to state a presumed or conditional value for IP that they can anticipate, and postpone confirmation of this amount pending a determination of the actual IP created and the market conditions at the time the IP is “available for license.”
There is another alternative to consider. What is a “market” for what often is a poorly documented and tested research result? In university experience, there hardly is any such “market.” There is no bustling bazaar full of extroverted patent sellers haggling over deals that university folks can trot on down to. There is no on-line exchange, no eBay, no shopping mall. What, then, is “fair market value” for a research result? Let’s dismiss such things as “salvage value,” “the cost of replication,” “the cost to invalidate the IP,” and “the cost of beating back another savage university infringement claim.” Each of these represent a kind of “value,” but not a fair market value. Let’s also set aside “the value to a competitor to the research sponsor if the competitor could prevent the sponsor from benefiting from the research the sponsor so foolishly supported.” That, too, is not a fair market value.
What do we have left? A negotiation between a university and a sponsor over generally iffy IP that lacks a market. What does such a negotiation entail? No, not an analysis of comparables–there usually aren’t any. How about a “business plan”? No, most research sponsors are not going to supply a business plan so clever university administrators can burn the plan into a contract and demand one quarter of the sponsor’s pre-tax profit from sale of products covered by licensed IP. Actually, what is most likely to happen is that the sponsor will propose a figure, and the university will accept it, perhaps after fussing around, but maybe gracefully. That figure represents market value in the absence of a market. So an alternative clause might be:
The royalty rate will be set by the sponsor at the time the technology is available for license, provided that if the sponsor desires a non-exclusive license and sets a royalty rate that the university believes is above fair market value, the parties will negotiate a rate that reflects fair market value.
That is, if the sponsor is willing to pay a high value for non-exclusive access, then the university would be constrained to charge all others that same high value to preserve the appearance of “fair market value.” But no one else would be willing to match that price, then, again, there’s no market and hence the sponsor’s price is not at fair market value. In this version, no price is set in advance, but the sponsor is permitted to set a price, so long as it does not exceed fair market value–whatever fair market means in a context in which there is no market.
The upshot of all this is that universities should be able to navigate readily federal tax law and provide industry with a favorable research environment. There is no need to force industry sponsors to be uncertain about pricing for research results they support. Further, as Serreze points out, one can avoid Rev Proc 2007-47 by forming relationships that do not involve sponsored research, such as material transfer arrangements. One also is outside the safe harbor if the research is not ‘basic” research. What matters is the motive behind the institution’s practices involving the research. And here is where things go bad.
Universities with No Brain
A university, like other institutions, has no brain. It cannot think and it cannot have motives. A university’s motives are those of its executives, as may be stated in public or private, in policies and procedures, in contracts, in correspondence, and in internal documents. A university, however, does have a charter, and that charter is used to determine the university’s federal tax standing. Thus, when university executives state purposes that are outside of those of its charter, they create their own trouble.
Here, for instance, is the University of Delaware’s statement of purpose from its charter of 1953:
The leading object of the University shall be to promote the liberal and practical education of persons of all classes in the several pursuits and professions in life through the teaching of classical, scientific and agricultural subjects, the mechanical arts, military tactics, and such other subjects as are related to and will contribute to the achievement of the objectives of a Land-Grant, State University.
Now what happens when administrators decide to own faculty IP, and then dedicate university resources to the “commercialization” of that IP?
Here is the University of Delaware’s Intellectual Property Protection, Onwership, and Commercialization provision on ownership of IP:
It is policy of the University that all inventions and discoveries, together with any tangible research materials, know-how and the scientific data and other records of research including any related government protections (collectively “Intellectual Property”), which are conceived or reduced to practice or developed by University faculty, staff, or students in the course of employment at the University, or result from work directly related to professional or employment responsibilities at the University, or from work carried out on University time, or at University expense, or with the substantial use of University resources, shall be the property of the University.
It has all the usual defects–it is an assertion of ownership, not a delegation of authority to follow a procedure or to negotiate with faculty inventors; it conflates patentable inventions with a host of other things such as “know-how”; it has a sketchy allusion to the Bayh-Dole Act; it leaves open what constitutes “employment” of faculty; it expands its claim of ownership beyond “employment” to include anything “related” to “professional” responsibilities (which would not be responsibilities to the university, as those would be covered by “employment”); and with the ambiguity of what “university time” means for faculty members. In short, this is a finely mutated university IP ownership policy. The policy asserts ownership of what the state does not own by federal law–faculty authored works and faculty inventions. Beyond the usual defects, the policy fails to discuss commercialization and spends most of its effort on who will pay for patent work and how the spoils of licensing and litigation will be divided up (hint–as the money goes up, the faculty inventors get less).
The policy justifies this taking (and the rest of its foolishness) with a preamble that includes:
The University of Delaware strives to support its faculty, staff, and students by pursuing the commercial development of intellectual and tangible research property resulting from University research. The early transfer of such knowledge and rights is consistent with the University’s mission of creating new knowledge and facilitating its application to the benefit of the public.
One might note that “transfer of knowledge and rights” is cover for “operating a patent licensing shop.” But the real problem is that the policy states a “mission” for the university that is not in the purpose statement of the university–that of “creating new knowledge and facilitating its application for the benefit of the public.” While it may be a popular thing to say–that universities have a mission to create new knowledge–it does not mean that this mission is formally an element of a university’s charter.
Whoever drafted this policy was just making things up. It is easy to argue that licensing patents is consistent with a mission of licensing patents–and that’s about all this policy claims. We will refrain from discussing what “early transfer” means and what the restriction of “to the benefit of the public” does to alter licensing practices. We will not wonder what “such knowledge” refers to, since the preceding sentence is concerned only with property, not knowledge.
Consider, instead, the claim that the university “pursues” the “commercial development of intellectual and tangible research property.” This is a plain statement of purpose that is outside any reasonable interpretation of the university’s charter. Perhaps the intent of “commercialization” is “to attempt to license patents to anyone willing to pay us more than it costs for us to take state ownership of inventions from faculty inventors and file patents.” But that is not the plain words of the policy. The policy claims that the university intends to commercialize intellectual and tangible properties, and to that end, the university policy demands ownership of such property
If we turn to the University of Delaware’s technology licensing program, called the Office of Economic Innovation & Partnerships, we find general statements under the heading “What We Do”:
. . . the University of Delaware Office of Economic Innovation and Partnerships (OEIP) works as a gateway, enabling outside entities to access the University’s knowledge-based assets—and UD personnel to form partnerships outside the University.
. . . OEIP has worked with the state, Delaware Technology Park, numerous researchers and companies in creating a culture where innovation and entrepreneurship can thrive in Delaware.
With part of its focus on discovering inventions, OEIP helps individuals develop and market intellectual property in addition to building those assets into businesses.
The particular portion of the OEIP that handles the licensing work is the “Technology Transfer Center”:
The Technology Transfer Center (TTC) within OEIP is responsible for the protection, management and commercialization of the University of Delaware’s intellectual property assets.
It’s clear–the university is involved in commercialization. The OIEP language about “helping individuals develop and market intellectual property” leaves out the bit that the university claims to own that intellectual property and intends to deploy it by “commercialization” as distinct from, say, instruction, publication, and broad public access.
Revised University IP Policies Outside Unrelated Business
If we take these elements together, we find a university that has a stated motive to take ownership of faculty property and knowledge (not in the charter) and commercialize it (not in the charter). Who needs private commercial sponsors when the university administration itself has set out a pattern of trade that is not consistent with the university’s charter, has asserted commercial motives for its activity, and indeed has stated a specific commercial intent with regard to the research that faculty conduct under its auspices. Such activity does not come directly under Rev Proc 2007-47 because there is no sponsor of research. However, this activity may constitute unrelated trade or business.
Providing services to faculty inventors does not necessarily run outside a university’s charter. Even accepting IP for management doesn’t. But when one states that the purpose of management is to commercialize the IP, then there’s a problem. Commercialization is not instruction, though instruction may be involved. Commercialization is not regional economic development, though there may be some economic development. Commercialization is not public benefit, though there may be public benefit. Commercialization, here, means the use of university-owned IP to create new products for commercial sale, presumably with the help of investors and companies. But much as such new products may be a happy result, the activity itself is not a university purpose. Certainly, people can choose to do whatever they want at a university, but just because they do so does not mean that the university’s charter must be stretched to include whatever they do.
If one were to challenge a university with a licensing operation such as the University of Delaware’s–and the UD policy structure is typical of how technology transfer folks have gone about creating their empires on top of faculty and public interests–the challenge would be that the university itself is acting outside its charter and thus is not a qualified user of the facilities built with tax-free bonds.
Consider 26 USC 513(a) (I have focused on the parts pertaining to universities):
The term “unrelated trade or business” means, in the case of any organization subject to the tax imposed by section 511, any trade or business the conduct of which is not substantially related (aside from the need of such organization for income or funds or the use it makes of the profits derived) to the exercise or performance by such organization of its charitable, educational, or other purpose or function constituting the basis for its exemption under section 501 . . . , except that such term does not include any trade or business—(1) in which substantially all the work in carrying on such trade or business is performed for the organization without compensation; or(2) which is carried on, . . . in the case of a college or university described in section 511 (a)(2)(B), by the organization primarily for the convenience of its members, students, patients, officers, or employees,(3) which is the selling of merchandise, substantially all of which has been received by the organization as gifts or contributions.
The upshot is that a university’s IP program has to be “substantially related” to the purpose “constituting the basis for exemption” under the tax code. That is, it means nothing for a university to assert that its IP program is related to whatever administrators assert they can do from their university base. To avoid a finding of unrelated trade or business, a university IP program has to show that its activities relate to the university’s purposes as considered by tax law. It is possible to have an IP program that avoids a finding of unrelated trade or business–it may relate directly to the purpose for which an exemption is granted, such as conducting technology transfer is a form of instruction. Or, the IP program can operate without compensation, or can be for the convenience of faculty, students and the like–for instance, by not demanding ownership of all IP, but only accepting IP for management at the request (for the convenience of) faculty inventors and others.
Over the past thirty-five years, universities have changed their programs. Where previously universities did operate their licensing programs, if they had them at all, on a largely volunteer basis, universities now uniformly have changed their policies to demand ownership and participate directly in “commercialization”: they have made their practice systematic, they operate venture funds, they start companies which they house in their facilities, they seek profit, they litigate and threaten litigation. There is nothing that substantially differentiates such activity from comparable activity in the private sector, other than happy talk about public benefit and puffy words about “transferring knowledge and rights” instead of “licensing IP for profit.”
IRS Audits to Restore University IP Policy to Productive Strategies
If I were a bored IRS agent, I’d be looking to audit any university that has changed its IP policy in the past ten years. I would be looking at how the stated practice of the university directly relates to the stated purpose under which the university enjoys an exemption from unrelated business income taxes. I would ask how a university’s practices and motives may have changed, based on changes in policy statements and practices, whether formal or informal. I would expect to find most universities–ones with comprehensive, compulsory, systematic intellectual property policies tied to a licensing operation focused on commercialization are not within the exceptions of 26 USC 513(a).
Such a finding would not be merely one of a dispute over wording. It would be a dispute over the treatment of faculty (and student, and volunteer) knowledge and freedom to innovate. For that, the IRS may be a key player in requiring universities to comply with the interpretation of the Bayh-Dole Act provided by the US Supreme Court in Stanford v. Roche.
Universities have a valuable role to play in the broad scheme by which we discover and create new technology and new social practices. University administrators have an important role to play, as well. But the role of the university is not as bureaucratic patent bully. Nor is the role of the administrator to systematize profit-seeking from intellectual property and anything else administrators can take. The role of the university is to provide an environment in which instruction can take place. Research is a branch of that instruction, and the publication of research is one established form of public instruction. Patenting is as well a form of publication, but one with implications for commercial and non-commercial uses alike. A university might better write a policy that limits the uses of patenting than to write one in which the university seeks to use patents in the pursuit of profits, to “commercialize” faculty knowledge rather than to facilitate instruction.
Prior to the Bayh-Dole Act, most universities operated in the manner I have just described. That description is no ideal impossible world never to be witnessed. We had that world. It was operating well. It was the metrics of that world that were cited in support of Bayh-Dole. It was that world to which Congress thought it might be good to make available as well more of the results of federally funded research conducted by university faculty. It was this world that university administrators dismantled over the past thirty-five years. It was this world that AUTM and others aimed to finally destroy altogether in Stanford v. Roche. But the Supreme Court rejected their argument. That better world still lives. There can still be a return to that world–though it will be more like a re-discovery, more like the bewilderment of freed slaves working out what freedom means than the “free men” of Vannevar Bush’s day with the confidence to get on with new things.
It is easy to undo most university compulsory IP policies with tiny edits. For “shall be the property of the University” read “may be assigned voluntarily to the University.” Simply change the policy mandate on administrators from one that demands ownership to one that authorizes administrators to make university resources available if inventors choose to assign ownership to the university. That’s the only snip-snip necessary to make sure that university IP policies do not continue to breed bureaucracy, bitterness, and antagonism to innovation.