Professor Paul David will be giving a “Farewell Lecture” for the Oxford Internet Institute, at 1 St. Giles’ , Oxford OX1 3JS at 4pm on July 29, 2008. Attendance is open, but space is limited. If you would like to attend, please e-mail events@oii.ox.ac.uk to confirm a place. You are also invited to comment on this blog. The following text is the title and an extended abstract of Paul David’s talk, posted with the author’s permission:

The ‘Oxford model’ and the ‘anti-commons’ threat to the long-run vitality of collaborative e-Science

The case for new moves in ‘legal jujitsu’ to mitigate adverse unintended consequences of growing university engagements in commercial exploitation of publicly funded research

By

Paul A. David
Quondam Senior Fellow, OII & Emeritus Fellow of All Souls College, Oxford

PRÉCIS

The rapid spread during the past two decades of policies that have encouraged universities and other public research organizations (PROs) to claim and commercially exploit intellectual property rights based upon their faculties’ research findings has stirred both expressions of worried concern and enthusiasm. In a recently published article, two distinguished scientists (with Oxford connections) join the current and former managing director of Isis Innovation (the university-owned technology transfer company established by Oxford in 1988) in presenting a restrained version of the enthusiastic view (Cook et al., 2008). The authors contend that the controversy surrounding the growing commercialization of university research arises not so much from any real conflicts between the academic and business missions and their respective research cultures as from mutual suspicions and distrust about the division of the financial benefits and the efforts entailed in a successful “technology transfer.” These difficulties, and related conflicts of interest, they suggest, are resolvable by following the principles of ‘the Oxford model’—by separating the phases of research, technology transfer and actual market exploitation of the patents, and providing a trusted intermediary operating on the lines of Isis Innovation to develop and manage relationships between academic researchers the executives of the business companies involved.

Apart from the university taking equity-ownership positions in the companies “spun-out” by Isis Innovation, and the latter company taking an active role in “assisted’ consultancy contracts between university researchers in external parties, the arrangements described as ‘the Oxford Model’ are appear quite unexceptional in the general run of university-based technology transfer organizations (TTO’s). What is remarkable about this particular article is the authors’ presentation of the problematic issues as arising from the unfortunate (and avoidable) confusion of discussions about technology transfer by the interjection of considerations about “the setting of priorities of the university itself.” Their concluding message is simply this: “When developing technology transfer systems, the active support of senior university personnel, both researchers and administrators, is essential.”

This parochially ‘Oxonian view’ fails to confront most of the issues that have been raised in the extensive empirical research and policy discussions that economists and other social scientists in the science and technology policy field have devoted to issues arising from the growth of “academic patenting”, and the activities university TTO’s in the U.S., the EU, and other OECD countries. These concern the systemic consequences affecting the long-run vitality of fundamental, exploratory scientific research. They are about impacts on the social missions of universities as direct and indirect recipients of public patronage, about the congruence between the latter’s requirements of independences from external pressures and the acceptance by these institutions of the parity of the “third stream” contributions to society that they are being called upon to by actively exploiting “their knowledge assets” in the private business sphere. Even that does not exhaust the list of questions that have been vaguely alluded to vaguely otherwise disregarded, and which includes doubts about the efficacy of university as opposed to business ownership of IP rights, about the economic efficiency of the system of individual university-based TTO’s, and about the potential for distortion of university internal procedures and resource allocation decisions, including the time and attention that academic administrators have to devote to questions arising from these new university missions. (See David and Metcalf 2008, for an extract from a more extensive EC DG-Research Expert Group review of these questions.)

The lack of a “systems perspective” on these questions is perhaps the most surprising and serious deficiency among those in the foregoing catalogue. Having given a considerable amount of attention during my time at Oxford to elaborating such perspectives issues in science and technology policy – this one being prominent among them, I will take the occasion of this lecture to try to articulate what for me is a central source of concern for the long-term vitality of publicly funded, academic science, and what can be done to protect is it from the unintended damage that can ensue from many universities around the world embracing ‘the Oxford model.’ My focus necessarily must be narrow, and it will be restricted to the important stream of ongoing lively discussion and debate among legal scholars and economists that deals with the problem of the ‘anti-commons’. Following Heller and Eisenberg (1998), this concerns the existence and the seriousness of obstacles to scientific discovery, invention and innovation that may arise as an unintended consequence of “over-patenting”— particularly, the extensive, wide-spread patenting of research methods and tools that emerge in “frontier” fields, where exploratory science spawns new techniques with novel applications.

Although initially framed in terms of “over-patenting,” the anti-commons as a conceptualization of the perverse resource allocation effects of the distribution of private ownership rights has a considerably wider potential range of practical relevance, and so warrants commensurately more careful examination. My first purpose, then, is to underscore that analytical point by taking the discussion beyond the domain of academic patenting, presenting a stylized model of the impediments imposed upon the conduct of research by the burdensome licensing charges that can result from the distribution of monopoly rights over multiple complementary database resources. Doing so acknowledges the emergence and growing role of digital databases as critical facilities of the research infrastructure in many scientific and technical domains, a central theme in current policy thinking about the so-called cyberinfrastructure for e-Science. Substantively, it points to the generality of the phenomenon of “multiple marginalization” due to the uncoordinated exercise of market power by database owners in setting licensing charges. The latter is likely to emerge from the concurrent efforts of many different university-based research groups being encourage to disclose findings on which their respective host institutions can file patent claims, but that also is the economic core of the anti-commons problems created by intellectual property rights and technical impediments to the federation of databases.

Several analytically distinct aspects of the “anti-commons problem” may arise from the dispersed distribution of intellectual property ownership and consequent exclusion rights. Searching to locate the owners of relevant rights, negotiation with those rights owners from whom access rights are needed, and paying the pecuniary charges for the licenses that are granted, each may impose costly burdens on enterprises that require access to the use of such assets when they are both numerous and in the hands of many other parties. Distinguishing among these potential sources of cost for scientific and technological research enterprises is important, because their economic implications are differently affected by the structure of productive relationships among the resources that enter into the resource process, and particularly by the degree of technical complementary among various “research inputs that fall under the control of diverse owners of “exclusion rights” (such as are conveyed by IP ownership). Further, dealing separately with these parts of the anatomy of the anti-commons recognizes that such inefficiencies in the allocation of research resource as they would occasion may differ in seriousness, be amenable in different degrees to market solutions, or, failing that, require distinctive institutional remedies.

By focusing within this framework upon on the differential incidence of “multiple-marginalization” effects on exploratory and applications-oriented R&D underscores the need for a more nuanced approach in empirical efforts to assess the ways in which this and other cost-imposing dimensions of the anti-commons problem would manifest themselves. Reconsidered from that angle, the particular questions posed to academic researchers by several of the pioneering survey- and interview-based studies of the impact of patented research tools in the biomedical area are seen to have invited overly sanguine conclusions. This is because they implicitly supposed that the existence of a “serious anti-commons” effect would take the form of the blocking or abandonment of research projects. That such events, reportedly, are “as rare as the White Tiger,” are not surprising and, instead, are consistent with the existence of more subtle but cumulatively distorting long term effects on the advance of fundamental science, upon which the capability of modern economies to sustain innovation and long-term economic progress rests.

This perspective, however, obliges us to inquire whether the problems created by the distribution of fragmented IP ownership cannot be mitigated, if not solved, by the same market processes that created them. The second part of the paper takes up that question in two specific connections, considering first the proposal that the existence of transferable IP rights would allow the problems of search and negotiation costs to be mitigated by the development of institutionalized solutions modeled on copyright collection societies; that these might also be a palliative for the “royalty stacking” created by uncoordinated pricing of bundles of patents that constitute “thickets.” The second connection is simply a more general formulation of the latter claim — viz. that owners of complementary intellectual property rights may well have private profit-incentives to exploit those rights in a collectively managed “pool,” and could therefore act spontaneously to mitigate the worst inefficiencies of multiple-marginalization. But, on closer inspection, the copyright collecting society-like mechanisms that have been proposed are revealed to be inadequate in dealing with the core source of the inefficiency arising from widely distributed exclusion rights to complementary research assets that are protected either by patents or by technical means, such as encryption in digital rights managements system. Furthermore, like unregulated patent pools, their history exhibits considerable potentials for the abuse of IP-based monopoly power. Thus, the burden of this part of the argument is that substantial doubts surround the wisdom of the implied policy recommendation to leave “anti-commons problems” to be remedied by the workings of new institutions engendered by forces in the markets for valuable intellectual property.

Some positive policy action therefore appears to be called for, particularly in view of the distribution of IP rights to exploit results of publicly funded R&D projects that already has been produced by the workings of the Bayh-Dole Act of 1980 in the U.S. and other kindred legislative and administrative measures introduced in the EU and elsewhere. The discussion of suitable policy measures in the paper’s aims to (a) clarify the meaning and practical significance of the idea of legally creating a “information common” for scientific and technical research communities by means of common-use contracting, (b) inquire into the conditions under which these are likely to emerge spontaneously as “clubs” or “pools” among holders of IPR in research tools and databases, rather than having to be pro-actively encourage by public agencies, and (c) consider specific policy measures that would be appropriate and effective in promoting participation of universities and other public research organizations in IPR licensing arrangements of that kind.

In sum, the policy thrust of my argument is tantamount to a proposal to pursue an indirect path towards reform of the workings of regimes initiated by the Bayh-Dole and Stevenson-Wydler Acts in the U.S., and the parallel legislative measures subsequently were introduced in the UK and other EU member states. The closing part of the lecture sketches in outline form specific institutional arrangements for the administration of “scientific research commons” (SRCs) formed by IP right-holders. These arrangements must address: (i) conditions of eligibility to participate, (ii) limitations upon the scope of legally protected content that can be placed within the commons, (iii) principles for the management and pricing of licenses granted to non-commoner for use of intellectual property rights contributed to, and arising from the utilization of pooled research assets, (iv) relationships among independent SRCs and between SRCs formed by universities and other public sector research organizations that presently maintain technology licensing/transfer offices, and (v) the implications of competition policy safe-guards against the creation of inefficient pools, and the abuse of patent cartel power.

References

Cook, Tim, Baruch Blumberg, Raymond Dweck, and Tom Hockaday, “Commercializing University Research: Threats and Opportunities – The Oxford Model,” Capitalism and Socialism, 3(1), 2008 (Art.4). [Available at: http://www.bepress.com/cas/vol3/iss1/art4/%5D.

David, Paul A., and J. Stanley Metcalfe, ““’Only Connect’”: Academic-Business Research Collaborations and the Formation of Ecologies of Innovation,” forthcoming in Knowledge and Innovation in the Triple Helix, H. Etzkowitz and R. Viale, ed., Cheltenham, Eng.: E. Elgar, 2008. [Available as SIEPR Discussion Paper No.07-033 (January 2008) at: http://siepr.stanford.edu/papers/pdf/07-33.html .]