Intellectual Property Policy Reform: Fostering Innovation and Development

Intellectual Property Policy Reform Intellectual Property Policy Reform Fostering Innovation and Development Edited ...

Author: Christopher Arup | William Van Caenegem

58 downloads 1996 Views 1MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

Intellectual Property Policy Reform

Intellectual Property Policy Reform Fostering Innovation and Development

Edited by

Christopher Arup Professor, Faculty of Business and Economics, Department of Business Law and Taxation, Monash University, Melbourne, Australia and

William van Caenegem Professor, Faculty of Law, Bond University, Queensland, Australia

Edward Elgar Cheltenham, UK • Northampton, MA, USA

© Christopher Arup and William van Caenegem 2009 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical or photocopying, recording, or otherwise without the prior permission of the publisher. Published by Edward Elgar Publishing Limited The Lypiatts 15 Lansdown Road Cheltenham Glos GL50 2JA UK Edward Elgar Publishing, Inc. William Pratt House 9 Dewey Court Northampton Massachusetts 01060 USA

A catalogue record for this book is available from the British Library Library of Congress Control Number: 2009928607

ISBN 978 1 84844 163 7 Typeset by Cambrian Typesetters, Camberley, Surrey Printed and bound in Great Britain by MPG Books Group, UK

P EE

Contents Acknowledgements List of contributors

vii viii

1 Themes and prospects for intellectual property law reform Christopher Arup and William van Caenegem PART I

INTELLECTUAL PROPERTY GRANTS

2 Why patents need reform, and some suggestions for it William Kingston 3 What are the costs and benefits of patent systems? Hazel V.J. Moir 4 Strong patent rights, weak patent standards and innovation in biomedicine Dianne Nicol 5 The jewel in the crown: India’s patent office and patent-based innovation Peter Drahos 6 The first steps in remedying the relationship between patents and competition Charles Lawson PART II

1

11 29

55

80

101

OPEN INTELLECTUAL PROPERTY SYSTEMS

7 An introduction to open source biotechnology Janet Hope 8 Intellectual property, innovation and openness Ulf Petrusson and Caroline Pamp 9 Wikipedia, collective authorship and the politics of knowledge Matthew Rimmer 10 Copyright and the new street literature Megan Richardson and Jason Bosland

v

129 154 172 199

vi

PART III

Contents

INTELLECTUAL PROPERTY TRANSFERS

11 Commercialization of university research and free diffusion – what does experience show works best in and for Australia? Ann L. Monotti 12 Pervasive incentives, disparate innovation and intellectual property law William van Caenegem 13 Commodifying sheer talent: perverse developments in the law’s enforcement of restrictive covenants Joellen Riley 14 Split entitlements? Intellectual property policy for clusters and networks Christopher Arup 15 Conclusion William van Caenegem and Christopher Arup Index

221

250

267

285 306

311

Acknowledgements This book grew from an idea we had when researching worker mobility and knowledge transfer. We felt there was much thoughtful work being done concerning the reform of intellectual property law generally. The collection we proposed would draw attention to the kind of reforms that in fact are worth considering, taking account of the changing context – that in fact are already being tried whether the orthodoxy has accommodated them as yet or not. The contributors willingly responded to an invitation and came to a workshop at Bond University on the Gold Coast, with little more than the promise of a sunny sky outside the window of the Faculty of Law’s new resource centre. One came via videolink from early morning Dublin. Everything went smoothly (except for a lost wallet) because we were able to rely on Cherie Daye’s peerless organisation of the workshop. Then, we were much obliged to the contributors, our colleagues, for the timely and professional way they reworked their papers and responded to the requests of the editors. At this point, we were fortunate to obtain the assistance of Ginny O’Grady, who has done the copy editing at Monash so thoroughly and sensitively. In completing the volume, we should also like to thank Luke Adams, Nep Elverd and Joanne Betteridge at Edward Elgar for the confidence they have shown in the project. We acknowledge the crucial financial assistance that our Faculties, the Faculty of Business and Economics at Monash and the Faculty of Law at Bond, have given us. We hope you find sustenance in the volume. A list of contributors is included in this book and we are sure they would be interested to discuss their ideas further. Please note that the law and policy are current as at the date of submission, September 2008. Chris Arup and William van Caenegem September 2008

vii

Contributors Christopher Arup – Department of Business Law and Taxation, Faculty of Business and Economics, Monash University, Melbourne, Australia Jason Bosland – Faculty of Law, University of New South Wales, Sydney, Australia Peter Drahos – Centre for Governance of Knowledge and Development, College of Asia and the Pacific, Australian National University, Canberra, Australia Janet Hope – Centre for Governance of Knowledge and Development, College of Asia and the Pacific, Australian National University, Canberra, Australia William Kingston – School of Business, Trinity College, Dublin, Ireland Charles Lawson – Australian Centre for Intellectual Property in Agriculture, School of Law, Griffith University, Brisbane, Australia Hazel V.J. Moir – Crawford School of Economics and Government, College of Asia and the Pacific, Australian National University, Canberra, Australia Ann L. Monotti – Faculty of Law, Monash University, Melbourne, Australia Dianne Nicol – Faculty of Law, University of Tasmania, Hobart, Australia Caroline Pamp – Centre for Intellectual Property Studies, Chalmers University of Technology, Goteborg, Sweden Ulf Petrusson – Centre for Intellectual Property Studies, Chalmers University of Technology, Goteborg, Sweden Megan Richardson – Melbourne Law School, Melbourne, Australia Joellen Riley – Faculty of Law, University of Sydney, Australia Matthew Rimmer – Australian Centre for Intellectual Property in Agriculture, College of Law, Australian National University, Canberra, Australia William van Caenegem – Faculty of Law, Bond University, Queensland, Australia viii

1. Themes and prospects for intellectual property law reform Christopher Arup and William van Caenegem When we set out to collect papers around our theme, we were encouraged by a mood of constructive criticism concerning the roles that intellectual property has come to play in innovation and creativity. Broadly the criticisms concern two developments: (1) the strength of intellectual property endowments and entitlements and (2) the proliferation and manipulation of claims to intellectual property rights (see Maskus and Reichman 2005). The criticism largely comes from academic scholars and sets that community apart from practitioners who are perforce involved in pursuing the possibilities within the current laws. The criticism is constructive. Very real questions remain about the social value of intellectual property, at least for some sectors of the economy and locations around the world. It is prudent to keep alternatives to intellectual property in mind when designing new policies, for there are many other ways that innovation and creativity can be furthered. Nonetheless, academic scholars are generally thinking of ways to improve intellectual property through the institution of reforms. That is the flavour of our contributions. Essentially they pursue two lines. The first is for more rigour and discernment when deciding whether to confer intellectual property rights. This, it could be said, is merely to insist that intellectual property only rewards efforts that are innovative and creative. The second is to influence how people use intellectual property rights; specifically how they transact in them. The aim is to encourage openness, once endowments have been made, in particular to promote coordination, circulation and sharing of intellectual property. We are not alone in this approach, for intellectual property scholarship seems to be in a healthy condition. Our particular contribution comes from a workshop we convened on the Gold Coast in December of 2007. The participation was international, though we do appreciate that a majority of the contributors are Australian based. The Australian interest in intellectual property reform is very high. What enables academic law scholars to identify and recommend reforms? 1

2

Intellectual property policy reform

Intellectual property scholarship forms part of a general trend within legal research to study the law ‘in context’. Intellectual property is examined and assessed as it operates within empirically mapped social fields. The fields of interest range over industry sectors, creative clusters, contract networks, media sub-cultures, developing economies and public domains. The contextualised approach is suited to reform. The contributors are able to ask whether intellectual property, as it is constituted, administered and put to use, is fulfilling the purposes ascribed to it. The context is changing. The conditions on which the old policies might properly have been based are being superseded. While the dominant model of corporate capitalism (large firm centralisation and internalisation) still fits some sectors and locations, the context changes with the onset of new technologies, business practices, work methods and social mores. If intellectual property is still an important security for investment in production, it must also facilitate dissemination and coordination of knowledge resources. To work in context, legal researchers need to call on assistance from other disciplines. Our contributors are well read in other disciplines, which is not to say that they adopt the methods of those disciplines (such as economic modelling) or indeed that they are sufficiently expert to do so. Yet they are receptive to the insights those disciplines offer and increasingly they also seek empirical evidence to inform their recommendations. Thus, Peter Drahos goes inside patent offices; Dianne Nicol talks with start-up biotech companies; Janet Hope engages with creators of open licence systems; Ulf Petrusson and Caroline Pamp with builders of technology platforms. While other disciplines have certain epistemological advantages over law, our experience suggests they will not always be as open-minded as lawyers are. For our contributors appreciate that law too forms part of social fields. In considering how legal agents influence the course of intellectual property, they have a rich and non-determinative sense of the possibilities of legal principles and legal processes. So Peter Drahos takes a close interest in patent office examiners; Charles Lawson in those commissioned to review competition policy; Ann Monotti in university technology transfer officers, and Joellen Riley in higher court judges. If this research is neither traditional legal analysis, nor the method of another discipline, how should we characterise it? The contributors see intellectual property as forming part of the regulation and governance of innovation (see Andersen 2006). So, they are interested in intellectual property in the hands of a variety of actors, including scientists, writers, artists, public institutes, universities, private firms, financial investors, government administrators, legal practitioners, legislators, judges and politicians. They seek to place intellectual property within a range of strategies or instruments of policy, some essentially outside law, but some varying within law as legislative directives,

Themes and prospects

3

default principles, administrative guidelines, licence specifications, contractual terms or conditions attached to grants. This approach is sceptical that one rule fits all. Policy making requires evidence that intellectual property will further innovation and creativity within a specific set of circumstances. William Kingston, Peter Drahos, Dianne Nicol, Hazel Moir and Charles Lawson all stress this point. Often, it will not do so directly. Instead, policy making should empower those individuals who want to work with intellectual property in this way. Janet Hope, Ulf Petrusson and Caroline Pamp, Ann Monotti, Matthew Rimmer, Megan Richardson and Jason Bosland, Joellen Riley, William van Caenegem and Chris Arup all express this preference. So, what reforms do our contributors pursue? We said that they fall into two broad categories. The first, for rigour and independence, is when new rights are created. Much of the discussion is about regulating the criteria for appropriation, that is, the outer limits of the knowledge that should properly be the subject-matter of a patent or copyright right, or a non-disclosure or noncompetition obligation. But our contributors also point to the need to be demanding when laws are being interpreted and applied to make individual grants. It is appropriate that patents are a focus of this collection because they are amongst the most powerful of intellectual properties in the knowledge they encompass and the control they enable. Copyright is addressed in the context of the new media, a little less so than patents because of its coverage elsewhere. The contributions also take an interest in the appropriation of tacit knowledge, which, in the absence of trade secrets legislation, is decided by judges fashioning the scope of contractual and equitable obligations. This too is a live issue. Within a service economy, technical know-how and the emerging forms of intellectual and social capital, such as customer and workplace relations, are growing in value. Hazel Moir takes on the whole patent system, wanting patents to be confined to the situations where they make a difference – where they induce innovation that otherwise would not eventuate. Similarly, William Kingston wants to ensure that patents reward real investments in innovation, not just the satisfaction of legal categories. Charles Lawson and William van Caenegem again are generally sceptical of grants: they want patents available only where they are needed to assist competition by lowering barriers to market entry and not, it follows, when they actually work anti-competitively. Moving down a layer into administration, Peter Drahos provides a detailed case why patent examiners in India should make their own rigorous assessments of applications from foreign companies, especially to avoid evergreening of pharmaceuticals. Dianne Nicol argues that, if patent rights are going to be strong, standards for granting patents should really insist on novelty,

4


inventiveness and utility. William Kingston recommends a cheaper means of dispute resolution – technical arbitration – than litigation of patent validity through the courts. Joellen Riley counsels judges in Australia to think hard about the consequences of surrendering to the appeals of employers for property in information. The uses made of intellectual property rights are our second target of reform. In this regard, regulatory reform does not so much direct how holders should exploit the intellectual property. It encourages and assists those who will use rights to expand innovation and open access. As these holders are regulating ‘privately’, the role required of the state is co-regulation or ‘metaregulation’ – the regulation of regulation (Hope, Nicol and Braithwaite 2008). The basic settings of intellectual property law shape the uses made of it. Policy making involves decisions about which use rights will be conferred on the holders, which uses the holders may prevent others from pursuing or controlling in other ways. Megan Richardson and Jason Bosland pursue this regulation through the recurring contest, largely in the courts, over the scope of the fair use provisions of copyright. Their case study is the revival of an old medium of expression, street news, now manifested in a new electronic internet form. They argue that the street news use of newspaper copy should not have to be ‘transformative’. Several contributions make the initial rights holders the subject of their study. The chapters are interested in those who do not want to use their rights to lock down intellectual property, but who wish to construct open licensing systems. This focus is recognition that the context is changing, that much innovation and creativity are network based or even the outcome of an all-comers knowledge bazaar or a commons-based peer production. They stress the interactive, roundabout, cumulative nature of innovation. As systems engineers, these holders are taking their own steps to work around the hold-ups and break-downs associated with strong entitlements and micro-protection of rights, as we see for example in patent thickets and a welter of copyright take-down notices. Such reciprocating and coordinating systems are largely self-generating; they are movements that are growing organically. At the same time, they are quite regulatory. Their protocols and mores are so sophisticated that some go as far as to consider them constitutions (see Joerges, Sand and Teubner 2004). But this belies their fragility. They also need a supportive public law environment in which to prosper. At this stage, they are most pronounced in the media sectors where copyright is the main intellectual property governor. Matthew Rimmer explains the dynamics of a successful system par excellence, Wikipedia, though he has his fingers crossed that copyright litigation will not foul it up. Its integrity depends principally on people’s discretion and restraint. If litigation is being brought, it will call for a sympathetic reception from the courts, sitting possibly


5

anywhere in the world, and remedial action by legislatures if the judicial environment turns hostile. Both the Janet Hope and the Ulf Petrusson and Caroline Pamp contributions point the way for similar systems or platforms to develop in the biotechnology/biomedicine patents sectors. The organisational innovations they recommend are perhaps not as wide open as the Wikipedia model. But they are more accessible than other collective responses in the past such as industry patent pools and relational contracting (Dreyfuss, Zimmerman and First 2001). These contributions envisage how intellectual property can be used for increasing the openness of innovation, citing examples that are working. Open licensing systems are crucial to avoiding lock-up of resources by upstream owners and break-downs in exchange because there are so many rights holders. Subject to their universities’ policies, university researchers can be involved in these open licensing systems. Ann Monotti captures them in more conventional intellectual property roles. They have become more like private holders, perhaps of necessity. Yet the right policies can encourage university decision makers to release the intellectual property they are not using and to seek out those licensees who want to work the inventions. Regulatory studies offer tools and Monotti identifies them at work in the proactive United States approach, including the legislative directives of the Bayh–Dole Act, the conditions the National Institutes place on money grants and the universities’ code of ethics for research. This perspective on the universities reminds us that intellectual property is mainly transacted in the commercial world. Nonetheless, as Benkler (2006) explains, the commercial interests that people show in intellectual property can vary. They can be commercial without being proprietary. Some are more positive for innovation. Rights can be used negatively against innovation. For example, litigation might be threatened to increase risks and costs, to bar others entry, to extract rents. As Dianne Nicol finds, not everyone uses intellectual property in this way. The first policy reflex is to be cautious about conferring rights at all, allowing freedom to operate. But if conferring rights is a positive, then policy should be concerned with who gets rights and what they do with them. Thus, Dianne Nicol argues that innovation is furthered if small biotech companies have some command of intellectual property; likewise Joellen Riley, William van Caenegem and Chris Arup if workers do. The final bunch of contributions to the book sees mobile individual workers as keys to the dissemination and recombination of knowledge. Large firms need some reward for the economies of scale and scope they bring to innovation. But increasingly innovation benefits from circulation either within localised networks (the high point being Silicon Valley) or inter-national movement of

6


knowledgeable and skilled people. ‘Freed up’ by organisational changes, the hollowing out of the large firms, these workers want to take intellectual property rights with them to other firms or to their own start-ups and collectives. Systems prosper if these individuals have incentives to invest in intellectual and social capital. Some of that investment returns to their first firm. However, workers should be free to put knowledge to new uses elsewhere. What intellectual property policy is conducive to mobility? Again, policy begins with a caution against over-appropriation. Where ownership can be justified, the question is who is to be free to use the rights to the knowledge? The property in question can be the rights to use explicit or codified knowledge – such as the right to work patents. In this regard, Chris Arup’s contribution considers whether it is good policy to split entitlements between employers and workers and between principals and consultants. Just as vital to innovation is the right to use tacit knowledge. In their contributions, Joellen Riley and William van Caenegem seek the appropriate limits to the scope and application of civil obligations, the non-disclosure and non-competition restraints that the judges will uphold as a matter of contract and equity. In these relationships, intellectual property and freedom of contract should be moderated by a common law competition law. In some jurisdictions, competition law becomes a legislative and administrative scheme. Competition law has the potential to allow for intellectual property while correcting for some of its abuses. That allowance can be on a categorical basis or a more searching case-by-case basis. On the basis of an Australian experience, Charles Lawson wonders if competition law policy makers have the will to regulate the uses made of intellectual property by those with market power, such as the restrictive conditions inserted into licences (where licences are given at all). Joellen Riley asks that judges recall the common law’s aversion to restraints of trade, especially before giving employers the procedural advantage of an interim injunction. While disposed to variety, our contributors do not suggest that intellectual property should be left entirely to the market. There is attraction in the argument that innovation is much too complex and unpredictable for it to be engineered socially. However, the final policy position is not one of pure freedom of contract or absolute contract autonomy. The market is already regulated when intellectual property is created. Our contributors do lean towards voluntarism when it comes to the deployment of intellectual property rights. They are showing faith that holders will be interested in promoting innovation and creativity. But the framework still matters and public regulation is still needed – at the pointy end of the regulatory pyramid – if rights are abused. One danger with contract autonomy is that the powerful will obtain more than their legal entitlement. But is it worth entertaining such reforms – or have the possibilities been


7

foreclosed? One check on our enthusiasm is the extent of intellectual property standardisation internationally. The World trade Organisation TRIPs agreement is a high point of standardisation. Checks on reform are also to be found in bilateral inter-governmental agreements and in memoranda of understanding between functional regulators such as the patent offices. Standardisation reduces the freedom to experiment with intellectual property regulation and to choose divergent legal paths. That includes the opportunity to learn from the experience in other jurisdictions, to avoid the same mistakes or simply to adapt the form to different local conditions. Yet we could pitch reform efforts at the international level. The international level provides another political and legal space in which to make reform efforts. If standardisation proceeds, the content of it is still contested, both within, and between, international organisations. Some aspects of intellectual property are not as standardised as others. They include trade secrets, employee inventions, moral rights, and the regulation of the distributions and uses of rights, including competition law – the second of the two broad pursuits of this book. Even the more established aspects of regulation, like the criteria for patents and copyright endowments, reveal gaps in coverage and explicit allowances for choices to be made at the national level. So, despite their own protestations at times, national leaders have opportunities. Their policies can be demanding of any grant of intellectual property endowments, geared to openness in the regulation of uses, and tailored to the conditions for local innovation and creativity. The failure to do so is often the result of domestic politics of interest or eagerness to please international constituencies, not for the lack of imagination of alternatives.

REFERENCES Andersen, B. (ed.) (2006), Intellectual Property Rights: Innovation, Governance and Institutional Environment, Cheltenham, UK and Northampton, MA, USA: Edward Elgar. Benkler, Y. (2006), The Wealth of Networks: How Social Production Transforms Markets and Freedom, New Haven and London: Yale University Press. Dreyfuss, R., D. Zimmerman and H. First (eds) (2001), Expanding the Boundaries of Intellectual Property: Innovation Policy for the Knowledge Society, Oxford: Oxford University Press. Hope, J., D. Nicol and J. Braithwaite (2008), ‘Regulatory Capitalism, Business Models and the Knowledge Economy’, in J. Braithwaite, Regulatory Capitalism, Cheltenham, UK and Northampton, MA, USA: Edward Elgar. Joerges, C., I.-J. Sand and G. Teubner (eds) (2004), Transnational Governance and Constitutionalism, Oxford: Hart Publishing. Maskus, K. and J. Reichman (eds) (2005), International Public Goods and Transfer of Technology under a Globalized Intellectual Property Regime, Cambridge: Cambridge University Press.

PART I

Intellectual property grants

2. Why patents need reform, and some suggestions for it William Kingston There is an emerging consensus amongst users and students of the patent system that it is not performing as it ought to, and that this performance is getting worse. A recent comprehensive analysis of a large number of US empirical investigations concluded that: The performance of the United States patent system deteriorated markedly during the 1990s as the private costs of patent litigation soared. By the late 1990s, the risk of patent litigation for public firms outside of the chemical and pharmaceutical industries exceeded the profits derived from patents. This implies that patents likely provided a net disincentive for innovation for the firms who fund the lion’s share of industrial R&D, that is, patents tax R&D. (Bessen and Meurer 2008, p. 144)

And since most other countries, as will be seen below, have replicated the US system, it is a reasonable assumption that this pattern is also the global one. The simple reason why this has happened is that it is nobody’s business to see that patents work properly. Consequently, they have been shaped by the interests of those who could benefit from them, rather than from any vision of the public good. The subtitle of the book from which the above quotation is taken is ‘How Judges, Bureaucrats and Lawyers Put Innovators at Risk’.

I.

INVESTMENT IN INNOVATION

Economic innovation depends upon investment, partly under uncertainty and partly under risk, which is uncertainty quantified. Investment under uncertainty can never be rational at all, and investment under risk can only be rational if there is the prospect of a reward that is commensurate with the risk as it is perceived. Since innovation is the turning of information into concrete reality, the risk has two dimensions: the attempt to innovate may fail; and even if it does succeed, it may not deliver the hoped-for returns because of others freeriding on the results. ‘Information is hard to produce, easy to reproduce, and therefore difficult to profit by’ (Nordhaus 1969, p. 144). 11

12


Preventing free-riding requires market power. This is not the power to make a market, in the sense of some physical or mental arena to which there is freedom of entry, and within which price is the arbiter between supply and demand, as in a stock exchange. Market power, in fact, is the precise opposite, being the power to unmake the market, by preventing entry to it by others. Freedom of entry drives price down, thus eliminating the reward that an investor in innovation has to have to justify his risk. Only market power, the power to prevent freedom of entry, can make it possible for an innovator to hope for the above-average returns that the above-average risk of investing in innovation demands.

II.

TYPES OF MARKET POWER

Although patents are invariably thought of as the characteristic means of making it rational to invest in innovation, in fact they rank in importance far behind large-scale investments in capability and persuasion. Empirical studies on how firms capture the rewards from the results of their research and development (R&D) in the US, Europe and Japan all agree that, except for chemicals, patents are ‘unambiguously the least central of the major appropriability mechanisms’ (Cohen et al. 2002, p. 1360). The market power of capability results from investment in productive assets, and the market power of persuasion results from similar investment in psychological assets, that is, in reputation. In each case, the power comes from scale. If a firm is able to produce on a larger scale than its competitors, it can then afford to innovate because its unit costs will be lower than theirs. Even if they can obtain access to the originator’s information, they will be at a disadvantage in making use of it. The same is true if a firm is able to invest more than its competitors in building its reputation, typically through advertising. Firms which have less established reputation, known as ‘brand equity’, are correspondingly less able to copy with commercial success. As long as these two types of market power are available to firms, therefore, we could expect that there would be economic innovation because of the protection they give to investment to produce information of certain kinds. But not all information can be protected in these ways, and consequently there is need for other kinds of protection also. For example, the end-product of pharmaceutical R&D is a chemical formula. Capability market power is a poor protector of this information embodied in a pill because, once the formula is known, manufacture of the pills in quantity does not require much investment in productive assets by someone who wants to compete. Similarly, if the pill is an effective remedy, it will sell easily without the need for much marketing effort, so again the market power of persuasion cannot set up much of a barrier

Why patents need reform

13

to entry by competitors. Consequently, if there is to be invention and innovation in cases where the information is of this kind, there is need for a third means of preventing free-riding. This is specific market power, of which intellectual property is an important component, and of course patents in turn are part of this.

III.

EVOLUTION OF PATENTS

The forces in society which shaped the patent system in its origins did so in terms of a vision of the public good. Early patents protected innovation rather than invention, since their subject matter was technology found abroad and introduced to another country. The granting of a 14-year monopoly in England for bringing ‘new manufacture within the realm’ was typical of this kind of patent. The root of protection of invention rather than innovation is Article 1, Clause 8 of the United States Constitution, and a vision of the public good was again the shaping force. There can be little doubt that the motivation of the Founding Fathers for this was genuine belief that it was in the interest of society as a whole that there should be patent protection for inventors and copyright protection for authors. Subsequently, until World War I, intellectual property laws, like all other legislation, were largely formulated by bureaucrats who were independent of politicians to a large degree and equally insulated from commercial interests. Civil servants regarded their function to be very much the definition and defence of the public good. Such an attitude is evident, for example, on the part of the German bureaucracy in shaping the famous 1877 Patent Act there (Murmann 2003). This Act played an indispensable role in changing Germany from a poor country in the mid-19th century into a very rich one by the end of it, as the leading world player in the new chemical and electrical technologies. It also influenced patent legislation throughout Europe for three-quarters of a century. However, World War I and the inflation which followed it destroyed the bureaucracies which could withstand political and commercial pressures. An early deployment of such pressures can be seen at the 1925 meeting of the Paris Convention at The Hague. From the Convention’s establishment as the basis of internationalisation of patents and trademarks in 1883, there was no prohibition on any member country making local manufacture a condition of validity of a patent. In 1925, however, on the initiative of Britain and the United States, such a condition was outlawed from then onwards. These were the countries with most capability market power, and the change meant that they could now supply their foreign markets with products made at home. This enhanced their scale of operation still further, and at the same time they could

14


enjoy patent protection in their markets. Patents should be the means of enabling those who do not have the market powers of capability or persuasion to protect investment in generating information; the 1925 change can be seen as the beginning of a movement whereby they became a reinforcement of other kinds of market power instead.

IV.

UNITED STATES 1952 PATENT ACT

The evolution of patent law then went on to reflect progressive shaping by the interests which benefited from it. A fully documented illustration of this is the way in which the US 1952 Patent Act was drafted and passed. Its cause was the incompatibility between the existing criterion for patentability and the new kind of research which lay behind the antibiotic revolution. In 1941, the Supreme Court used the expression ‘flash of creative genius’ to encapsulate the level of ingenuity which it claimed had been necessary since the first patent Act of 1790. Such a criterion suited inventions made by individuals, which of course was the only kind of invention there was at the time of that Act, and for more than half a century afterwards. However, inventions coming from purposive, cooperative research in company laboratories, and from the application of science to industry, were progressively less able to meet this requirement. Eventually, no less than two-thirds of the patents coming before Circuit Courts of Appeal were being ruled invalid. What was described as an ‘ever-widening gulf between the decisions of the Patent Office in granting patents and decisions of the Courts which pass upon their validity’ was even the subject of a concerned Message from the President of the United States in 1943.1 This ‘gulf’ became a real crisis for the US pharmaceutical industry with the coming of antibiotics during and after World War II. The discovery of penicillin opened up an apparently limitless prospect of new drugs and profits, but the necessary investment in R&D could only be made on the scale which was desirable if effective patent protection were available for its results. At the same time, nothing could have been clearer than that research for antibiotics based on the techniques of large-scale screening which would be used, would be quite unable to supply the ‘flash of genius’ which the Courts required for patent validity. Once firms in the pharmaceutical industry grasped that nothing could solve this problem for them except a change in the law, they moved very quickly. In 1948, the New York Patent Bar Association drafted a Bill and was able to get it introduced in Congress, and this, supplemented by other Bills and pressures, brought about the results that the industry wanted. This Act of Congress was given its form solely by those who would benefit by it. As a judge who, as a patent attorney had been the main drafter of the legislation, wrote later: ‘The


15

[1952] Patent Act was written basically by patent lawyers . . . A good 95% of the members [of Congress] never knew that the legislation was under consideration, or that it had passed, let alone what it contained.’2 The new US criteria for patentability were then copied by other countries, so that from the 1970s the patent systems of all the developed countries were based upon the United States model. Parallel with this, the Secretariat of the Paris Convention was becoming active in spreading patents through the poorer countries of the world. This was due to the zeal of its Director, Arpad Bogsch, which led to the recruitment to the Convention of most of the 141 countries that joined it after World War II. He developed a ‘model law’ of intellectual property which the poor countries could adopt, and cultivated and provided incentives for the politicians and bureaucrats who could arrange for the adoption of this law in their countries. The model law was effectively the intellectual property law of the advanced countries of the Western world, and its promotion worldwide was done in the name of ‘harmonisation’, claimed to be in the general public interest. It was in fact the last kind of law which these countries needed, as demonstrated by the way India was able to build up a world-class pharmaceutical industry, specifically by refusing to join the Paris Convention. If it had done, it would have had to grant patents to the giant international pharmaceutical firms and its own infant firms would have been strangled at birth.

V.

TRIPS

The most egregious example of how patent law came to be made by those whose interests are served by it is the Trade Related Aspects of Intellectual Property (TRIPs) annex to the World Trade Organisation agreement of 1994. The background to this was a normal ten-year updating of the General Agreement on Tariffs and Trade (GATT), but then, as described by Jerome Reichman (1998, p. 586), The momentum of the multilateral negotiations during the Uruguay Round carried the developed countries well beyond their initial goal, which was to limit the capacity of firms in developing countries to make and export free-riding copies of hightech goods produced at great cost in the developed countries. Instead, by ... 1991, the developed countries’ strategic goal was to impose a comprehensive set of intellectual property standards on the rest of the world.

Just how this was done has been well told by Peter Drahos and John Braithwaite in their Information Feudalism and in Susan Sell’s Private Power, Public Law,3 in which her expertise as a political scientist complements their expertise.

16


A key factor was the Intellectual Property Committee (IPC), initiated by the heads of Pfizer, IBM and Du Pont to lobby for change in international intellectual property arrangements. Sell points out that an important element in achieving this objective was extraction of US commercial policy out of its foreign policy. Once this had been done, the US Trade Representative was free to act on behalf of American business interests overseas in a completely predatory way, unconstrained by any of the factors that the President and the State Department also need to take into account in their decisions, such as the country’s international reputation or its military alliances. Since the objective was to impose a world-wide regime in which intellectual property would provide reinforcement of firms’ scale and scope, the IPC had little difficulty in recruiting support from similar interests in Europe and Japan. The upshot, as Sell (2003, pp. 92 and 95) put it, was that: The IPC, in conjunction with its counterparts in Europe and Japan, crafted a proposal based on existing industrialized country laws and presented its proposals to the GATT Secretariat. By 1994, the IPC had achieved its goal in the Trade Related Aspects of Intellectual Property … accord of the Uruguay Round … In effect, twelve corporations made public law for the world. (My emphasis)

VI.

BILATERAL AGREEMENTS

In spite of this triumph of commercial interests over any concept of the public good in lawmaking, TRIPs did not work out quite as anticipated for US firms. Consequently, in recent years the US Trade Representative’s policy has been to move towards bi-lateral agreements, in which America’s greater economic strength can be deployed to the full. As one illustration of this, Australian experience has been the subject of a perceptive study (Weiss, Thurbon and Mathews 2004). What is wrong with patents can therefore be summed up as follows: The justification for their existence is that they can enable investment in invention and innovation to be made rationally by those who do not have great capability or persuasive market power. Instead, patents have been allowed to become a reinforcement of these market powers, rather than an alternative to them. This means that smaller firms and smaller countries are greatly disadvantaged relative to larger ones in terms of innovatory capacity. Hope of changing this situation requires sharing Keynes’s belief that ‘in the last analysis, it is ideas rather than vested interests, which are dangerous for good or ill’ (Keynes 1936, p. 384). It gains some support from the way in which the educational efforts of organisations such as Oxfam and the Quaker United Nations Office produced a coalition of poorer countries at the Doha and Cancun TRIPs revision meetings. This has mitigated some of the worst features of


17

TRIPs, notably in relation to sourcing generic drugs. And even within the restrictions of TRIPs, some useful general improvements appear to be possible.

VII.

MAKING DISPUTE SETTLEMENT EASIER

A major deterrent to rational investment in invention and innovation is the cost of defending patents or of otherwise being involved in litigation. Firms with large resources for this can intimidate those that are weaker, and almost invariably do so. By far the greatest practical encouragement for small-firm innovation, therefore, would be an inexpensive, quick and simple means of resolving disputes. One practical approach to this would be to introduce compulsory technical arbitration before litigation is allowed to commence. It is because of the use of arbitration that so few disputes in technical areas other than patents ever reach the Courts. In fact, patents are unique in the world of technology in not using this device to reduce the cost of dispute resolution. The difference between other technical areas and patents is that in the former the parties involved are related by a contract, which invariably includes the arbitration condition; obviously, there is no contract between two parties in contention over a patent. For settling patent disputes, therefore, any provision for technical arbitration can only be in the contract which does exist – the one between the inventor and the state, that is, the patent grant. TRIPs requires that settlement of intellectual property disputes should be dealt with in the Courts of each country, so there would have to be a right of appeal to the Courts from an arbitrator’s decision. It might therefore be argued that compulsory technical arbitration of disputes would only push back the advantage of the party with more resources for a limited time period. Intimidation would re-emerge in the form of strong firms forcing weaker ones to face the costs of appeals to the Courts. This could be solved by provision of legal aid for the respondent party, that is, the one which accepts the arbitration and does not appeal. Small firms would never appeal, since to do so would move the dispute on to the ground where their larger opponents have all the advantage of resources for litigation; larger firms would be cautious about appealing because judges pay a lot of attention to technical expertise and there would be no certainty of winning an appeal, having lost the arbitration. It is likely enough, therefore, that the total amount required for legal aid payments would be low. In any event, whatever the cost to the State of legal aid for the party that accepts an arbitration might be, it should be regarded exactly as a counterpart to the necessary expense of conventional policing of other kinds of property. If exclusive privileges are to be real property rights, the State cannot escape having to spend money on their protection.

18

VIII.


LARGE-SCALE WORKING MODEL

The likelihood that appeals from compulsory technical arbitration would be few is confirmed empirically by results from the ‘Interference’ files of the United States Patent and Trademark Office. ‘Interference’ is a procedure necessitated by the US requirement that patents are granted to the first to invent, not to the first to file. Consequently, when it is noted that two (or more) applications that might possibly be for the same inventive entity are received by the Office, all parties must provide evidence as to their respective dates of ‘conception of the invention’ and of their efforts to reduce it to practice. There are about 200 such ‘interferences’ each year and the probability of an applicant for a patent becoming involved in this procedure is about three per thousand. The Board of Patent Appeals and Interferences, which decides on the evidence submitted, acts through Administrative Judges, who are recruited from the most experienced members of the Patent Office’s Examiner Corps. They deal with every possible element in a patent dispute – novelty, non-obviousness, unity of invention, nullity factors – everything which in a dispute after a patent grant could be the subject of litigation. Interference procedure, therefore, is precisely an actual large-scale working model of compulsory technical arbitration by experts. Two of these Administrative Judges made a particular study of the working of this procedure over a nine-year period, and found that between 40 and 50 final decisions were delivered each year. Just over one-third of these were appealed to the Courts, but only about 5 per cent of them were either wholly or even partially reversed.4 All in all, therefore, in terms of how few of its decisions are successfully appealed to the Courts and of how many of its cases reach voluntary settlement, the performance of compulsory expert arbitration in the United States interference procedure is a very good augury of how well a similar system could work for settling disputes after patent grant. Any arrangement that works 95 per cent of the time, after all, is giving at least as good a result as can reasonably be hoped for in human affairs.

IX.

DECLINE OF PATENT QUALITY

The issue of patent quality has become a contentious one everywhere, most of all in the United States because of the grant of patents for computer software and business methods there. One of the reasons is growth of the practice of ‘trolling’, where firms are held up to ransom by claims that they are infringing a patent. Those who make such claims generally manufacture nothing themselves and have often bought the patent in question from a bankrupt firm.


19

They make money through threats of litigation, especially by seeking injunctions which could bring part of their target firm’s manufacturing to a halt and wreck a market. Because the dangers, costs and delays of litigation are so great, firms frequently pay up rather than fight (Reitzig, Henkel and Heath 2007). It is an ominous sign that patent litigation costs in the US have been growing at a faster rate than firms’ investments in R&D. An important reason for the high cost of settling patent disputes is that the information on which a patent is based at the time of its grant is incomplete to one degree or another. The ideal, which of course is a patent issued after a perfect search carried out by an omniscient examiner, is impossible of achievement. If prior art searches could be perfect, trolling could scarcely exist and litigation would be greatly decreased. But, just as in the Diamond v. Chakrabarty case, the US Supreme Court ruled that ‘anything under the Sun that is made by man’ is patentable, so ‘anything under the Sun’ can be prior art. It can be foreseen, therefore, that no patent office will ever be able to offer an absolute guarantee that it has taken all relevant information into account in deciding on a particular case. As has been seen, the US Patent Act of 1952 delivered a patent system suited to the pharmaceutical industry, which had designed it. Since chemical inventions rely almost completely on the patent system for protection, virtually every chemical invention is patented. As a result, even if a prior art search is limited to issued patents, it works well for these inventions. For other kinds of invention, especially in fields where patents have only been operative in the fairly recent past, much of the prior art is outside the patent records. Examiners have tended to miss this, with the result that many patents have been shown to be invalid by searches carried out later by experts outside the patent Office, often as a result of litigation.

X. ‘OPEN REVIEW’ In terms of establishing conditions conducive to innovation, it cannot be right that a patent on the basis of which investment has been made should be invalidated, and the investment lost, because of the later discovery of prior art in a foreign-language journal which had never been translated, as has often happened. The information in question would probably only have been discovered because of intensive search effort by a competitor of the patentee firm after the latter’s investment had been made and even possibly when a product was on the market. An inventor’s or a firm’s competitors are far more likely to be aware of relevant non-patent prior art than any individual patent examiner could possibly be, and they will also be motivated to call attention to it.

20


This has led to a general call for means of improving the comprehensiveness of prior art searches, including the idea of opposition before examination, which would maximise the amount and quality of relevant information available to a patent Examiner. What is called open review would force this type of effort to be exerted before the patent grant, and would thus save wasted investment errors in both human and financial terms. It is clearly absurd to continue to issue patents on an inadequate basis of information, when this could be provided by bringing the proved enormous power of open source – the collective intelligence that the Net now makes possible – to bear upon it. Wikipedia, for example, the result of large numbers of voluntary submissions, has been shown to be little less accurate than Encyclopaedia Britannica, for all the costly resources which the latter requires. Open review should prevent many patents from being wrongly granted in the first instance. It could also significantly increase the value of the patent grants that are made, because a community of experts will have contributed to their prior art search. The Community Patent Review project of New York University and the United States Patent and Trademark Office (USPTO) is a pilot study of this approach. Several firms of the calibre of IBM, Microsoft, Intel and HewlettPackard, as well as some smaller ones, have agreed to allow 250–400 of their software-related patent applications to be published immediately for the world to evaluate (IBM, in fact, has decided to open all its future applications in this way).5 Part of USPTO’s contribution to this pilot study is to give priority to such applications in the subsequent process of deciding their patentability.

XI. ‘AGREED PATENT DATABASE’ AND INCONTESTABILITY PERIOD The Gowers Report (2006) recommended that the US pilot study should be replicated in the UK, and a group of reform-minded inventors there has proposed to the authorities that modern information technology now allows for an open source database to be developed for use in patent examination. Any party could add information to this, submitted presumably according to a primary reference in the Classification. It would then be the job of the Patent Office to do the cross-referencing which would enable all the data to be picked up for examination of a patent application. In the event of a later dispute, only information already in this database at the time of examination could be used to settle questions of novelty and non-obviousness. In the long term, the most valuable prize from success of open review and an agreed prior art database would be the ability to take the step of making patents incontestable for some period of their life. Just how great the gain


21

could be in terms of increased incentive for innovation can be seen from the results of the US Orphan Drug Act of 1983. The background to this is that there are numerous disorders (‘orphan’ diseases) which affect too few people to justify the investment that large drug firms have to make to produce a profitable product under present arrangements – including the cost of patents to protect them. This Act empowered the Food and Drug Administration (FDA) to fill this gap by offering to any firm that produced a relevant drug an undertaking that it would not license a competitor for seven years. This is effectively an incontestable grant of an exclusive privilege, since no drug can be put on the market without FDA approval. It is worth far more than any patent, since there can be no risk of being involved in litigation. The outcome of this legislation has been spectacularly successful, reflected in a 12-fold annual increase in new ‘orphan’ drugs, with both actual and relative declines in death rates from the diseases they treat. We can be certain, therefore, that before it, the potential for innovation of drugs of this kind was not being exploited because the arrangements for protecting the results of the risky investment needed for this – predominantly patents – were not considered to hold out enough prospect of profit. In contrast, the protection offered by the Orphan Drug Act is certain, complete and eliminates all danger of litigation costs. The benefits could not be clearer, nor could the lesson that can be drawn from them: appropriate protection results in more innovation. A period of incontestability might be offered only to smaller firms at the outset, since they could benefit most from it. The Research Fund of the European Patent Office funded research into this. When small-firm patentees with experience of infringement were asked, ‘what would it have been worth to you if the defence that your patent was invalid had not been available to your infringer?’, 75 per cent thought that it would be worth either ‘a lot’ or ‘very much’. In contrast, the overwhelming view of patent agents, presumably reflecting the conflicting interest of their large-firm clients, was against the idea of incontestability. If it were to be pursued, it would require smaller firms to agree to have their patent applications published immediately, that is, they would give up their entitlement to the 18 months’ period of grace for this under present arrangements. There would then be a short period of months during which all concerned could provide information to the Patent Office for consideration by the Examiner. The validity of patents which were issued after consideration of this and any other information available could then not be questioned for a specified period. There seems to be no reason why, if this proved successful in operation, it might not then be possible to grant an automatic first period of incontestability to all patents as from the date of grant, reflecting a basic level of presumption of

22


validity, and a second, longer one, once the outcome of some form of opposition procedure had strengthened this presumption. This would contribute to solving the disincentive effect of patents on innovation by removing the fear of litigation.

XII.

PROTECTING INNOVATION DIRECTLY

The way in which the patent system evolved, especially in terms of the cost of settling disputes, rendered it of relatively little use to smaller firms, and the spread of ‘harmonisation’, especially in its extreme form of TRIPs, rendered it positively disadvantageous to smaller countries. Some improvements which could be made, even within the TRIPs strait-jacket, have been suggested above, but by far the best potential is for quite new arrangements which would protect innovation directly. Invention is finding new things and innovation is getting new things done. At present, innovation is protected only indirectly, through whatever protection its related invention is able to obtain. It therefore depends upon the strength of the invention–innovation link. This explains why pharmaceutical inventions are well protected by the system: in this case the link between invention and innovation is very close, because what is discovered in the laboratory, what is described in the patent specification, what is circumscribed by the claims of the issued patent, what is made in quantity in the factory and what is ultimately prescribed by the medical practitioner, dispensed by the pharmacist and administered to the patient, are all absolutely identical. In contrast, for example in the case of engineering innovations, the actual link with the related invention is nothing like so close, as there will almost certainly be a number of incremental changes between the start of development work on such an invention and the time a product is finally put on the market. Since the investment required for innovation is on a far bigger scale than that for invention, it has correspondingly more need for protection. There is no reason why it could not be protected directly, through arrangements which would operate alongside the existing patent system and make up for several of its shortcomings. As the English Court of Appeal judge, Lord Jacob, put it in a recent case, One can, of course, postulate a different policy under which a [patent] monopoly might make sense. There are old or obvious ideas which take a lot of work, expense and time to develop and turn into something practical and successful. Without the incentive of a monopoly, people may not do that work or spend the time and money. The Fosamax case, Teva v Gentili [2003] EWHC 5 (Patent), [2003] EWCA Civ 1545, is an example of an obvious invention which cost lots to bring to market. But


23

patent law provided no protection for all that investment because the basic invention was obvious. The courts’ job is not, however, to uphold any claim to a monopoly for an idea which requires investment and risk to bring to market, only those for ideas which are new, non-obvious and enabled.6

XIII.

INNOVATION PATENTS AND WARRANTS

Two ways of achieving what the judge postulates have in fact been advanced. These are Herman Kronz’s Innovation Patent and the writer’s slightly different Innovation Warrant. The EU commissioned and published a study in which a number of world experts evaluated both approaches (see Kingston 1987). Although this was done before evidence of the success of the US Orphan Drugs Act of 1983 was published, that evidence is strong confirmation of the direct protection approach. The orphan drug protection is given, not for a concept of a new drug, or even for one that has reached the stage of clinical trials, but only for the actual drug, developed, fully tested and ready to go on the market. The EU’s Directive for the protection of databases is also a form of direct protection of innovation. Both versions evaluated in the EU study shared a number of features. The arrangements would be administered by an independent Innovation Authority, although a high value is set on eliminating official discretion. The subject matter of protection would be innovation, not invention. Protection would be offered before investment has to be made, but actual investment would be the condition of activating it and keeping it in force. Any economic object could be protected, not just technology. The novelty criterion would be ‘nonavailability in the ordinary course of trade’. The term of the protection would be variable, and examination would rely heavily on Third Party involvement – an anticipation of the open review concept. All grants would be incontestable except in cases where they had been obtained by fraud, and there would be no renewal fees to keep them in force. Terms of grant could differ between regions of a country (or could apply over more than one country by agreement).

XIV.

MEASUREMENT BY MONEY

A feature of these proposals which deserves special mention is that measurement of the protection granted would be through money, not time, since this would also be a much better way of dealing with classical intellectual property grants, if this could be reconciled with TRIPs. A time measure of exclusive rights can never be anything more than a surrogate for money. If all inputs and outputs could be measured accurately, then the logic of any grant of an exclusive right

24


would have it last until an investor in R&D had received as profit a socially acceptable multiple of the investment made. This multiple would take account of its subjectively assessed risk, and would in fact be the reciprocal of that risk. With a lower multiple, the risk would be regarded as too high and the investment would not be made, thus possibly depriving the public of something new and useful; with a higher multiple, the protection – and consequently the private benefit – would be more than it needs to be, and the public benefit (in terms of lower prices and/or improved products from competitors) would be correspondingly less. When the practicalities of this were investigated, however, it soon emerged that it would be very difficult to devise a tamper-proof system. In contrast to any accounting measurement after the event, focusing on the investment which had to be made beforehand to bring about an invention or innovation is much more promising. It is even more so if it is combined with compulsory licensing, so that an exclusive right would change from that of ‘making, using and selling’ to that of conditionally allowing others to ‘make, use and sell’ (Kingston 1994). The concept of the multiple in the ideal (but impractical) approach could then be applied instead to defining how much should be paid for a compulsory licence. Access to an invention by competitors is a most important point: inventions only become useful to the public (and consequently profitable) by incremental improvements, and no originating firm can devote its resources optimally to more than a single ‘trajectory’ of these. During whatever period the originator’s intellectual property protection of the present kind remains effective, other firms that are capable of developing the invention along alternative trajectories of incremental innovation may be prevented from doing so. Boulton and Watt’s patent postponed the development of high pressure steam and that of the Wright Brothers slowed down aircraft design. It is only through these alternative trajectories of relatively small technical changes that the widest range of user and consumer wants can be satisfied; that dynamic new firms are able to get a start in new technology; and – in the long run – that prices of the standardised products that eventually emerge are brought down through competition. But if competitors are to be allowed access to the results of an originator’s investment in R&D, the system must ensure that they pay appropriately for it. This means making a second- or later-comer share retrospectively in the investment which had brought the information they want to use into being. The amount payable would have to be weighted by the risk the originator had taken. That weighting should also reflect the reality that the very first money that is put behind an idea involves uncertainty rather than risk, and is to that extent irrational. No matter what the advantages of compulsory licensing, it would be essen-


25

tial that it did nothing to reduce the incentive to undertake the high risk of investment in invention and innovation. A safeguard for this is that the more important any information is seen to be by competitors, the more licences will be requested for it, and as each licence would earn the same amount, the originator could find that his risky investment was very well rewarded. This reward might be much greater than could have been achieved under traditional protection, because several trajectories of incremental development would be exploited simultaneously, which is the best possible way of expanding the total market for the originator to exploit. At the same time, no firm would be prevented from developing any new market as long as it was ready to share retrospectively in both the investment and the risk which had made that market possible.

XV.

CAPITAL PAYMENTS INSTEAD OF ROYALTIES

A further safeguard for originators would be if payment for licences was made through one-off capital amounts, rather than by royalties. This would reinforce the principle of having the second- or later-comer share retrospectively in both the amount and the risk of the investment which has been made. Any such investment is now a sunk cost for the originator, and there is no reason why any sharing of the result should be dependent on a licensee’s future success in using it, as would be the case with royalty payments. Also, a licensee who has made a capital payment for access to information is more likely to be motivated to get the best possible value out of the information which this licence has bought, by doing his own development work on it. This motivation is not necessarily present to the same extent when payment for the licence is contingent, that is, through royalties. Firms which can license information on a royalty basis may be content simply to produce clones of the originator’s product. The public interest is best served when the firms which obtain compulsory licences will themselves do R&D to exploit independent trajectories of incremental innovation. Having invested a significant capital sum for the licence adds greatly to the incentive for this. It should be stressed that the multiple would only set the price at which the originator of information would have to grant a licence for its use by another. The proposed arrangements would not prevent any type or number of licence agreements between mutually willing buyers and sellers. Any licence which would be granted under the present system, therefore, would equally be available under the new one. The public interest requires that inventions should be made, then innovated, and then diffused as quickly and as widely as possible. Wide diffusion implies the possibility of several trajectories of incremental change. Unfortunately,

26


much innovation diffusion at present is the result of failure of the patent system to deliver the protection it promises – especially in non-chemical fields. It would clearly be much better to use the proposed new arrangements to get the process of diffusion started in an orderly way and as early as possible.

XVI.

CONCLUSION

The various proposals outlined above could be useful in enabling patents to perform the only function which justifies their existence. This is that they deliver the market power necessary for rational investment in innovation to those who do not have this market power in other ways, such as through largescale investment in productive assets (for capability) or in advertising and marketing (for persuasive market power). Just as the US began to by-pass TRIPs when it discovered that it was not working as much to its advantage as had been hoped, smaller countries, all of which are disadvantaged by TRIPs, could work around its provisions in their own interest by adopting some of these proposals. Specifically, direct protection of innovation could be put into effect in parallel with the existing patent system and without being affected by TRIPs at all. It could be introduced without being applicable to chemical inventions, which are already well served. All smaller countries have suffered from ‘harmonisation’ and need to develop some countervailing force in intellectual property to that of the ‘trilateral’, the arrangement between the Patent Offices of the US and Japan and the European Patent Office, which arranges settlement matters in the interests of their largest users, the multinational firms, especially those in the pharmaceutical industry. The way in which the international patent system has evolved is proof positive of Mancur Olson’s theory of collective action (Olson 1965). This is that small cohesive groups have great advantages in getting laws that suit them, over large, diverse ones. In the patent case, the small number of large firms, especially multinational firms, has been able to shape the patent system in their own interest, to the disadvantage of smaller firms, which are greater in number but which do not act in unison. What is true of small firms is also true of smaller countries. The history of TRIPs makes it clear that their performance when it was being formulated and enacted was totally passive. It appears that they have never thought of taking advantage of Article 19 of the Paris Convention to improve the capacity of their firms to innovate. This Article is: It is understood that the countries of the Union reserve the right to make separately


27

between themselves special agreements for the protection of industrial property, in so far as these agreements do not contravene the provisions of this Convention.

Taking advantage of this would enable any group of member countries to agree to mutual arrangements for operating any or all of the suggestions for repair of the patent system outlined above, and others, as long as they also allowed any other country which was a member of the Paris Convention to share in the benefits. Just as Doha marked the beginning of the necessary reaction against TRIPs, such a group could begin a useful reaction against harmonisation.

NOTES 1. 2. 3. 4. 5. 6.

Transmitting the Report of the National Patent Planning Commission, 18 June 1943. Judge Rich, quoted by P.J. Federico in Witherspoon (1978). Drahos and Braithwaite (2002); Sell (2003). Calvert and Sofocleous (1989, 1992, 1995). See: http://www.peertopatent.org/. Angiotech Pharm., Inc. v Conor MedSystems Inc., [2007] EWCA Civ 5, 50 (Court of Appeal 2007 (Jacob, L.J.), aff’g, [2006] EWHC 260 (Pat) (Pumfrey, J.) (High Court 2006).

REFERENCES Bessen, J. and M. Meurer (2008), Patent Failure, Princeton, NJ: Princeton University Press. Calvert, I.A. and M. Sofocleous (1989), ‘Interference Statistics for Fiscal Years 1986 to 1988’, Journal of the Patent and Trademark Office Society, 71, 399–410. Calvert, I.A. and M. Sofocleous (1992), ‘Interference Statistics for Fiscal Years 1989 to 1991’, Journal of the Patent and Trademark Office Society, 74, 822–6. Calvert, I.A. and M. Sofocleous (1995), ‘Interference Statistics for Fiscal Years 1992 to 1994’, Journal of the Patent and Trademark Office Society, 77, 417–22. Cohen, W., A. Goto, A. Nagata, R. Nelson and J. Walsh (2002), ‘Research and Development Spillovers, Patents and the Incentive to Innovate in Japan and the United States’, Research Policy, 30, 1349–67. Drahos, P. and J. Braithwaite (2002), Information Feudalism: Who Owns the Knowledge Economy?, New York: The New Press. Gowers, Andrew (2006), Gowers Review of Intellectual Property, Norwich, UK: Her Majesty’s Stationery Office. Keynes, M. (1936), The General Theory of Employment, Interest and Money, London: Macmillan. Kingston, W. (ed.) (1987), Direct Protection of Innovation, Dordrecht/Boston: Kluwer Academic Publishers. Kingston, W. (1994), ‘Compulsory Licensing with Capital Payments as an Alternative to Monopoly Grants for Intellectual Property’, Research Policy, 23, 661–72. Murmann, J. (2003), Knowledge and Competitive Advantage, Cambridge: Cambridge University Press.

28


Nordhaus, W. (1969), Invention, Growth and Welfare, Cambridge, MA: MIT Press. Olson, M. (1965), The Logic of Collective Action: Public Goods and the Theory of Groups, Cambridge, Mass: Harvard University Press. Reichman, J. (1998), ‘Securing Compliance with the TRIPs Agreement after US v India’, New York University Journal of International Economic Law, 1, 586–601. Reitzig, M., J. Henkel and C. Heath (2007), ‘On Sharks, Trolls, and their Patent Prey: Unrealistic Damages Awards and Firms’ Strategies of “Being Infringed”’, Research Policy, 36, 134–54. Sell, S. (2003), Private Power, Public Law: The Globalization of Intellectual Property, New York: Cambridge University Press. Weiss, L., J. Thurbon and J. Mathews (2004), How to Kill a Country: Australia’s Devastating Trade Deal with the United States, Sydney, New South Wales: Allen and Unwin. Witherspoon, John F. (ed.) (1978), Non-obviousness – the Ultimate Condition of Patentability, Washington, D.C.: Bureau of International Affairs, Sections 1:10, 11.

3. What are the costs and benefits of patent systems? Hazel V.J. Moir Patent systems are one of the oldest policies to promote innovation. So it is surprising how little factual information is available about their economic costs and benefits. The data that are available seem to be regularly ignored in patent policy discussions (Mazzoleni and Nelson 1998). Macdonald (2004) suggests this imperviousness to fact shows that the idea that innovation will not occur without patents has achieved the status of myth. The objective of this chapter is to set out the range of benefits and costs of a patent system,1 identify those elements most critical to sound patent policy and review the evidence available so far, so as to set an agenda to establish an evidence basis for sound patent policy. A sound patent system is defined as one where social benefits exceed social costs, and the system therefore improves a nation’s economic well-being.

I.

INTRODUCTION

Patent policy is based on a conundrum: designed to increase innovation, it operates by initially suppressing the dissemination of new patented technologies. Balance is therefore central to patent policy. The benefits deriving from any induced higher level of innovation must offset, at least at the level of society, the costs due to the grant of monopoly privileges. As yet, no country appears to have undertaken a cost–benefit analysis of their patent system. The earliest, most comprehensive approach remains Machlup’s 1958 report to the US Congress, but needed data are still missing. The sole economist on the 1984 Australian patent review tabled a dissenting statement saying that there was nothing economic about the supposedly economic review (Industrial Property Advisory Committee (IPAC) 1984, pp. 79–80). Since then, a significant body of academic empirical research has been undertaken concerning the role patents play in protecting returns to industrial innovation (detailed below) and this research clearly shows that the benefits of patents to innovation have been greatly over-stated. Nonetheless, in 1994, while the Bureau of Industry Economics (BIE) did make references 29

30


to the earliest instances of this evidence, it was to place greater weight on econometric estimates of high social returns to R&D (BIE 1994a). So too the 2000 Intellectual Property and Competition Review Committee (IPCRC) review referred only briefly to the empirical evidence and it maintained a strong faith in the absolute need for patents to encourage innovation (IPCRC 2000). Both reports relied heavily on assumptions, including the requirement of a high standard of inventiveness for patent grants. To the extent that costs and benefits have been measured, the major focus has been on private benefits and private costs. Yet the most recent study shows that if pharmaceutical companies are excluded, the net private return from patents to publicly listed US firms is substantially negative (Bessen and Meurer 2008). This chapter is concerned with social costs and social benefits. Private costs and benefits form part of this assessment too.2 But patent systems involve significant positive and negative externalities and these externalities need to be added to private costs and benefits in order to estimate the net return to society. Anticipated positive externalities are the underlying rationale for the patent protection intervention in the economy. Patents are expected to induce more innovation, generating increased consumer surplus, spillovers of knowledge, enhanced productivity and higher economic growth. The negative externalities are the losses due to the exercise of monopoly power (lower output, less competition and reduced consumer surplus). Because knowledge develops in a cumulative fashion, there may also be negative externalities in the form of lower levels of subsequent innovation. Finally, any form of market regulation gives rise to direct transaction and rent-seeking costs and indirect costs due to misallocation of resources. A starting point for identifying the costs and benefits of patent systems is those areas of economic activity to which the system is directed, or through which it works. Thus, impacts on innovation, competition and resource allocation become priority categories, within which benefits and costs should be identified. Innovation impacts need to include any benefit from improved technology information dissemination. Naturally any regulatory system has associated transaction costs. For example, public choice theory directs us to pay attention to lobbying costs as an important element of the costs associated with regulatory intervention. Nonetheless, these headings do not form a clear framework which can be used as a classification system. For example, if a patent system induces more innovation, this affects resource allocation and, through the grant of the patent monopoly, competition. Under which category should induced innovation then be included? The ‘headings’ are useful in ensuring as complete as possible an accounting of the full range of economic impacts but they do not provide a good classification system.

Costs and benefits of patent systems

31

Implicit in any measurement of costs or benefits is the question, relative to what? Given the substantial evidence that in most industries market mechanisms provide an adequate return to innovation investment, the comparison chosen here is compared with a situation where there is no patent system. This could be refined to a comparison with patent system variants where, for example, patents were restricted to technology-based inventions, where there was a genuine inventive step of a reasonable height, where clear boundaries to claims were a condition for grant, and/or where scope of a grant was limited to the knowledge actually contributed. Where it is clear that such changes to patent policy would make a difference to a particular cost or benefit, this will be noted. Naturally any effort to measure as complex a phenomenon as innovation is fraught with hazards. A full discussion of these measurement problems would leave room for little else. Carter (1996) provides useful discussion of some specific measurement difficulties.

II.

INDUCED PATENTS AND USED PATENTS

From a policy perspective there are two key absolutely critical measures. First, any policy is effective if it induces the desired change: here, more innovation. Do patents induce innovation which would not otherwise occur? It is only from induced innovations that the potential benefits of a patent system flow. Where innovation would have taken place, absent patents, no benefits from that innovation can be attributed to patent policy. This remarkably central policy issue has received little comment since Machlup (1958). Innovations that are patented, but not induced by the patent system, create no benefits attributable to patent policy but they do create costs attributable to patent policy. It is not possible to quantify the economic impact of a patent system without first estimating the proportion of patents induced by it. The second key policy measure is the proportion of patents used. Some analysts assert that because of close substitutes, the monopoly losses from patent policy are small (e.g. Gans, Williams and Briggs 2004). This seems counter-intuitive – if a monopoly is needed to gain a return to innovation investment, then it will not deliver a return unless it is used. Certainly the degree of monopoly power exercised in the use of any individual patent can vary widely. Some authors assert that only a very small proportion of patents are actually used, but provide no supporting evidence (Lemley 2001; Blonder 2005). The one major empirical study on patent use found that 50 per cent of Canadian patents were worked in at least one country. Firestone also reported on the proportion of patents worked in the UK (30 per cent) and the USA (49 per cent for large companies and 71 per cent for small companies) (Firestone 1971, pp. 148–9).3

32


This issue is complicated by the question of what manner of use. Use was once simply thought of as local working, but with the massive growth in the volume of patents and strategic uses of patents, other uses need to be considered. Each type of use may confer different types of cost on other parties, including intermediate firms. Even patents which are renewed but not used can increase costs of boundary search for other firms (Bessen and Meurer 2008). Because of defensive patenting, patents may have value even where they are used only as insurance. Patent renewal data thus reveal more about private value than about use. Information on the proportions of patents used in different ways is also central to measuring the economic impact of patent systems. There are some known facts relevant to patent use. First, the proportion of non-use is likely to be higher among domestic than foreign patentees. The majority of patentees first patent domestically (lower patent costs and a more familiar system).4 For some domestic applicants, the very grant of the patent might fulfil the innovator’s prestige or marketing objectives and the patent may not be used or renewed (Bartow 2000). Second, certain patentees are known to take out very large numbers of patents for strategic reasons. These include (particularly large) firms in the industries well known for strategic patenting – particularly the electrical and computing industries (Bessen and Hunt 2004).5 A small number of firms obtain the bulk of patents: one-third of grants to organisations in each of the USA and Australia go to just 100 firms (Moir 2008). Many of the patents owned by such firms may not be used in the sense of being ‘worked’. They are however used for strategic purposes, including misleading competitors about the key directions of research, making significant patents harder to find, challenging other firms in cross-licensing negotiations and preventing entry to a market. There is now substantial evidence that the proportion of patented innovations induced by the patent system is low. This possibility was first documented by Scherer et al. (1959), confirmed by Taylor and Silberston (1973) and further documented by Mansfield (1986). The large-scale Yale survey in the early 1980s and the 1994 Carnegie-Mellon survey (CMS) confirmed that, in most industries, imitation time and cost were substantial, patents were poorly regarded as a mechanism for obtaining a return to R&D investment, and most innovations would occur, absent patents (Levin et al. 1987; Cohen, Nelson and Walsh 2000).6 The principal findings, particularly the low frequency with which patents are used as a means of ensuring a return to R&D investment, have since been replicated in national innovation surveys in many European countries (Mairesse and Mohnen 2003; Crowley 2004; Larsson 2004; Robson and Ortmans 2006), in Canada (Hanel 2008) and in Australia (Australian Bureau of Statistics 2007). These very large surveys include many small R&D performing firms. Their results confirm that there is no general failure in the market for inno-


33

vation. First mover advantages, market segmentation (including product differentiation, branding and reputation), speed of technological change, complementary sales, service and manufacturing capabilities, knowledge imperfections, transaction costs, learning and switching costs and network effects, all operate to ensure that firms are both motivated to innovate and gain good returns from their R&D investments. There is a significant exception – the pharmaceutical and fine chemicals industries. These are industries where patent boundaries are clear, where the time and cost of imitation are lower than average, and where industry leaders report that innovations would not take place absent patents.7 It makes sense to look separately at the economic consequences of patents for such clear and highly codified technologies. Bessen and Meurer (2008) separate out the pharmaceutical industries and find a large private net benefit for such firms, in contrast to the negative outcome for firms in other industries. Even here, however, it is not clear that patents are essential. One of the few natural experiments in patent policy was the extension of patents to chemical products in Italy in 1978. While this led to an increase in patenting, there was no consequent increase in R&D expenditure, nor any increased focus on new chemical entities (Scherer and Weisburst 1995). There will always be individual exceptions to these general findings. Such selective anecdotes are frequently used to illustrate the benefits or costs of patent systems. This is not a sound basis for policy design. If there are special circumstances in certain technologies – and there is substantial evidence that there are – then tailored programmes should be designed if these would deliver a net welfare gain. But the fact that a few companies may benefit from the patent system as currently designed is not a rationale for maintaining it if the overall impact is negative (Edwards 1949). The empirical surveys referred to above can be used as a basis for estimating the proportion of patents induced by patent systems. Unfortunately, industry and patent classifications do not cross-map well. Nor do the later surveys provide estimates of the proportion of innovations that depended on the provision of patents. However, the CMS found that 63 per cent of process patents and 74 per cent of product patents were acquired for defensive purposes (Cohen, Nelson and Walsh 2000, p. 18) and that they may well not be needed if there were no patent protection available. So we might estimate roughly that around 37 per cent of process and 26 per cent of product patents would cover innovations that depended on the provision of patents in order to be induced.8 Thus, estimating induced patents (in industries other than pharmaceuticals/ chemicals) at 25 to 50 per cent will provide an upper estimate of induced patents. In small nations, where the vast majority of patents are granted to foreigners, one needs to consider whether there is likely to be an inducement effect

34

Table 3.1


Estimated induced patents, Australia and USA Granted Proportion Estimated patents induced induced (%) patents

Australia, 1990–2005 Patents granted to residents of USA, Canada, Europe and Japan Patents to residents of Australia, New Zealand and other countries: Pharmaceutical and fine chemical patents All other classes Total granted patents Estimated maximum induced Estimated minimum induced USA, 2001–2005 Pharmaceutical and fine chemical patents All other classes Total granted patents Estimated maximum induced Estimated minimum induced

170 610

0

23 635

(total)

4349 19 286

100 50 25

Patents not induced

0 170 610

4349 9643 4822

0 9643 14 465

194 245 13 992 9171

109 362 701 125

100 50 25

0 109 362 350 563 350 563 175 281 525 844

810 487 459 925 284 643

Notes: Australian data are from http://www.ipaustralia.gov.au/about/statistics.shtml (accessed 5 March 2007), Table P31 (Mar06) Countries of Origin of Granted and Certified Patents, and Table P63 (May06) Grants of Standard Patents by Technology (‘pharmaceuticals’ and ‘fine organic chemicals’). US data are for utility patents, from http://www.uspto.gov/go/taf/reports.htm (by patented technology), June 2006 version (accessed 5 March 2007). Estimates for pharmaceuticals and fine chemicals are based on utility patents granted in the Life and Agricultural Sciences and Testing Method and the Compositions and Synthetic Resins; Chemical Compounds classes.

specifically for overseas innovators. In Australia, for example, 88 per cent of patents are granted to innovators from Japan, North America and Europe. It is unlikely that the possibility of gaining an Australian patent would make a difference to innovation investment for most innovators in these large markets, so it is assumed that none of these innovations are induced by the Australian patent system. For innovators from smaller markets, including New Zealand and a range of other countries, the possibility of an Australian patent grant might be critical to the investment. Based on this reasoning, it can be estimated that between 1990 and 2005, between 9171 and 13 992 innovations were induced by the Australian patent


35

Table 3.2 Patent system costs and benefits: importance and empirical evidence Type of effect

Importance

Existing empirical evidence?

Impacts on innovation Induced innovation

critical

substantial; can use appropriability studies to estimate non-existent anecdotal

Earlier timing Inventive resources shifted to patent work Negative impacts on other innovators: no copying inventing around complete blocking fences and thickets

marginal unknown/moderate

Technology market

unknown

Licensing costs and benefits Information dissemination

unknown

Secrecy

minor

Competition and resource allocation Used patents Dynamic efficiency gains from induced innovations: knowledge spillovers productivity improvements Increased foreign direct investment Static efficiency losses Slower diffusion of patented innovations Impact on entry of firms Patent races Defensive patenting Waste of non-invention resources

moderate? moderate? low risk/large impact

limited

critical important

marginal important unknown possibly important important small? frequent but modest unknown

almost none almost none some historical evidence anecdotal except for 19th century scattered evidence suggests zero value, with possible exception of large firms single study

almost none some indirect estimates of knowledge spillovers; little on productivityenhancing share of induced innovation evidence suggests insignificant almost none almost none several studies with varying findings scattered could estimate from appropriability studies almost none

36

Table 3.2


continued

Type of effect

Transaction costs Applying for and renewing patents Establishing patent boundaries/noise Monitoring and enforcing patents Lobbying Abuses of the system

Importance

Existing empirical evidence?

moderate

a few studies

possibly very significant

cost impact on search not documented some useful litigation studies; one on total cost impact very scattered

important small but central unknown

system (see Table 3.1). Thus, we can say that between 180 000 and 185 000 (92.3 to 95.3 per cent) of granted patents were for innovations that would have occurred in the absence of the Australian patent system. In the much larger US market, a larger share of patents is induced. The assumptions lead to an estimate of between 284 643 and 459 925 induced innovations, or 35 to 57 per cent of granted patents. This generous estimate of induced innovation still leaves a minimum of 43 per cent of patents granted to innovations which would have occurred in the absence of a patent system. The remainder of this chapter reviews, very briefly, the evidence regarding the range of costs and benefits associated with patent systems. The discussion is summarised in Table 3.2.

III. IMPACTS ON INNOVATION The extent of induced innovations is critical to estimating the major social benefit from patents. If the embodied new knowledge spills over to other firms and industries, it may create positive social benefits. These knowledge flows translate into productivity improvements and subsequent innovation in the recipient firms. There is remarkably little direct evidence on such social benefits.9 Mansfield’s seminal work (Mansfield et al. 1977) measured gross consumer surpluses generated by 17 innovations. One had negative social returns. Of the remainder, social returns exceeded private returns for 12, but half of these had private returns of over 25 per cent, so would have proceeded anyway. The BIE’s in-depth study of social returns to 16 Australian industrial innovations attempted to trace the likelihood and magnitude of both consumer


37

surplus benefits and knowledge spillovers. Fourteen of the 16 innovations were intermediate products. Knowledge spillovers were found to be low or very low for 13 of the innovations, and consumer surplus increases were low for six and moderate for another three. General community spillovers (improved quality of life or environmental impacts) were low or non-existent for 14 of the innovations (BIE 1994b). While such positive externalities form the most important likely benefit of patent systems, they are clearly difficult to measure and may be more tenuous than often supposed. They can be counted as a benefit of patent systems only if they flow from induced innovations. The higher the inventive step requirement, the more likely that patent innovations contain new knowledge and thus generate spillover benefits. It is well known that most of the private and social value of patents lies in a very small proportion of them (Griliches 1990; Sena 2004). But for innovations generally, there is no direct or systematic correlation between private and social returns.10 The small literature on patent renewal rates focuses largely on estimating the private value of patents. Provided renewal costs are more than symbolic,11 it could be assumed that firms will only renew patents that remain of real value to their business.12 In the mid-1990s Mischlewski and Dormer estimated that in Australia only about 10 per cent of patents ran the then full life of 16 years (Mischlewski and Dormer 1995). Others have used the idea that those patents which are renewed to the statutory limit are the most valuable and found that a very small proportion of patents contribute a large share of the value (see, for example, Harhoff, Scherer and Vopel 1998). Do private returns necessarily tally with positive social externalities? Some innovations are productivity-enhancing. While theoretical treatments of such innovations are frequent, there appear to be no empirical estimates of the proportion of innovations which have this effect. Such innovations – particularly where they are general purpose technologies like electric power or computers – can have major positive impacts on economic growth. But major positive benefits do not occur until the innovation has been widely adopted (David 1990). There are no estimates of the proportion of induced patents which have major or minor impacts in improving productivity. It is to be expected that new patented technologies will diffuse slowly, due to the exercise of the granted monopoly power to raise price and restrict output. National innovation surveys show that the bulk of innovative activity in high income countries is innovation that is new to the industry or the firm rather than new to the country or the world. For example, the 2004–05 Australian Innovation Survey found that 82 per cent of product or process innovations fell into these categories (Australian Bureau of Statistics 2007, p. 12). It is this spread of technology through the economy that provides the major boost to national productivity levels.

38


Another suggested benefit from patent systems is bringing innovations forward in time. There seems to be no documented evidence of this, even anecdotal, so this benefit, if it exists at all, seems likely to be small. Innovation resources may be shifted from unpatentable to patentable areas of research as a consequence of patent systems. Evidence from 19th century World Fairs suggests that patent systems do not affect the level of innovation but do affect the structure of the industries and the technologies in which it occurs (Moser 2005). Cole (2001) considers this a cost. He refers to research that finds, in some fields, the patentability of alternative options has a major impact in determining research directions (Rivette and Kline 2000). There is little disagreement that such an effect is a logical consequence of patent incentives, but there appears to be no systematic evidence on its magnitude. Unlike copyright, patents do not simply prevent copying. They prevent the commercial use of independently made identical inventions.13 To a large extent losses due to the prevention of copying are implicit in losses due to slower dissemination of new technology. But the loss due to reduced ability of competitors to innovate independently is important (Nelson 2006, p. 1108). Bessen and Meurer (2008) argue that most litigation involves cases of inadvertent infringement, and that the costs of determining the boundaries of patented technology are substantial. All these effects need to be assessed against the understanding that knowledge cumulates and that every innovation draws on previous knowledge, including knowledge in patented artefacts. There is no point at which an ‘initial innovator’ can be clearly identified. Acquisition of knowledge is costly, and imitation involves substantial knowledge inputs. It can also lead to improvements, and sometimes sequential incremental innovation leads to radical breakthroughs (Mandeville 1996). Given the objective of encouraging innovation, the very substantial cost to innovators of establishing the boundaries to property already granted through the patent system needs to be recognised (Bessen and Meurer 2008). This transaction cost relates directly to determining the technological space within which innovators may freely operate. If a technological space becomes crowded with large numbers of patents, the costs to other innovators can rise substantially. The higher the inventive step threshold, the less likely this crowding is to occur. Companies devote considerable resources to ‘inventing around’ patents. Plant suggested this was an important aspect of the maldistribution of resources caused by the patent system (Plant 1934, p. 46). In a remarkable piece of ‘spin’ (or ‘framing’ to use a more academic term), the costs of inventing around are now often claimed as a benefit of patent systems. Machlup took short shrift with this:


39

from merely defending the need of ‘inventing around a patent’ as a minor item of waste, the discussion has recently proceeded to eulogize it as one of the advantages of the system, indeed as one of its justifications … The production of the knowledge of how to do in a somewhat different way what we have already learned to do in a satisfactory way would hardly be given highest priority in a rational allocation of resources. (Machlup 1958, p. 51)

At its extreme, the prevention of independent invention can hold up major technology developments. Perhaps the most famous example is James Watt, whose refusal to license his patents is generally considered to have held up the Industrial Revolution by several decades (Ashton 1948; Boldrin and Levine 2008). Other famous examples are the Wright brothers’ stabilisation and steering systems for aircraft and Edison’s incandescent lamp (Cohen 2005). Examples of technological hold-ups do not need to be pervasive to have a large social cost: if ‘the technology in question is sufficiently important, only one or a few instances … may impose considerable social cost’ (Cohen 2005, p. 63). In some industries there is a very high degree of concentration in patent ownership. For example, 75 per cent of agricultural patents are held by just six companies (Phillips et al. 2004).14 Another consequence of a patent system for innovation can be efforts by an inventor to tie up all possible applications of a new idea. The use of resources to discover and claim ownership to all possible means of doing something is inefficient. The CMS shows a large majority of firms report taking out patents either to block rivals, or to improve their position in negotiating licences with other firms (Cohen, Nelson and Walsh 2000). The associated resources, including resources to develop alternative variants, are resources that would not be needed if a patent system did not exist. The net effect of such an array of techniques [patent pools and cartels, bottleneck patents, patent blitzkrieg, umbrella patents and accordion patents] is that competitors would not be able to differentiate valid and invalid patents and the cost of litigation or, alternatively, of inventing around becomes so high that they would refrain from indulging in any activity that is protected by the network of patents. (Endeshaw 1996, p. 92)

A further cost is consequent on the prevention of independent invention – so-called ‘defensive patenting’. Cohen, Nelson and Walsh (2000) found that a large majority of patents were taken out for defensive reasons (63 to 74 per cent). The greater the volume of patenting, the more is the pressure to patent defensively (Macdonald 2002, p. 144). This cost would disappear if independent invention were allowed as a defence in infringement cases (Scotchmer 1998). A recently popular argument in support of patent systems is that they encourage development of technology markets. There are two aspects to this:

40


outright sale and licensing. Patents appear to have been more often traded in the late 19th century (Lamoreaux and Sokoloff 1997; 1999) than they are now. A study of patent renewal rates notes that patent rights are seldom marketed (Pakes and Simpson 1989, p. 356). The sole empirical estimate of recent trading of patents indicates that 18 per cent of US patents held by small innovators are traded at least once in their lifetime (Serrano 2006). Given the boundary problems associated with most patents (Bessen and Meurer 2008), it is surprising that so many are traded. Licensing of patents usually involves arrangements for the purchase of associated know-how as well as the patented technology (Boehm 1967). Firms are notoriously loath to disclose information about the terms and conditions of such licences.15 Between 23 and 28 per cent of firms in the CMS indicated that a reason for obtaining patents was to earn licensing revenue (Cohen, Nelson and Walsh 2000). A large minority of patents are used in some way in a technology market.

IV.

DISSEMINATION OF TECHNOLOGY INFORMATION

A number of authors consider information dissemination to be the principal rationale for a patent system, while others consider it an important secondary benefit. Even in countries where the principal rationale is the innovation incentive, patent statutes often require clear disclosure, sometimes as an explicit condition of grant. Machlup (1958) dismissed as entirely fallacious the idea that technology disclosure is a benefit of patent systems.16 But what empirical evidence is there on the use of patents for information on new technologies? In an early 1980s survey of Australian engineers, only 14 per cent reported using patent information to track new technologies. This was partly due to legalistic language: one respondent commented, ‘I could hardly recognize my own inventions in legalese’ (Mandeville, Lamberton and Bishop 1982, p. 209). A companion survey indicated that the major use of patent data was for using the patent system. The 1984 IPAC review noted that ‘much of the technical information in patent documents is not, or is only some years later, disclosed elsewhere. Indeed, one study found that less than 10 per cent of the information contained in patents appears in other technical media’ (IPAC 1984, p. 53).17 The IPAC committee considered this might be due to poor technical content, lack of awareness by potential users and difficulties in interpreting the information. Other reasons can be added. The important ways in which information flows along supply chains and the role of customers and suppliers in the innovation process have been increasingly understood as key drivers of competi-


41

tive advantage (Porter 1990). In a 1996 study of small UK manufacturing firms, Macdonald (2003) found they looked principally to customers, suppliers and competitors for ideas. He found the two major reasons for patent searches were checking on infringements and preparing patent applications. Another frequent use was to check on competitors – but on their patenting activity not their technology. Oppenheim (2000), also looking at small UK manufacturing firms, found the major reason for non-use of patent data to be irrelevance to company needs (54 per cent), with a further 17 per cent saying such information was available from other sources.18 Patent offices regularly write about the high value of patent information, often despite survey work they have commissioned. A very large survey of European R&D performing manufacturing firms, conducted in 1994 for the European Patent Office (EPO), found that patent data were ‘rated lowest as sources of information on technical developments’ (EPO 1995a, p. 105).19 The one empirical study finding that firms do use patent data is a small survey, undertaken through the auspices of a business association, reporting that 90 per cent of large firms find patent data useful for their own R&D, and that 44 per cent had found usefulness increased over the past ten years (Sheehan, Martinez and Guellec 2003). Whether the value for their own R&D derives from purposes related to the patent system (avoiding infringement, inventing around etc.), or for purposes to do with following the latest developments in technology, is not reported. Besides the complexity of patent language, non-use of patent data as a source of technology information may be due to the very low level of inventiveness apparent in many patents. Further, built-in delays in publication can eliminate any value in fast-moving areas. Patent classes are not well designed for identifying and tracking developments in specific technologies. They often contain widely differing inventions (Desrochers 1998).20 At the same time, searching for a single type of innovation can require searching dozens of sub-classes.21 Word search is not always a good alternative as there is evidence that patentees deliberately hide critical information. Where keywords are missing, finding relevant patents can be impossible. Stallman provides a very telling example of this problem: a US patent on natural order recalculation in spreadsheets.22 He notes that the key terms ‘spreadsheet’, ‘natural order recalculation’ and ‘topological sorting’ did not appear in the specifications (Stallman 2001). The obscurity of patent titles has been highlighted (Murphy 2002) and Labich commented two decades ago: ‘A company’s patent lawyers can protect the company’s proprietary position without giving away too much in the application process’ (Labich 1988, p. 30). The empirical evidence thus suggests that the possible benefit of disclosure of detailed information about new technologies may be minimal.23 This value may have fallen recently if preliminary work indicating increased secrecy

42


about patented innovations is more widely confirmed. Grushcow measured scientists’ secrecy by the delay between presentation at a scientific meeting and formal publication of the work in a peer-reviewed journal.24 On this basis he shows that scientists seeking to patent their work withhold presentation at scientific meetings (Grushcow 2004). His data also suggest increased secrecy among university scientists, even when patents are not being sought. These findings from the US are paralleled in the UK, where Macdonald (2003) notes: ‘The British Technology Group, which specialises in exploiting university patents, is especially keen to suppress academic discussion.’ These findings suggest that it may even be necessary to assign a negative value to the impact of a patent system on the dissemination of new technology information. There is now an extremely large literature suggesting that the novelty and inventiveness criteria for issuing patents have fallen to a very low level.25 Given this strong consensus, many granted patents will not contribute any leading-edge technology information. With the very large volume of patent applications and grants, where a high proportion are minimally inventive (if at all), the costs of searching would be very high compared with any expected gain. Overall, then, it seems that this benefit is rarely evidenced in the real world. The possible exception is large firms. But it would be useful to establish if their use is for reasons to do with the patent system or as a source of technological information.

V.

COMPETITION AND RESOURCE ALLOCATION

The current dominant economic paradigm, and the perspective taken in this chapter, is that competition in markets for goods and services is most likely to lead to both an efficient allocation of resources and to maximising consumer satisfaction. An essential characteristic of a competitive market is numerous suppliers and buyers. This also implies that alternative suppliers can freely enter and exit any market. In sharp contrast to competitive markets are those that are monopolised by a single supplier. Monopolists are likely to restrict output in order to be able to charge higher prices, thus earning greater profits. Because this prevents marginal cost and marginal price equality, it leads to a poorer outcome in terms of resource allocation, reducing both overall output and consumer satisfaction. Competitiveness and resource allocation are thus, from an economic viewpoint, flip sides of the same coin. Any market situation characterised by at least a degree of monopoly power is likely to mean less than optimal resource allocation. As noted above, some consider that only a small proportion of patents are


43

actually used and that most patented innovations have close substitutes, meaning little market power can be exercised. This is to suggest that patents are largely ineffective, in that they do not prevent competitors from offering very similar innovations. It also suggests that patents are granted for very similar innovations, implying that the boundaries of granted patent property may overlap. Certainly, if a patent is not used there can be no static efficiency loss associated with it. Given use, there will also be variable degrees to which an innovator is able to charge higher prices, so the actual estimation of such losses for a patent system involves two steps, both challenging. Information on what proportion of patents is used, and in what manner, is thus critical to assessing the economic impact of patent systems. The 1984 IPAC review recommended that patentees should be required to provide details of patent use when they renew their patents. This need involve very little cost to patentees but would provide invaluable information about the impact of patent systems. A second source, which would provide complementary information, would be to use national innovation surveys to find out the proportion of innovating firms affected by patents owned by other companies. Firms could be asked whether, and how often, they have received advice that they were infringing others’ patents, and what this has cost them. If the patent system did not divert resources from other activities, it would fail in its principal purpose. The rationale for patents is that the net social return from induced activities is higher than from the abandoned activities. Is this demonstrated empirically? Because of the massive difficulties in accurately measuring such variables as the quality of education, knowledge and capital equipment, it is hard to demonstrate conclusively whether the returns to one are greater than the returns to another. All are generally agreed to be significant inputs into economic growth. But whether a country will benefit more at a particular point from more input into skilled labour formation, investment in infrastructure or capital equipment, or industrial innovation is not a question that can be answered with any certainty.26 An important variant on this idea that patents will draw more resources into inventive activity is the argument put strongly to low income countries during the Uruguay Round GATT negotiations – that stronger patent, copyright and trademark law would lead to higher levels of foreign direct investment (FDI). This idea was based on a study undertaken for the World Bank (Mansfield 1994), but this has been substantially criticised by Heald (2004). A dissertation completed at around the same time as Mansfield’s study found that at best the relationship between strong intellectual property rights and FDI was insignificant (Kondo 1994). Certainly, modern econometric studies show substantial estimated social returns to industrial expenditure on R&D. But these studies are plagued by measurement problems. Measures of the real input to improvements are likely

44


to miss much of business input into improvements (Carter 1996). Total factor productivity (frequently the dependent variable) is, of course, the famous residual in economic growth, with all the measurement problems that implies. Cole considers Griliches’ rough guesstimate – that not more than a quarter of total productivity growth is attributable to patented inventions – to be an overestimate (Cole 2001). In addition, Griliches’ estimate (Griliches 1990, p. 1699) was in respect of all patented innovations, not induced patented innovations, which we have said are a small proportion of patented innovations (see above). Boldrin and Levine (2008) compare productivity growth in US agriculture generally and corn in particular to changes in intellectual property ‘protection’ for plants, and find no evidence of any increase in productivity following the introduction of the 1970 US Plant Variety Protection Act. So the central question remains: do we really know if we get a better return by tilting the playing field in this way? A possibly more tractable, but equally important, issue is the entry and exit of firms. There has been much discussion over the role that patents play in this. Historically, there have been examples of patents both assisting in the entry of firms from outside an industry, and in entrenching dominant firms in an industry. For example, industry outsiders are thought to have played an important role in the development of railway technology in the USA in the 19th century (Merges 2003). Equally, there have been examples of the ownership of patents being used to keep competitors out of a market (see above). One empirical study has found that the fear of litigation may cause smaller entrant firms to avoid areas where incumbents are heavy users of patents (Lerner 1995). Nonetheless, ‘[d]irect evidence on the “stifling” versus “stimulating” impacts of patents on innovation and competition is not easy to find’ (Cockburn and MacGarvie 2006, p. 1). Cohen (2005) notes that this is an area where there is no systematic evidence. Despite the importance of new firm entry to the functioning of competitive markets, there is as yet insufficient data for determining whether the patent system has a positive or negative effect. In the winner-takes-all patent market, if two or more inventors develop a practical application at the same time, then only one wins the monopoly grant. Does this constitute a waste of resources? There is substantial evidence of simultaneous invention,27 so that cases where several people invent approximately the same innovation at around the same time may be the norm rather than the exception, and may be a natural phenomenon rather than one induced by patent systems. In fact, there are likely to be substantial practical variations in the resulting artefacts, and investment in R&D is an important part of the learning needed for the adoption and adaptation of new technologies (Cohen and Levinthal 1989). So ‘patent races’ may not be unambiguously positive or negative in terms of their economic impact. Recently patent races have expanded to become portfolio patent races (see, for example, Cohen 2005, p.


45

62). Such broader races might impose significantly different costs and benefits, including the substantial blocking of innovation by rival firms.

VI.

TRANSACTION/REGULATORY COSTS

The most frequently mentioned costs of patent systems are the private costs of applying for, renewing and defending patents, and the private costs incurred through patent litigation. Only four papers with systematic evidence on filing and renewal costs have been found.28 The sole academic study involved considerable estimation and was presented with an error rate estimate of +/– 37 per cent (Mooney and Oppenheim 1994a; 1994b). Nonetheless the estimates – that filing, renewal, litigation and information service costs were around 6 per cent of the value of gross trade (imports and exports) of patented goods – provide an interesting perspective on this aspect of the cost equation.29 The EPO has estimated that the cost of obtaining a ‘standard’ patent through the EPO was two to four times higher than in the USA or Japan (EPO 1995b). Van Pottelsberghe and François (2006) attempt to relate these costs to market size, but this explains little of the cost difference. A more recent EPO survey confirmed this, and also found costs to be slightly higher for small and medium-sized enterprises than for larger companies.30 Nearly 60 per cent of the surveyed companies had their own intellectual property department; onethird was unable to estimate their total expenditure on patents. One conclusion was that there is considerable variation in the cost of applying for a patent between regions, and between patents. One Patent Attorney responded that some ‘inventions are closer to the prior art than others – it is more difficult to draft applications for the closer ones and to prosecute the application through to grant’ (Roland Berger Market Research 2004, p. 43). For Australia, the Patent Office indicates a cost range of between $A6000 and $A10 000 for filing a standard application and paying relevant costs through to grant, including Patent Attorney fees. It indicates a further likely outlay of $A8000 if the patent is renewed for the full 20-year period possible.31 Data on grants and renewals could be combined with this information to generate estimates of total outlays. There is a much larger empirical literature on the costs of monitoring and enforcing patents. This focuses on patent litigation, especially its risk, validity or infringement outcomes, and impacts on firms of different sizes. The probability of a particular patent being involved in a litigation suit is significantly lower when the patent holder has a large portfolio of patents, except for pharmaceutical patents (Lanjouw and Schankerman 2001). In a further study, these authors find that individuals and firms with small patent portfolios have a

46


much higher risk of patent litigation and conclude that, ‘small patentees are at a significant disadvantage in protecting their patent rights because their greater litigation risk is not offset by more rapid resolution of their suits’ (Lanjouw and Schankerman 2004, p. 45). These findings are confirmed for Europe: 30 per cent of smaller European firms with patents granted in 1994–97 reported copying by larger firms. They also reported particular difficulty in winning infringement cases in the USA (European Union 2000). These studies focus on cases where lawsuits are filed, which may well be a minority of infringement cases. There seem to be no studies on infringement costs, where cases are settled before any lawsuit is filed; nor do there seem to be any on the cost of monitoring to identify infringements. Recent work by Bessen and Meurer (2008) provides very useful new estimates of the total private cost of patent litigation, focusing on the impact on firm market value. Costs are shown to substantially exceed (already substantial) direct legal costs. Their estimates cover US publicly traded firms and indicate that, at least for non-pharmaceutical firms, the costs due to litigation now exceed the private benefits of the patent system. They point out that litigation risk is an inevitable consequence of innovation in a world with patents, and suggest that most infringement is inadvertent. If so, an independent invention defence would eliminate much of this cost. There is now a large literature on public choice theory and regulatory capture. Public policy is most often captured when there are few well-organised beneficiaries of the intervention, with those who pay being widely dispersed. This of course typifies the patent system. The patent community, particularly patent lawyers and major users of the system, engage in extensive lobbying in their own interests.32 But patents, like tariffs, have hidden costs, spread throughout the community. One reason the economics profession is so opposed to regulation of markets (except where large net gains are clearly identifiable) is that there is an incentive to spend very large sums on lobbying. Barton provides an interesting perspective on the costs and benefits of the patent system by providing, for the USA, data on the ratio of intellectual property lawyers to $US billions spent on R&D. This ratio increased from under 45 in 1970 to about 75 in the late 1990s (Barton 2000). A problem with any form of regulation is argument and dispute over boundaries. Bessen and Meurer (2008) demonstrate vividly that most patent boundaries are so poorly drawn that third party innovators are totally unable to identify, at any remotely reasonable cost, the boundaries of technology that is owned and technology that is free to be exploited. But the very volume of patents, many of them over-lapping, written in extremely hard-to-read language and with poor classification and keyword systems, means a vast amount of noise in the system. This creates significant costs for innovating firms.


47

Commentators often allege abuse of the patent system. Extreme care is needed in this area. It is not an abuse of the patent system to charge such a high price for a new and more effective HIV/AIDS drug that it is out of reach of many who would benefit from it. This is a direct consequence of the design of the patent system. Abuses are those actions that undermine the purpose of patent legislation, and attitudes to this change over time. Machlup considered that combining trademarks with patents, to effectively extend the term of market power, was an abuse (Machlup 1958, pp. 10–11). Nowadays it is standard practice for patented pharmaceuticals to be heavily marketed under a trademark, and it is evident that this operates to extend market power beyond the patent term.

VII.

PRIORITIES

Given the importance of innovation, the possibility that poor patent policy can significantly inhibit innovation, and increasing evidence that the costs of patent systems are higher than suspected, and the benefits lower, developing a solid evidence base for patent policy is a priority. The development, since the early 1990s, of a series of comparative national innovation surveys indicates the importance of innovation. Surprisingly these surveys contain very little useful data on patents. They could easily be expanded to throw light on two critical issues: the proportion of innovations induced by patents and the extent to which patents held by one firm create costs for other innovating firms. Requiring data on use at the time of patent renewal would provide important complementary evidence on patent system costs. Other important issues for research are sounder evidence on dynamic benefits, including spillover benefits, and impacts on the entry of firms. Further information on the extent to which patent holders are able to extract monopoly rents is also needed for a fuller understanding of the competitive impact of the patent system. In the area of transaction costs, the issue of identifying patent boundaries has clearly become urgent, and research is needed on how this impacts on firms’ innovation decisions and investments. Finally, given the small number of beneficiaries of the system,33 and evidence of their substantial impact on patent policy (Drahos with Braithwaite 2002; Sell 2003), it would be of public interest to know how much they spend lobbying to tilt the patent playing field in their preferred direction. Noting the dearth of relevant data for patent policy analysis, Bakels and Hugenholtz (2002) called for the establishment of a patent observatory in Europe, so far to no avail. This is not the first call for better evidence on patents, but such recommendations do not seem to be acted on, even when they are by formal review committees. Kahin (2003) has noted several

48


instances where lobbyists have prevented research or data collection on the impact of software and business method patents. Patent policy is too important to national economic well-being to allow narrow sectional interests to prevent the collection of evidence on its impact.

NOTES 1. 2. 3. 4.

5. 6. 7. 8.

9.

10. 11.

12. 13.

14.

A wide range of possible economic impacts are identified here drawing on many authors’ writings, not all of which are identified here. Doubtless there are other economic impacts which I have missed, and which ought to be added to this accounting. But they are not the direct subject of this chapter. And these figures are consistent with an earlier US study reported by Griliches (1990, p. 1679) that 41–55 per cent of patents were used commercially, with up to 71 per cent used among small firms. Evidence from recent EPO market research on companies receiving EPO patents in 2002 and 2003 indicates that 74 per cent of companies initially filed in their country of domicile (Roland Berger Market Research 2004). This sample is biased towards more frequent patentees and larger companies. It is probable that smaller companies would be even more likely to make their initial filing domestically. For an unknown share of applications one could reasonably argue that some non-US firms might choose to first patent in the USA because of the wider subject matter scope and low novelty standards in US patent law, and because of the size and importance of the US market. Hall also identifies a set of industries with significant strategic patenting behaviour (Hall 2005). Another useful study focused solely on imitation is Mansfield, Schwartz and Wagner (1981). Though in the CMS, patents were not regarded as the most important appropriability mechanism for pharmaceuticals. They were reported as a close second to secrecy and were closely followed by lead time (Cohen, Nelson and Walsh 2000, Table 1). Mansfield (1986) specifically asked what proportion of innovations would not have been commercially introduced without the protection of patents, and reported that this per centage was 65 for pharmaceuticals, 30 for chemicals, 15 for machinery and zero for office equipment, motor vehicles, rubber and textiles. Most studies of knowledge spillovers are econometric models attempting to measure the return to R&D spending in other firms, industries or in the economy generally. These studies are reviewed in Sena (2004) and (US) CBO (2005), who reach opposing conclusions as to the likelihood and magnitude of knowledge spillovers. Nor are there any theoretical arguments to suggest why there should be any such relationship (Machlup 1958). Machlup also notes that both private and social returns will depend on how an innovation is exploited. This is actually a very questionable assumption. Users of the patent system have been very successful in persuading Patent Offices of the need to reduce the price they pay for their legislated monopolies. As a result there has been a real decline in the cost of application and renewal fees in some jurisdictions. Though Mooney and Oppenheim (1994a, pp. 154–5) suggest that some firms simply renew all patents for the full term. They suggest this may cost no more than reviewing each patent each year. Patents are thus quite unlike the ownership of tangible property. Property rights in a house prevent others from using that house. But they do nothing to prevent others from acquiring their own houses, nor making them identical. Patents provide the right to prevent anyone else making commercial use generally of a particular area of technology. There seem to be no studies on patent ownership concentration since Watson and Holman (1970).

Costs and benefits of patent systems 15. 16. 17. 18. 19. 20. 21. 22. 23.

24. 25. 26. 27.

28. 29. 30.

31. 32.

33.

49

Mooney and Oppenheim (1994a; 1994b) were unable to obtain any useful information of this and had to omit it from their survey of UK patent costs. Arguing that as there was an incentive only to patent that which could not be kept secret, the dissemination of information could not be claimed as a benefit of the patent system. This view is also put forward by Boldrin and Levine (2008, p. 186). The study referenced is Liebesny et al. (1974). Indeed, Macdonald (2003) has suggested that, for small and medium-sized enterprises, the patent system is a net demander of information rather than a net supplier. More important sources reported were: ‘talks with customers, specialised literature and trade magazines, trade fairs and talks with suppliers as their main sources of information on the latest technology’ (EPO, 1995a, p. 105). Water pistols and holy water dispensers are in the same patent class (see Griliches 1990, p. 1666). Sorensen and Stuart searched 2400 patent classes to ensure a complete identification for semiconductor product, device and design inventions (2000, p. 91). Meaning cell values are recalculated in logical order (i.e. cells that depend on others are calculated after the earlier cells are recalculated). An early quantitative estimate of the private value of patent information to firms estimated this at 0.75 per cent of R&D expenditure (Taylor and Silberston 1973, p. 212). Using data for 1992, Mooney and Oppenheim (1994b, p. 148) estimated that only about 3 per cent of outlays by British industry on patents and patent information were on patent information. These values include the value of using patent data for purposes associated with using the patent system as well as for any use for technology search in itself. With a short gap indicating that the data were shared at a scientific meeting and a long gap suggesting they were not. A small selection includes Merges (1999); Hunt (1999); Lunney (2001, 2004); Jaffe and Lerner (2004); Lemley, Lichtman and Sampat (2005–06). A good review of studies in this area is (US) CBO (2005). See, for example, Mokyr (2002). Machlup and Penrose (1950) also comment that this understanding was widespread during the Great Patent Debate of the 19th century. In an 1878 US Supreme Court decision, where dominant firms in an industry won a significant victory over patents held by small industry outsiders, Justice Bradley commented: ‘Like almost all other inventions, that of double brakes came when, in the progress of mechanical improvement, it was needed; and being sought in many minds, it is not wonderful that it was developed in different and independent forms’ (quoted in Perelman 2002, p. 19). The American Intellectual Property Law Association regularly gathers data on such costs in respect of the US market, but these data are not available in the public domain except where quoted elsewhere, for example in Bessen and Meurer (2008, p. 132). A more appropriate yardstick would of course be expenditure on research and development. Based on a sample of companies with EPO patent grants in 2002 or 2003, and discussions with Patent Attorneys and translation service firms (Roland Berger Market Research 2004). This study had a very low response rate and respondents are dominated by frequent patentees. http://www.ipaustralia.gov.au/patents/fees_index.shtml, accessed 4 July 2008. For example, ‘The penultimate Senate draft of the American Inventors Protection Act of 1999 mandated a General Accounting Office study of business method patents, but this was removed at the behest of the patent bar’ (Kahin 2003). Bessen and Meurer comment that the Federal Trade Commission ((US) FTC 2003) recommendation that was most prominently rejected by the Intellectual Property Owners’ Association was Recommendation 10: ‘expand consideration of economic learning and competition policy concerns in patent law decisionmaking’ (Bessen and Meurer 2008, pp. 293–4). Bessen (2006, p. 19) notes that a very small number of companies take nearly 80 per cent of the benefits of the US patent system in respect of publicly listed US firms.

50


REFERENCES Ashton, T.S. (1948), The Industrial Revolution, 1760–1830, Oxford: Oxford University Press. Australian Bureau of Statistics (2007), Patterns of Innovation in Australian Businesses 2005, Canberra: Australian Bureau of Statistics, Cat. No. 8163.0. Bakels, R. and P. Hugenholtz (2002), The Patentability of Computer Programs: Discussion of European-Level Legislation in the Field of Patents for Software, Study Commissioned by the European Parliament, Brussels: European Union. Barton, J. (2000), ‘Reforming the Patent System’, Science, 287, 1933–4. Bartow, A. (2000), ‘Separating Marketing Innovation from Actual Invention: A Proposal for a New, Improved, Lighter, and Better-Tasting Form of Patent Protection’, Journal of Small and Emerging Business Law, 4, 1–25. Bessen, J. (2006), Estimates of Firms’ Patent Rents from Firm Market Value, Boston, University School of Law Working Paper No. 06-14. Bessen, J. and R. Hunt (2004), An Empirical Look at Software Patents, Working Paper No. 03-17/R, Philadelphia, Federal Reserve Bank of Philadelphia. Bessen, J. and M. Meurer (2008), Patent Failure: How Judges, Bureaucrats, and Lawyers Put Innovators at Risk, Princeton and Oxford: Princeton University Press. BIE (1994a), The Economics of Patents, Occasional Paper 18, Canberra: Bureau of Industry Economics. BIE (1994b), Beyond the Innovator: Spillovers from Australian Industrial R&D, Occasional Paper 16, Canberra: Bureau of Industry Economics. Blonder, G. (2005), ‘Cutting Through the Patent Thicket’, Business Week, 20 December. Boehm, K. (1967), The British Patent System, Cambridge: Cambridge University Press. Boldrin, M. and D. Levine (2008), Against Intellectual Monopoly, Cambridge: Cambridge University Press, also at http://www.dklevine.com/general/intellectual/ againstfinal.htm. Carter, A.P. (1996), ‘Measuring the Performance of a Knowledge-Based Economy’, in D. Foray and B.-Å. Lundvall (eds), Employment and Growth in the KnowledgeBased Economy, Paris: OECD. Cockburn, I. and M. MacGarvie (2006), Entry, Exit and Patenting in the Software Industry, NBER Working Paper 12563, Cambridge, Mass: National Bureau of Economic Research. Cohen, W. (2005), ‘Patents and Appropriation: Concerns and Evidence’, Journal of Technology Transfer, 30, 57–71. Cohen, W. and D. Levinthal (1989), ‘Innovation and Learning: The Two Faces of R&D’, Economic Journal, 99, 569–96. Cohen, W., R. Nelson and J. Walsh (2000), Protecting Their Intellectual Assets: Appropriability Conditions and Why U.S. Manufacturing Firms Patent (or Not), NBER Working Paper 7552, Cambridge, Mass: National Bureau of Economic Research. Cole, J. (2001), ‘Patents and Copyrights: Do the Benefits Exceed the Costs?’, Journal of Libertarian Studies, 15, 79–105. Crowley, P. (2004), Innovation in the New Member States and Candidate Countries: Output, Barriers and Protection, Statistics in Focus: Science and Technology 13/2004, Luxembourg: Eurostat. David, P. (1990), ‘The Dynamo and the Computer: An Historical Perspective on the Modern Productivity Paradox’, American Economic Review, 80, 355–61.


51

Desrochers, P. (1998), ‘On the Abuse of Patents as Economic Indicators’, Quarterly Journal of Austrian Economics, 1, 51–74. Drahos, P. with J. Braithwaite (2002), Information Feudalism: Who Owns the Knowledge Economy?, London: Earthscan. Edwards, C. (1949), Maintaining Competition: Requisites of a Governmental Policy, New York: McGraw-Hill. Endeshaw, A. (1996), Intellectual Property Policy for Non-Industrial Countries, Aldershot, UK, Brookfield, US, Singapore and Sydney: Dartmouth. EPO (1995a), ‘Utilization of Patent Protection in Europe’, World Patent Information, 17(2), 100–105. EPO (1995b), Cost of Patenting in Europe: Results of Surveys Carried Out by the European Patent Office, Munich: European Patent Office. European Union (2000), Enforcing Small Firms’ Patent Rights, CORDIS/Innovation Policy Studies, ftp://ftp.cordis.europa.eu/pub/innovation-policy/studies/studies_ enforcing_firms_patent_rights.pdf, 6 Nov 2006. Firestone, O. (1971), Economic Implications of Patents, Ottawa: University of Ottawa Press. Gans, J., P. Williams and D. Briggs (2004), ‘Intellectual Property Rights: A Grant of Monopoly or an Aid to Competition?’, Australian Economic Review, 37, 436–45. Griliches, Z. (1990), ‘Patent Statistics as Economic Indicators: A Survey’, Journal of Economic Literature, 28, 1661–1707. Grushcow, J. (2004), ‘Measuring Secrecy: A Cost of the Patent System Revealed’, Journal of Legal Studies, 33, 59–84. Hall, B. (2005), ‘Exploring the Patent Explosion’, Journal of Technology Transfer, 30, 35–48. Hanel, P. (2008), ‘The Use of Intellectual Property Rights and Innovation by Manufacturing Firms in Canada’, Economics of Invention and New Technology, 17, 285–309. Harhoff, D., F. Scherer and K. Vopel (1998), Exploring the Tail of Patent Value Distributions, Mannheim: Centre for European Economic Research. Heald, P. (2004), ‘A Skeptical Look at Mansfield’s Famous 1994 Survey’, Information Economics and Policy, 16, 57–65. Hunt, R. (1999), Nonobviousness and the Incentive to Innovate: An Economic Analysis of Intellectual Property Reform, Working Paper 99-3, Philadelphia: Federal Reserve Bank of Philadelphia. IPAC (1984), Patents, Innovation and Competition in Australia, Canberra: Industrial Property Advisory Committee. IPCRC (2000), Review of Intellectual Property Legislation Under the Competition Principles Agreement: Final Report, Canberra: Commonwealth of Australia. Jaffe, A. and J. Lerner (2004), Innovation and its Discontents: How Our Broken Patents System is Endangering Innovation and Progress, and What to Do About It, Princeton: Princeton University Press. Kahin, B. (2003), ‘The Expansion of the Patent System: Politics and Political Economy’, First Monday, 6, www.firstmonday.org/issues/issue8_3/kahin/index. html. Kondo, E. (1994), ‘Patent Laws and Foreign Direct Investment: An Empirical Investigation’, Harvard University, PhD dissertation. Labich, K. (1988), ‘The Innovators’, Fortune, 6 June, 27–32. Lamoreaux, N. and K. Sokoloff (1997), Inventors, Firms, and the Market for Technology: U.S. Manufacturing in the Late Nineteenth and Early Twentieth

52


Centuries, NBER Working Paper H0098, Cambridge, Mass: National Bureau of Economic Research. Lamoreaux, N. and K. Sokoloff (1999), Inventive Activity and the Market for Technology in the United States, 1840–1920, NBER Working Paper 7107, Cambridge, Mass: National Bureau of Economic Research. Lanjouw, J. and M. Schankerman (2001), Enforcing Intellectual Property, NBER Working Paper 8656, Cambridge, Mass: National Bureau of Economic Research. Lanjouw, J. and M. Schankerman (2004), ‘Protecting Intellectual Property Rights: Are Small Firms Handicapped?’, Journal of Law and Economics, 47, 45–74. Larsson, A. (2004), Innovation Output and Barriers to Innovation, Statistics in Focus: Science and Technology 1/2004, Luxembourg: Eurostat. Lemley, M. (2001), ‘Rational Ignorance at the Patent Office’, Northwestern University Law Review, 95, 1495–1532. Lemley, M., D. Lichtman and B. Sampat (2005–06), ‘What to Do about Bad Patents?’, Regulation, 28, 10–13. Lerner, J. (1995), ‘Patenting in the Shadow of Competitors’, Journal of Law and Economics, 38, 463–95. Levin, R., A. Klevorick, R. Nelson and S. Winter (1987), ‘Appropriating the Returns from Industrial Research and Development’, Brookings Papers on Economic Activity, Special Issue on Microeconomics, 783–831. Liebesny, F., J. Hewitt, P. Hunter and M. Hannah (1974), ‘The Scientific and Technical Information Contained in Patent Specifications: The Extent and Time Factors of its Publication in Other Forms of Literature’, Information Scientist, 8(3), 165–77. Lunney, G. Jr (2001), ‘E-Obviousness’, Michigan Telecommunications Technology Law Review, 7, 363–422. Lunney, G. Jr (2004), ‘Patent Law, the Federal Circuit, and the Supreme Court: A Quiet Revolution’, Supreme Court Economic Review, 11, 1–79. Macdonald, S. (2002), ‘Exploring the Hidden Costs of Patents’, in P. Drahos and R. Mayne (eds), Global Intellectual Property Rights: Knowledge, Access and Development, Basingstoke: Macmillan. Macdonald, S. (2003), ‘Bearing the Burden: Small Firms and the Patent System’, Journal of Information, Law and Technology, http://www2.warwick,ac.uk/fac/soc/ law/elj/jilt/2003. Macdonald, S. (2004), ‘When Means Become Ends: Considering the Impact of Patent Strategy on Innovation’, Information Economics and Policy, 16, 135–58. Machlup, F. (1958), An Economic Review of the Patent System, Study No. 15 of the US Senate Subcommittee on Patents, Trademarks and Copyrights, Washington DC. Machlup, F. and E. Penrose (1950), ‘The Patent Controversy in the Nineteenth Century’, Journal of Economic History, X, 1–29. Mairesse, J. and P. Mohnen (2003), ‘Intellectual Property in Services: What Do We Learn from Innovation Surveys?’, in OECD (ed.), Intellectual Property Rights (IPR), Innovation and Economic Performance, Paris: OECD. Mandeville, T. (1996), Understanding Novelty: Information, Technological Change, and the Patent System, Norwood, NJ: Ablex. Mandeville, T., D. Lamberton and E. Bishop (1982), Supporting Papers for the Economic Effects of the Australian Patent System, Canberra: Australian Government Publishing Service. Mansfield, E. (1986), ‘Patents and Innovation: An Empirical Study’, Management Science, 32, 173–81. Mansfield, E. (1994), Intellectual Property Protection, Foreign Direct Investment and


53

Technology Transfer, Discussion Paper 19, New York: World Bank, International Finance Corporation. Mansfield, E., J. Rapoport, A. Romeo, S. Wagner and G. Beardsley (1977), ‘Social and Private Rates Return from Industrial Innovations’, Quarterly Journal of Economics, 71, 221–40. Mansfield, E., M. Schwartz and S. Wagner (1981), ‘Imitation Costs and Patents: An Empirical Study’, Economic Journal, 91, 907–18. Mazzoleni, R. and R. Nelson (1998), ‘The Benefits and Costs of Strong Patent Protection: A Contribution to the Current Debate’, Research Policy, 27, 273–84. Merges, R. (1999), ‘As Many as Six Impossible Patents before Breakfast: Property Rights for Business Concepts and Patent System Reform’, Berkeley Technology Law Journal, 14, 577–615. Merges, R. (2003), ‘The Uninvited Guest: Patents on Wall Street’, Economic Review – Federal Reserve Bank of Atlanta, 88, 1–14. Mischlewski, D. and M. Dormer (1995), Patents and Innovation: An Investigation of the Commercialisation of Inventions Patented by Australian Organisations, Ashfield, NSW: Watermark Patent and Trademark Attorneys. Moir, H. (2008), ‘Who Benefits? An Empirical Analysis of the Ownership of Australian Patents, European Consortium for Political Research (ECPR)’, Workshop 14: The Politics of Intellectual Property, Rennes, France. Now available as a RegNet working paper at: http://www.cgkd.anu.edu.au/menus/workingpapers.php Mokyr, J. (2002), ‘Innovation in an Historical Perspective: Tales of Technology and Evolution’, in B. Steil, D. Victor and R. Nelson (eds), Technological Innovation and Economic Performance, Princeton: Princeton University Press. Mooney, B. and C. Oppenheim (1994a), ‘The Cost of Patent Protection to U.K. Industry’, Journal of the Chartered Institute of Patent Agents, 23, 144–60. Mooney, B. and C. Oppenheim (1994b), ‘How Much Does British Industry Pay for Patents and Patent Information?’, World Patent Information, 16, 140–49. Moser, P. (2005), ‘How do Patent Laws Influence Innovation? Evidence from Nineteenth Century World Fairs’, American Economic Review, 95, 1214–36. Murphy, F. (2002), ‘The Occasional Observer <www.uspto.gov> a New Source for What is Happening in Operations Research Practice’, Interfaces, 32, 26–9. Nelson, R. (2006), ‘Reflections on David Teece’s “Profiting from Technological Innovation…”’, Research Policy, 35, 1107–9. Oppenheim, C. (2000), Patent Information Usage in British Small and Medium Sized Enterprises, http://scientific,thomson,com/free/ipmatters/bti/8199738/ (last accessed 25 August 2008). Pakes, A. and M. Simpson (1989), ‘Patent Renewal Data’, Brookings Papers on Economic Activity, Special Issue on Microeconomics, Washington: Brookings Institution, pp. 331–401. Perelman, M. (2002), Steal This Idea: Intellectual Property and the Corporate Confiscation of Creativity, New York: Palgrave. Phillips, R., J. Chen, R. Okediji and D. Burk (2004), ‘Intellectual Property Rights and the Public Good’, The Scientist, 18(14), 8. Plant, A. (1934), ‘The Economic Theory Concerning Patents for Inventions’, Economics, New Series, 1, 30–51. Porter, M. (1990), The Competitive Advantage of Nations, New York: Free Press. Rivette, K. and D. Kline (2000), ‘Discovering New Value in Intellectual Property’, Harvard Business Review, 78, 54–66. Robson, S. and L. Ortmans (2006), ‘The First Findings from the UK Innovation Survey, 2005’, Economic Trends, 628, 58–64.

54


Roland Berger Market Research (2004), Study on the Cost of Patenting, Munich: European Patent Office. Scherer, F. and S. Weisburst (1995), ‘Economic Effects of Strengthening Pharmaceutical Patent Protection in Italy’, International Review of Industrial Property and Copyright Law, 26, 1009–24. Scherer, F., S. Herzstein, A. Dreyfoos, W. Whitney, O. Bachmann, P. Pesek, C. Scott, T. Kelly and J. Galvin (1959), Patents and the Corporation: A Report on Industrial Technology under Changing Public Policy, Boston: privately published. Scotchmer, S. (1998), The Independent-Invention Defense in Intellectual Property, Berkeley Olin Program in Law & Economics, Working Paper Series: 1132, Berkeley. Sell, S. (2003), Private Power, Public Law – The Globalization of Intellectual Property Rights, Cambridge: Cambridge University Press. Sena, V. (2004), ‘The Return of the Prince of Denmark: A Survey on Recent Developments in the Economics of Innovation’, Economic Journal, 114, 312–32. Serrano, C. (2006), The Dynamics of the Transfer and Renewal of Patents, University of Toronto Department of Economics, Working Papers. Sheehan, J., C. Martinez and D. Guellec (2003), ‘Understanding Business Patenting and Licensing: Results of a Survey’, presented at the OECD Intellectual Property Rights (IPR), Innovation and Economic Performance Conference, Paris. Sorensen, J. and T. Stuart (2000), ‘Aging, Obsolescence, and Organizational Innovation’, Administrative Science Quarterly, 45, 81–112. Stallman, R. (2001), ‘The Danger of Software Patents’, Speech at Government Model Engineering College, India, http://www.gnu.org/philosophy/stallman-mecindia.html. Taylor, C. and Z. Silberston (1973), The Economic Impact of the Patent System: A Study of the British Experience, Cambridge: Cambridge University Press. (US) CBO (2005), R&D and Productivity Growth: A Background Paper, Washington: Congress of the United States (Congressional Budget Office). (US) FTC (2003), To Promote Innovation: The Proper Balance of Competition and Patent Law and Policy, Washington: Federal Trade Commission. van Pottelsberghe, B. and D. François (2006), The Cost Factor in Patent Systems, Working Paper 06/002, Brussels: Solvay Business School/Centre Emile Bernheim, University Libre de Bruxelles. Watson, D. and M. Holman (1970), ‘The Concentration of Patent Ownership in Corporations’, Journal of Industrial Economics, 18, 112–17.

4. Strong patent rights, weak patent standards and innovation in biomedicine Dianne Nicol I.

INTRODUCTION

There can be little doubt that the relationship between patenting and innovation is complex. Innovators who patent are provided with the advantage of a head start in bringing their inventions to the marketplace, but for follow-on innovators these same patents could block their route to market. An optimally functioning patent system will properly balance the innovation advantage provided to patent holders with the concomitant risk of innovation blockage for follow-on users. With the advent of high technology, the pace of innovation has increased dramatically. Questions need to be asked about whether the current patent system, which has essentially remained unchanged for centuries, is properly adapted to facilitate innovation in new areas of technology. In this chapter I explore the relationship between patenting and innovation in one specific area of high technology – that of ‘translational’ biomedical research. This sector focuses on taking ‘the outcomes and innovations developed within the research context into clinical practice’ (Kaye, Hawkins and Taylor 2007, p. 739). This type of research is to be distinguished from pure or basic research, traditionally performed in universities and other public research organizations, and downstream research that is tied to manufacture, marketing and product development, which would normally be conducted inhouse by large pharmaceutical companies or contracted out. Much has been written about the extent to which patents and patent licences in the biomedical sector could negatively impact on innovation by others, by creating hold-ups and anticommons effects. These issues will be considered briefly later in this chapter. Somewhat less attention has been focused on their potential to impact positively on innovation. This chapter focuses attention on this issue by assessing the extent to which patent owners and their licensees are actually reaping the benefits of the patent monopoly. Patenting and licensing are the primary mechanisms currently used by the biomedical industry for 55

56


translating research into practice. Patentable technologies that are utilized by this sector include gene sequences, proteins, vectors, vaccines, drug targets and other research tools. Patents of this nature create strong rights, because they have the capacity to provide extensive domains of exclusivity. Patent holders working within these domains can add value to their inventions both through their own follow-on research and development and by licensing out to other users. If strong patent rights are used in this way, they should foster innovation. The biomedical research sector includes public and private research organizations, spin-offs, small and medium sized core biotechnology companies and some of the large pharmaceutical companies (but only to the extent that they are involved in translational, rather than downstream research). To a large extent, then, this sector is clustered around the interface between the public and private sectors. I have chosen to focus on this sector because this is the area where some of the most intense debate about the relationship between patenting and innovation has arisen. It is the archetypal example of a research sector falling within Pasteur’s quadrant, as propounded by Donald Stokes (1997), where we see a merger of basic and applied research. This is an area of elision between discovery and invention, public knowledge and proprietary rights, public funding and private investment, and scientific norms and market forces. This is a sector where, to quote Rebecca Eisenberg’s oft cited phrase: ‘One firm’s research tool is another firm’s end product’ (Eisenberg 1997 p. 374). Central to the arguments that I will canvass in this chapter are the observations: (a) that governments around the world have fostered the development of indigenous biomedical research sectors by offering participants strong intellectual property rights and other incentives; and (b) that in many countries these sectors are the seat of biomedical innovation. As a consequence, I conclude that great care is needed in modifying existing arrangements. Changes to the law, for example, excluding gene patenting, could have deleterious consequences on the viability of this sector. Changes to government policies, for example, fostering closer relationships between public research organizations and large pharmaceutical companies, could create similar problems if small biotechnology firms are excluded. In either case, innovation could be stifled if small biotechnology firms lose the benefits associated with strong patent rights and the freedom to use those rights. I recognize that there has not been perfect harmony in the way that the patent system has operated in biomedicine for the past couple of decades – far from it. Changes to patent standards in the late 20th century facilitated patent claims associated with gene sequences and related subject matter at the cusp of the discovery–invention interface. Because of the way that the invention and disclosure requirements were assessed, many of the early patents that were

Strong patent rights, weak patent standards

57

granted included broad claims over these new technological developments. As a result, they provided strong patent rights to their owners and licensees, thereby laying the foundation for the development of the biomedical sector. Broad rights of this nature can be justified in new areas of technology if they foster truly innovative developments. However, if novelty, inventive step and other patent standards are set too low, it is not just true innovations that are rewarded. Insignificant developments will also receive the benefit of strong rights. In biomedicine, weak application of patent standards created a veritable ‘gold rush’ mentality in the industry in the 1980s and 1990s, and the subsequent flood of patents created the potential for detrimental consequences. My assessment of the innovative potential of the biomedical industry and the need to modify the current patent system in this chapter will require consideration of all of the consequences flowing from the availability of strong patent rights and the lowering of patent standards. I recognize that analysis of other industry sectors is also of importance in any debate about patenting and innovation. Agricultural biotechnology, semiconductors, software and business methods are also the subject of intense scrutiny and debate and different considerations do apply.1 Although patent law is supposedly technology neutral, some of the issues discussed in this chapter are unique to the biomedical industry sector, and indeed some are specific to particular sub-sectors. This is because, in its application, patent law has different effects in different industry sectors. The current debate about reform of the patent system in the US is illustrative of this point. Small biotechnology companies fear the proposal to reduce damages awards in patent infringement actions, whereas small IT companies see it as a boon (Wadman 2007, p. 651). Moreover, the way that the law is interpreted by the courts is becoming increasingly technology specific, at least from the perspective of the US courts (Burk and Lemley 2002). As a consequence, the conclusions that are reached in this chapter are not necessarily generic to industry as a whole. But my intention is that this analysis will be illustrative of the complexities involved in determining the relationship between patenting and innovation, and the diverse interests that need to be taken into account in considering the way forward. Because patents play such a dominant role in biomedicine, their impact in this sector must be considered in any future modelling.

II. THE STATE OF PLAY IN BIOMEDICINE: A POTTED HISTORY Recombinant DNA technology is at the core of all modern biotechnology research. The fundamental breakthrough in the development of the technology

58


was made by Herbert Boyer and Stanley Cohen at Stanford University and the University of California in the 1970s, opening up vast new areas of research possibilities. It has been described as ‘arguably the defining technique of modern molecular biology’ and ‘the founding technology of the biotechnology industry’ (National Research Council 1997, p. 40). This is of interest with regard to the issues being debated in this chapter not only because of the value of the technology as a research tool, but also because the technology itself was patented and Genentech, the first ever biotechnology company, was established by Herbert Boyer and others to utilize the technology to make therapeutic proteins (Hope 2008, p. 32). The technology itself was non-exclusively licensed by Stanford University and the University of California, the assignees of the patents, to many research organizations for small fees. This strategy turned out to be remarkably lucrative, with returns of US$139 million by 1995 (National Research Council 1997, pp. 40–41). Genentech, too, was highly successful in making innovative use of the technology. In 1982, Genentech and Eli Lilly obtained marketing approval for recombinant human insulin, the first biomedical therapy to do so (DiabetesHealth 2008). This would thus seem to be the perfect marriage of patenting and innovation. However, subsequent events in the modern history of patenting of biotechnological inventions have led many to question whether the relationship between patenting and innovation is quite so harmonious. In particular, events in the early 1980s, also occurring in the US, laid out the framework within which the subsequent debates about patents and innovation would be formulated. While these events were not precisely mirrored in other jurisdictions, to a large extent the major European and Japanese trading blocs, as well as Australia and other industrialized nations, followed the lead of the US. For this reason, key US developments are briefly outlined below to give context to subsequent discussions. Later in this chapter, I will turn some attention to the Australian biomedical industry and Australian patent law, focusing particularly on my empirical work with Jane Nielsen in 2002–03 (Nicol and Nielsen 2003). In 1980 the US Supreme Court handed down its landmark decision in Diamond v Chakrabarty, which allowed a patent for a bacterium that had been modified so that it could break down hydrocarbons.2 The holding in Chakrabarty has been interpreted expansively by US courts and the US Patents and Trademarks Office, and courts and patent offices in other jurisdictions, facilitating patenting of a wide range of biotechnological inventions (Nicol 2005, pp. 824–8). The only requirement to bring material that has been isolated or purified under the umbrella of patentable subject matter would seem to be that it offers some material advantage in utility over the naturally occurring material. In the same year as the Chakrabarty decision, the Bayh–Dole Act entered


59

into force. This legislation provided clarification that funding agencies would not claim ownership of intellectual property created using their funds. This legislation is widely regarded as providing the necessary impetus for patenting and licensing of inventions created in the public sphere, although David Mowery and his colleagues question this linkage (Mowery et al. 2004). They argue that the increase in patenting and licensing by US universities in the late 20th century would have happened anyway and was already happening prior to the introduction of the legislation (Mowery et al. 2004, pp. 1–2 and 99–128). The outcome of this debate is not central to this chapter, but what is important to note is that the 1980s and 1990s saw a massive increase in patenting by universities and other public research organizations (Hall and Ziedonis 2001, pp. 105–106; Hope 2008, p. 35), together with the emergence of technology transfer as one of the themes of the university mission. While other jurisdictions do not have legislation directly equivalent to the Bayh–Dole Act, the push for universities to engage in patenting and technology transfer is a common feature in the industrialized world more generally. In 1982, the US Court of Appeals of the Federal Circuit (CAFC) was created as a specialist court to hear patent appeals. In its early years, the CAFC was widely seen as being pro-patent, tending to uphold patent infringement actions and reject challenges to patent validity. The fact that the court could award triple damages for wilful infringement was an additional benefit to patent owners. Added to this, the court’s willingness to grant interlocutory injunctions against alleged infringers further strengthened the position of patent owners (Jaffe and Lerner 2004; Hall 2007, p. 11). Over the next two decades various industries, including the biomedical research sector, took advantage of the availability of these strong patent rights. Meanwhile, the technological developments continued apace. In the late 1980s, plans to map and sequence the human genome were crystallized into the transnational Human Genome Project. Scientists like Craig Venter, who at the time was working at the US National Institutes of Health (NIH), were developing new, faster methods for sequencing DNA. One of Venter’s strategies to speed the sequencing process was to generate vast numbers of short fragments of DNA (known as expressed sequence tags, or ESTs). Once numerous patent applications relating to these ESTs were filed by the NIH, the stage was set for the patent portfolio races of the 1990s (Nicol 2005, p. 813). These events also brought simmering academic concerns about the capacity for patents to impact detrimentally on innovation out into a wider audience. Terms such as ‘cumulative innovation’ (Scotchmer 1991), ‘hold-ups’ and ‘thickets’ (Shapiro 2001) and the ‘tragedy of the anticommons’ (Heller and Eisenberg 1998) entered the lexicon. The basic concerns were that broad patents over foundational discoveries could fence off whole areas of research and an overabundance of fragmented ownership rights could create too many fences, both

60


of which, together or independently, could deter innovation. The impact of strong patent rights and weak patent standards on innovation in biomedicine has to be assessed against this historical backdrop.

III.

STRONG PATENT RIGHTS AND THE UNIVERSITY

Ever since strong patent rights have been available in biomedicine, universities and other public research organizations have been some of the most active patenters (Pressman et al. 2006). It is not within the mission of these organizations to engage in product development and hence innovation inevitably requires transfer of the information protected by the patent to industry, together with associated technology. The optimal way to transfer technology is to put it into codified form: a patent specification, a refereed journal, a tangible reagent or assay. It is much more difficult to transfer tacit information like know-how, unless the personnel possessing the know-how are themselves transferred. Either that or they engage in training programmes to pass on their know-how to others. In the traditional university, codified information is freely disseminated through publication in peer-reviewed journals and conference proceedings, and materials are freely shared. Tacit information is exchanged through training and mentoring of postgraduate students. In the new university, information that can be codified is protected by patents. Tacit information can also be caught up in complex contractual transfer arrangements, which suppress disclosure to protect future codification and patenting. While acknowledging that this is an over-simplification, it is fair to say that free exchange is only part of the picture now and is much less the dominant norm than in the past. Researchers and institutions must consider confidentiality obligations and future intellectual property rights when making decisions as to what can and cannot be freely disseminated (e.g. Australian Government 2007, section 4). From empirical work that I carried out with my colleague Jane Nielsen in 2002–03 (Nicol and Nielsen 2003; Nicol and Nielsen 2005), we were able to form a picture of technology transfer strategies in Australian universities and research organizations. It appears that assignment was a favoured technology transfer strategy early in the biotechnology revolution because it was seen to reduce risk. However, the cost is that it also removes all control. If the assignee decides not to pursue development of the technology, all the potential for further innovation is lost (Nicol and Nielsen 2003, p. 108). More recently, there has been a growing trend either to retain ownership and license out to industry or to form a spin-off company for the purpose of commercializing a particular product and to assign or exclusively license the technology to the spin-off. Such exchanges are likely to be particularly effec-


61

tive from the perspective of individual firms if the personnel with tacit knowledge are transferred alongside the transfer of the coded information in the patent. For example, when Herbert Boyer established Genentech, he brought his tacit knowledge of recombinant DNA technology with him from the public sector to the private sector, enhancing Genentech’s prospects of successfully developing innovative products, which did in fact eventuate. The question of whether, more generally, the transfer of codified and tacit information encourages innovation at the level of the firm will be canvassed more fully in the next section. The problem for universities when personnel are transferred to the private sector is that they lose their best and brightest researchers. The tacit information that those researchers would have continued to use in future scientific research, the codified results of which would have been widely disseminated, becomes locked up in the firm.3 The firm may also limit the extent to which researchers are free to pass on their tacit knowledge to students, which those students would have used in their future academic careers and freely disseminated to future generations of students (although intra-firm exchange of this tacit information would, of course, be facilitated). In the case of recombinant DNA technology, locking up Boyer’s tacit knowledge within Genentech was not so much of a problem because the technology itself was made widely available to other research organizations and they did not require further tacit knowledge to utilize it. But this will not always be the case in sectors like biomedicine, where innovation is cumulative (Scotchmer 1991; Scotchmer 2004, pp. 127–59). If this tacit information is locked up in personnel transfers, licence deals or material transfer agreements, innovation may be seriously undermined both in terms of early stage creativity in universities and other public research organizations and product development in the downstream sector. This paints a rather bleak picture of the new university. How is it that we have gone down the path of putting upstream information exchange into the marketplace? It has been argued that forcing the flow of information into the marketplace delays exchanges and increases costs because it requires arm’s length negotiations, and this is particularly troublesome in areas like biotechnology where a great deal of information remains uncodified (Hope 2008, pp. 84–6). The reason why this path has been chosen goes to the heart of the purpose of the patent system (Macdonald 2003). While free publication and exchange of scientific information is vital for the progress of science, it does not provide an incentive for commercial development of the information. For large downstream companies with deep pockets this does not present too much of a problem, because they will have the opportunity to recoup their investment in research and development from sales of products during the time that they are protected by downstream patents. But for small firms in the biomedicine

62


sector, patenting at this upstream level is seen as being vital, both for bargaining with the downstream players and for attracting financing (Nicol and Nielsen 2003, p. 84). It is also of importance for universities because it encourages commercial sponsorship of research (e.g. Australian Government 2007, Introduction to Section 4). For better or for worse, both patenting and formalized technology transfer have become mainstream strategies for disseminating the results of biomedical research, in addition to the more traditional forms of publication in peerreviewed journals. One consequence is that it has become more difficult to transfer codified and tacit information by other means. This may result in deleterious consequences from the broad innovation perspective. However, it is important to acknowledge that the transfer of strong patent rights to small firms benefits them by enhancing their capacity to innovate and this has flowon benefits to downstream innovators. The extent of this beneficial effect will be debated later in this chapter. Additionally, strong patent rights may have indirect benefits for the university, if the licensing revenue or collaborative industry funding that they provide can be invested in further innovative research that would not otherwise be funded. But as with all such statements, this must come with a rider. There is evidence that licensing revenue rarely covers the cost of administration. While some US universities have flourished under their modern technology transfer strategies, this is not the norm (Mowery et al. 2004, pp. 142–9). However, there are encouraging signs that as technology transfer offices mature and become more strategic the situation is improving. It is also encouraging that commercial sponsorship of university-based research tends to provide an increasing percentage of university revenue streams, particularly when compared with licensing of the patentable outcomes of publicly funded research (Benkler 2004, p. 1111). But at the same time we should be concerned that commercial funding of this nature could turn university scientists into mere contract researchers, lessening the prospects for their research to be truly innovative and increasing the risk that confidentiality obligations will restrict broad dissemination of research results. Where innovative research does occur, if it is tied to collaborative funding there is a risk that industry partners may be more inclined to suppress it than to develop it. Such arrangements need to be scrutinized with care, to ensure that they do not unjustifiably impede innovation and disclosure. One of the difficulties for technology transfer offices is that successful strategies are highly technology specific, even within a single sector. Mowery and colleagues provide five case studies that illustrate this point (Mowery et al. 2004, pp. 152–78). They say that this illustrates the need for caution in making generalizations about technology transfer and the role of intellectual property rights (Mowery et al. 2004, p. 154). There will be instances where, if


63

managed properly, strong patent rights enhance innovation, but other instances where they may not.

IV. A.

STRONG PATENT RIGHTS AND THE SMALL FIRM The Positives of Strong Patent Rights

One of the defining features of the biomedical industry is that it is dominated by small to medium sized enterprises rather than large firms. In Australia, for example, the number of core biotechnology companies in 2006 was 445 (Innovation Australia 2007), only a small per centage of which could be classified as large. While there are considerably more large companies in other industrialized nations, small to medium sized enterprises still dominate, and new players continue to join the market. For example, in 2005 there were 1415 biotechnology companies in the United States, 329 of which were publicly held (Biotechnology Industry Organization 2005). Government grants are the most important source of funding for these young companies, which struggle to attract venture capital (Australian Law Reform Commission 2004, p. 405). It could well be the case that for these companies, strong patent rights might be even more important for attracting investment than for facilitating their capacity to innovate. The longevity of these small players is an issue that warrants deeper analysis, but for now the important message is that the global biomedical industry sector has been an active growth phase for the past 25 or so years and it continues to be dominated by small innovative firms, most, if not all of which, have patent portfolios. The structure of the biomedical sector is significantly different from agricultural biotechnology, where a small number of large multinationals dominate. Concerns about the detrimental impact on innovation in a market dominated by small numbers of mega-firms have led to the establishment of a number of collaborative initiatives in the agribiotech sector aimed at working around some of the blockages created by such market dominance. Two notable examples are PIPRA, a clearinghouse-like organization originally established by US public sector organizations, and BiOS, an open source-like organization based in Australia (Nicol and Hope 2006). Equivalent mechanisms have not yet emerged in the biomedical sector, although there is growing interest in exploring them, at least at the academic level. There may be many reasons why agribiotech and biomedicine have different industry structures. One may be that biomedical technologies are too disparate to develop meaningful collaborations; another may be attitudinal. Arguably, the so-called ‘bunker mentality’ of the exclusive rights model continues to prevail in biomedicine (Horn 2003, p. 123). However, one other reason may simply be that there is

64


less perceived need. It may be that the mix of small, specialized biomedical firms and strong intellectual property rights produces optimal conditions for innovation in this sector. We have seen that in biomedicine, relaxation of the thresholds for patenting of gene sequences and other research tools has occurred in parallel with a surge in the number of small companies entering the field. William Kingston postulates that the effect of lowering standards of patentability should be to reinforce the business ascendancy of large firms (Kingston 2004). This may well be the case with regard to downstream inventions, but in biomedicine the large pharmaceutical firms tend to leave this biomedical sector of the market to the smaller specialist firms. The extent to which the availability of strong patent rights is actually causative of the growth in numbers of spin-offs and small to medium sized enterprises in biomedicine is difficult to assess. The views of participants in the industry clearly support the notion that patenting is vital to their success. In my empirical work with Jane Nielsen in 2002–03 we found that respondents in Australian biotechnology firms generally considered that patents were essential for recovery of research and development expenditure and to ensure there was adequate incentive for innovation and investment in the industry. For example, one respondent stated that: ‘patent protection must be sought for our company’s survival’. Another suggested that in the absence of patents ‘…there would be no industry’ (Nicol and Nielsen 2003, p. 84). In contrast, much earlier empirical studies in other industry sectors suggest that, at best, patents provide a small incentive for innovation (Arora, Ceccagnoli and Cohen 2004, pp. 4–5). The question to be addressed here is whether theory supports the strong patent model for achieving optimal innovation capacity within the small firm, whether empirical data on the performance of participants in the industry support that theory, and whether the industry as a whole benefits from this model. B.

Market-Based Exchanges

Robert Merges has presented an interesting case as to why strong intellectual property (particularly patents) should facilitate the viability of small specialist firms within the marketplace (Merges 2005; Arora and Merges 2004). Whether or not the small firm, market-based model is, of itself, the optimal strategy is an important point – which will be returned to later. Theoretical economists of the so-called post-Chandlerian school certainly view 21st century markets in this way, seeing them as comprising a large number of independent firms that provide their technology to meet the various needs of the market.4 So, in biomedicine, we see that ‘large pharmaceutical companies contract with dozens of smaller research-intensive firms … to develop specific products’ (Merges 2005, p. 1518).


65

Other studies of the biomedical industry lend support to this assertion (Walsh, Arora and Cohen 2003, pp. 322–323; Straus, Holzapfel and Lindenmeir 2002, pp. 6–7). In the Nicol–Nielsen study, we found that, while the Australian industry is not quite so dynamic as in other jurisdictions, it still mostly comprises small firms with an extensive network of collaborations and licensing arrangements (Nicol and Nielsen 2003, p. 209). The study of the semiconductor industry by Bronwyn Hall and Rosemary Ziedonis also lends support to this argument, albeit in a totally different industry sector (Hall and Ziedonis 2001). I recognize that the thesis presented by Merges is contradicted by other theoretical economic literature, which suggests that new technological developments within an industry will tend to result in consolidation into fewer larger entities (Scherer 1984. p. 287, Nielsen 2008). It is difficult for the non-economist to enter into the debate, except to reiterate that the evidence points to a preponderance of small firms in the biomedical sector. The analysis by Merges and others is premised on the notions that information needs to be codified to be transferred and that transfer of information should occur in the marketplace (Hope 2008, p. 73). The argument put by Merges is that strong intellectual property rights encourage small firms to enter into transactions by facilitating disclosure of other sensitive information during negotiations and enhancing enforcement options (Merges 2005, p. 1488). Dealing with his more straightforward enforcement argument first, an action for patent infringement can be more advantageous to the patent holder than breach of contract because greater damages may be awarded, and contractual privity issues and limitations periods may be avoided (Merges 2005, pp. 1504–9). The disclosure issue needs to be teased out in more detail. As a general rule, disclosures in patent specifications are likely to be kept to the minimum necessary to fulfil the statutory requirements for obvious strategic reasons. This means that even when the technology under negotiation is protected by a patent, further tacit knowledge may need to be disclosed during negotiations to facilitate transfer. Merges sees patents as facilitating this exchange within the marketplace. Indeed, he suggests that the primary purpose of patents is, in fact, to ‘spearhead the transfer of the really valuable stuff’, and points to economic literature that supports this suggestion (Merges 2005, p. 1501). The extent of disclosure in modern patents is the subject of growing concern in the academic literature. Stuart Macdonald, for example, argues that the conventional theory that the patent system encourages innovation through disclosure no longer holds true (Macdonald 2003, section 1.7). Rather, for Macdonald, the only parties who are encouraged to innovate are patent owners, because of the protection provided by the patent (Macdonald 2003, section 1.7). While this has always been considered as one of the ways in which the patent bargain promotes innovation, the question is whether this is

66


enough in terms of public benefit. Macdonald’s concern is that larger firms, particularly those in the pharmaceutical and chemical industries, will be the only ones to benefit, because they are best equipped to innovate (Macdonald 2003, section 1.8). But their controlled and codified models of innovation are not suited to other sectors. The result is that ‘the patent system is essentially anti-innovative’ (Macdonald 2003, section 2). Yet the empirical evidence points to the fact that small firms have found ways to innovate within the confines of the existing patent system, at least in biomedicine. This is not to deny that there are imperfections, but it questions whether the picture is quite so gloomy as portrayed. The changing relationship between disclosure and innovation raises important questions that warrant further analysis, but it is beyond the scope of this chapter to explore this issue more fully. In any case, the arguments advanced by Merges suggest that patents do in fact continue to encourage innovation by disclosure, albeit not through disclosure in the patent specification itself. Although the ‘valuable stuff’ disclosures envisaged by Merges are not public, they still have the capacity to encourage downstream innovation. There is good reason why small firms might not be willing to freely disclose unprotected information, particularly when it is valuable, because of the risk that others will free ride on it. Merges argues that while disclosure of his ‘valuable stuff’ during negotiations will not have the protection of the patent if negotiations break down, patents do nevertheless encourage disclosure because they provide a level of protection from opportunism by offering some relief if the other party subsequently markets similar technology (Merges 2005, p. 1503). Arora and Merges attempt to explain further how this might work in practice, with a concrete example involving Genentech as a licensee of technology owned by Alkermes, a small, ‘patent intensive’ drug delivery firm. They state that the agreement executed by the parties has the dual effect of protecting Genentech against inadequate efforts by Alkermes to customize the technology to suit its own purposes, and protecting Alkermes from opportunism by Genentech because its patents allow it to ‘take back’ its technology once the agreement terminates (Arora and Merges 2004, p. 458). What Arora and Merges conclude from their theoretical discussions, concrete examples and economic modelling is that although integration of small firms into larger entities avoids rent dissipation, independent firms are more likely to specialize if they are provided with sufficiently strong incentives. The joint gains from the synergistic relationship between a supplier and multiple users outweigh rent dissipation, particularly because the supplier is able to learn something from each customer (Arora and Merges 2004, pp. 470–71). Arora and Merges do recognize the twin challenges to innovation created by patents that are too strong (and, what is argued in this chapter to be a corollary, standards that are too weak): the problem of hold-up caused by


67

broad patents over foundational discoveries in cumulative technologies and the anticommons problem (Arora and Merges 2004, p. 472). Working out optimal patent strength is a challenging exercise. Much of what Merges and Arora say is supported in a more recent article by Dan Burk and Brett McDonnell (Burk and McDonnell 2007) and by Oren Bar-Gill and Gideon Parchomovsky (Bar-Gill and Parchomovsky 2005). Burk and McDonnell argue that intellectual property rights should be neither too weak nor too strong, but ‘just right’ (Burk and McDonnell 2007, pp. 577–8). According to their ‘Goldilocks hypothesis’, if intellectual property protection is too weak, smaller firms will integrate into larger firms and larger firms will develop their ideas and undertake their transactions within the firm. Specialization is likely to decrease as a consequence, and where new developments do occur they will be locked up within the firm. Burk and McDonnell add that if patent rights are too strong, the transaction costs may be too high for buyers, and those that choose to invent around may be caught in the patent thicket. Moreover, strong patents may encourage over-specialization and overtransacting. Either that or they may also lead to integration and intra-firm rather than inter-firm transactions. Hence, there is an optimal level of intellectual property protection which, according to Burk and McDonnell, will vary between industries and societies and with time (Burk and McDonnell 2007, pp. 615–17). Working out the right level of patent strength is unlikely to be an easy or straightforward task. While Arora and Merges do not appear to have reached any firm conclusions about the state of the biotechnology industry, Burk and McDonnell conclude that the fairly high rate of failure or acquisition by big firms suggests that all is not right. They posit that patent rights are currently too strong because, according to the Goldilocks hypothesis, the predicted response to an overabundance of patents and an inefficiently large number of disintegrated firms is re-aggregation through acquisition or disappearance, which, they claim, is exactly what is being seen (Burk and McDonnell 2007, p. 635). On this basis, a case could be made for arguing that patent strength in biomedicine is not quite right. This assumes that attrition of small companies is either solely or mostly attributable to strong patent rights. There may be many other reasons why small firms fail. In Australia, for example, although attrition rates for small biomedical companies are high,5 they can be sheeted home to a range of factors other than strong patent rights per se. In fact, it is questionable whether there is even an overabundance of strong upstream patent rights in this jurisdiction. A number of the more contentious biomedical research tool patents have not been filed or have not been issued here (Nicol and Nielsen 2003, pp. 41–9). Other reasons for failure from the Australian perspective might include the tyranny of distance from overseas markets, lack of adequate venture capital and a rich

68


market allowing downstream developers to shop around for the most attractive technology to develop (Nicol and Nielsen 2001, p. 374). More empirical data need to be collected to test the assumptions underlying the arguments presented by Merges and others. It is also important to determine whether the assumptions made about the US system are transferable to other jurisdictions. In particular, it would be useful to explore what are the essential features of a ‘successful’ firm in various industry sectors and jurisdictions. C. Conflicting Rationales for Patenting: The Problem of Weak Patent Standards The notion presented by Merges that small firms acquire strong patent rights to facilitate specialization and innovation has some attractiveness. However, not all patents are acquired for the purpose of enhancing further research and development opportunities. It has been argued that strong patent rights undoubtedly encourage the use of patents as a business tool, but there is more doubt as to whether the primary focus of this business strategy is innovation (Hall 2007, p. 7). The empirical data on the relationship between strong patent rights and investment in research and development indicate that in most sectors the effect is small at best (Hall 2007, pp. 7–8). William van Caenegem also argues that the fact that research and development expenditure has not kept pace with patent growth tends to indicate that patents are not or not only seen as a protection for significant innovations, but for other strategic behaviour, such as erecting barriers to entry (van Caenegem 2007, p. 67). However, the pharmaceutical, biotechnology and medical instrument sectors are notable exceptions. Arora, Ceccagnoli and Cohen, for example, suggest that there is a positive relationship between strengthening patent rights and increased research and development spending in these sectors, but less so in other industries. They do, however, acknowledge that they do not address the question of whether patents create barriers to entry (Arora, Ceccagnoli and Cohen 2004, p. 32). Hall and Ziedonis point out that one of the consequences of the availability of strong patent protection and lowered patent standards in the late 20th century is that it led to races to build patent portfolios (Hall and Ziedonis 2001). Rather than patenting to facilitate innovation, the trend in the 1990s was to patent for defensive purposes. Effectively a positive feedback loop was created: the more patents that were filed, the more competitors had to file. It has been suggested that as a general rule firms are likely to patent more aggressively when rights to complementary patents are widely distributed (Ziedonis 2004). The rationale is that such patents have strategic value in subsequent licensing transactions. To round off this discussion, it is worthwhile to return to the empirical study of the semiconductor industry by Hall and Ziedonis. Based on their find-


69

ings that small firms tend to use patents to promote research and development, whereas large firms tend to use them more as bargaining chips, they conclude that the effects of strengthening patent rights are multifaceted, even in a single industry sector (Hall and Ziedonis 2001, p. 125). This comment illustrates the difficulty in reaching firm conclusions about the relationship between strong patent rights and innovation and the best means for changing that relationship, should change, of itself, be desirable.

V. STRONG PATENT RIGHTS, WEAK PATENT STANDARDS AND FOLLOW-ON INNOVATION In areas of cumulative innovation it is vital that patents are of the ‘right’ strength, whatever that might be. Some of the small specialist firms described above might be primary innovators whereas others will be following on. This, then, is a classic example of Eisenberg’s scenario where one firm’s research tool is another firm’s end product (Eisenberg 1997, p. 374). How might we measure what is the right strength for patent protection? There is a large, but conflicting, body of theoretical economic literature on optimal intellectual property strategies in areas of cumulative innovation. In particular, there has been extensive economic theorizing on the best model for allocating profits to first generation and subsequent innovators. The messages from these analyses are decidedly mixed. There are those that provide compelling arguments as to why broad upstream rights encourage investment in research and development (Scotchmer 1991; 2004). Others provide equally compelling arguments as to why such broad protection is not optimal in areas of cumulative innovation (Merges and Nelson 1995). The risk is that broad patents and patent thickets will cause hold-ups and anticommons effects. Hold-ups occur when the owner of a patent over upstream technology refuses to deal with a developer of downstream technology. In complex areas of cumulative innovation, particularly where patent standards are changing and the legitimate breadth of patent claims is uncertain, follow-on developers may inadvertently or consciously neglect to license in crucial upstream technology prior to development, putting the upstream technology owner in a much superior bargaining position than would otherwise have been the case. Heller and Eisenberg assert that there are two ways in which an anticommons may be created: through the existence of numerous overlapping property rights; and through the use of reach-through licence agreements leading to licence and royalty stacking (Heller and Eisenberg 1998, pp. 698–9). In their view, the explosion of patent grants within the biomedical industry and the increasing prevalence of restrictive licensing practices mean that an anticommons is more likely to arise in biomedicine than in other industries (Heller and

70


Eisenberg 1998, p. 701). The nature of the products being produced by this sector should also make us more concerned about deterrents to innovation than in other sectors. The complexities of the patent landscape inevitably increase as the translation from research into clinical practice moves further downstream. The risk is that, if there are hold-ups and anticommons effects, these are likely to significantly hinder the timely delivery of new drugs, diagnostics and therapies. Since the anticommons theory was first postulated, various methods have been developed to test whether there is concrete evidence that an anticommons exists in practice. There has been a deluge of patent applications over the last 25 or so years in biomedicine. For example, in a recent study by Kyle Jensen and Fiona Murray they conclude that nearly 20 per cent of all human genes have been claimed in patents granted in the US, with some genes featuring in up to 20 separate patents (Jensen and Murray 2005). They found that over 75 per cent of these gene patents have only one patent owner, but the remainder have fragmented ownership. The authors report that the two genes with the most fragmented ownership have 8 assignees for 9 patents, and 12 assignees for 14 patents. Similar findings are reported elsewhere (Verbeure, Matthijs and van Overwalle 2005). Such fragmented ownership could increase significantly the cost of access for downstream innovators. On the other hand, actual evidence that downstream innovation is being deterred in a significant way is not that easy to find. Stern and Murray used citations in peer-reviewed journals in an attempt to measure the anticommons (Stern and Murray 2005). Their hypothesis is that when knowledge is protected by intellectual property, follow-on research by others will be diminished. To test this hypothesis, Stern and Murray identified scientific knowledge that had been both published and patented (the patent–paper pair) and measured the number of citations pre- and post-patent grant. The control group was made up of other publications from the same journal that were not patented. They report that while papers from the patent–paper pair group were more highly cited, citations do significantly decrease post-grant, but only in a modest way (Stern and Murray 2005, pp. 26–7). While their results suggest that patents do impose some restrictions on the transfer of scientific knowledge by other means, they are hardly a ringing endorsement of the anticommons theory. Another body of empirical evidence suggests that while it may well be the case that multiple broad patents have been issued in biomedicine, this does not appear to have resulted in hold-up or an anticommons. This set of evidence comes from questionnaires and interviews conducted with industry participants. In particular, the work of John Walsh and his colleagues in the US suggests that while the preconditions for an anticommons may well exist, the industry as a whole is finding ways of working around the problem (Walsh,


71

Arora and Cohen 2003, p. 287; 2003b). My research in Australia and other research in Europe supports these findings (Nicol and Nielsen 2003; Straus, Holzapfel and Lindenmeir 2002; Intellectual Property Institute 2004). These studies consistently report prolific licensing activity. Industry players are able to license out their own technology or license in the technology they need to secure freedom to operate. It is common, in fact pretty much the norm, to see non-exclusive licensing of foundational research tools. It seems that, for the most part, research tool patent holders tend to license widely and users engage in a number of strategies to ensure that their research and development programmes can continue, including licensing in, inventing around, litigating to challenge patent validity or simply ignoring the patents that would otherwise block their research. In the public research sector, in particular, the available evidence suggests that researchers are rarely impeded in their research programmes by patents (Walsh, Cho and Cohen 2005), despite the absence of a clear exemption from infringement for research use in most jurisdictions and well-publicized case law in the US indicating that public sector research is not immune from infringement.6 However, the lack of compelling evidence that hold-ups and anticommons effects are being experienced should not make us overly confident that all is well. The simple fact of the matter is that more evidence is needed and new methods need to be developed for collecting that evidence (Gold et al. 2007).

VI. STRONG PATENT RIGHTS AND WEAK PATENT STANDARDS IN THE 21ST CENTURY: FACT OR FICTION? The analysis so far presents a mixed picture on the impact of strong patent rights and weak patent standards on innovation in biomedicine. There is some support for the notion that strong rights do encourage the establishment and sustainability of small specialist firms. These firms use their patents to their advantage in adding value to upstream research, attracting venture capital and bargaining. In this sense, then, innovation is promoted. However, weak patent standards encourage patent portfolio races. There is a risk that broad rights and fragmented ownership could deter follow-on innovation. If the industry is to continue to survive within the existing business model, then one important aspect is to ensure that patent rights are not too strong or too weak; that they are ‘just right’. In biomedicine, despite the developing body of jurisprudence, clear guidance as to the legitimate scope of upstream patents remains elusive. There is still ambiguity and uncertainty, particularly with regard to the ambit of the inventive step and novelty requirements, and the distinction between inventions and discoveries. Significant steps have

72


been taken in improving patent standards in some jurisdictions, particularly in the US, both through judicial decisions and through legislative and administrative guidelines, particularly with regard to the utility7 and the disclosure8 requirements (Nicol 2005). The recent decision in Re Fisher is particularly important, and would seem to finally put to rest concerns about the legality of EST patents.9 The inventive step requirement remains problematic in the US and Australia, and steps need to be taken to adopt a more stringent test, akin to that used in Europe. But further legislative reform is not necessarily the best option at this stage. Many of the early biomedical patents were probably overly broad. But this tends to be the case in any new area of technology. As more patents are granted in a new field, their scope necessarily decreases in order to satisfy the patentability criteria (Commission of the European Communities 2005, p. 4). There is other evidence to suggest that the negative aspects of weak patent standards and the patent portfolio races that they fostered in the 1990s are being dissipated. A group led by Parimal Patel in the UK released a report in November 2006 detailing the results of a study analysing the numbers and types of DNA patents by the major patent offices and the views of the major players in the industry (Hopkins et al. 2006). The report provides further data on the number of human gene patent applications and grants. But more importantly for the present purposes, it states that around a third of the patent applications had been withdrawn and just under a third of the patents granted in the 1990s had been abandoned by 2005 (Hopkins et al. 2006, Executive Summary). It appears that increased stringency in demonstrating utility and the other patent requirements is playing at least some role, both with regard to withdrawal and abandonment. While patenting in this area will continue, the authors of the study conclude that the negative impact of patenting in this area ‘may turn out to be more limited than some had feared’ (Hopkins et al. 2006, Executive Summary). In my study with Jane Nielsen, there was also a sense of optimism in the industry that concerns about overly strong patent rights and weak patent standards were being resolved. A number of our interview respondents commented that the early patents in the biomedical area were exceedingly broad, particularly those granted in the US in the early 1990s. However, there were various comments to the effect that the system appears to have worked itself out given that broad patents that have already been granted are not being upheld in the courts and it is much more difficult to be granted a broad patent now than in the past. Given that a considerable number of patents granted in the1990s and early years of this century may be invalid, some respondents said that they were content to ignore relevant patents and challenge their validity if approached by the patent holder. A number of respondents found that inventing around more recent, narrower patents had become more straightforward.


73

Patents that were broad but genuinely valid were not considered to be a problem given that most respondents were strong advocates of the patent system (Nicol and Nielsen 2003, pp. 86–91). The UK report indicates that the motivation for at least some of the filings in the 1990s was to ensure freedom to operate, rather than actually seeking patent exclusivity (Hopkins et al. 2006, pp. 28, 37). In such instances, the patent application would normally be abandoned before grant once it has been published (Hopkins et al. 2006, p. 28). While freedom to operate was a motive for filing in only a minority of circumstances overall, it was more prevalent in the pharmaceutical sector than elsewhere (five out of nine firms) (Hopkins et al. 2006, p. 28). This reinforces the argument that public sector organizations and small biomedical firms patent for the genuine purpose of facilitating research and development, whereas the bigger firms patent for other strategic purposes. If they patent with a view to withdrawing their applications or abandoning their patents, then hold-ups and anticommons effects should be minimized. If the market can sort itself out, perhaps we should not be unduly worried about weak patent standards and patent portfolio races. One difficulty, though, is that patent offices around the world are flooded with worthless patent applications, creating an administrative bottleneck. There are repeated criticisms in the literature of the time it takes for patent offices to examine patents and the quality of examination (Jaffe and Lerner 2004, pp. 127–50), but perhaps one strategy to deal with the issue raised above would be to delay rather than to speed up the examination process, allowing for as many worthless applications as possible to fall by the wayside. While this suggestion might not meet with wide approval, it should give some pause for thought about what we demand of our patent offices and why. It may create too much uncertainty to delay examination in some industries, but it may nevertheless be an option that is worthwhile to consider in new areas of technology to give the market time to settle.10 Another option is to take heed of Mark Lemley’s ‘rational ignorance’ argument (Lemley 2001). He presents a persuasive case as to why it would be irrational to pay increased attention to validity and to put increased resources into the assessment of patent applications. He argues that a massive increase in resources would be required to ensure validity of issued patents and that challenging the validity of questionable patents in court is more efficient (Lemley 2001). This would seem to be a pragmatic solution if many of these patents are abandoned post-grant. However, Lemley’s views are not universally shared (Rai 2003, pp. 1080–84). There are legitimate fears that improper laws and bad patents will impact negatively on innovation. Perhaps it would be worthwhile to explore the introduction of a system that filters out the freedom to operate patent applications. Something like the innovation patent system in Australia could be coupled with the more usual

74


examination system. Patents could be granted but not enforceable for an initial five-year period provided that they fulfil the formal requirements, after which they would lapse unless there was an application for examination. Requests could be made for earlier examination, but only for the purpose of enforcing the patent. This proposal is unlikely to find favour across the whole spectrum of patent-reliant industries. The pharmaceutical sector, for example, would doubtless prefer the certainty of timely examination. While technological specificity in examination procedures has some attraction, it is doubtful whether this is allowable under Article 27.1 of the World Trade Organization international Agreement on Trade-Related Aspects of Intellectual Property Rights.11 As such, although I argue that there is a need to explore more fully strategies for modifying current patent examination practices, this is likely to be a slow and frustrating process at best. In the short term, if the relationship between patenting and innovation in biomedicine does in fact need to be improved, it is likely to be more immediately productive to explore mechanisms for doing so within the existing patent framework. Some such possible mechanisms are canvassed in the next section of this chapter.

VII. THE WAY FORWARD: CHANGING THE BIOMEDICAL BUSINESS MODEL In this chapter I have attempted to highlight some of the key positives and negatives in the relationship between patents and innovation in the biomedical research sector. There are genuine concerns about the potential for patents to impact detrimentally on the primary research conducted in universities and other public research organizations that feeds in to the innovation cycle. Concerns about hold-ups and anticommons effects also need to be taken seriously. However, these concerns need to be balanced against the positive role that strong patents can play in encouraging innovation in small specialized firms and their licensees. I may have presented too optimistic a picture of the role that these firms play in biomedicine and inadvertently suggested that large pharmaceutical companies play a more benign role than is actually the case. But this analysis is presented with a view to moving the debate along out of the 1990s and into the 21st century. It was the case that too many upstream patents were filed and patent grants were too broad in the last century, but adjustments are being made. It was also probably the case that too many small firms were established on the back of the patent boom. Again, adjustments are being made, but the consequences are that some otherwise important innovations might be lost. For example, a recent report on the biotechnology industry in Australia suggests that the growth phase may be nearing an end, indicating that there is


75

evidence of consolidation in the industry (Innovation Australia 2007, p. 3). This is all well and good, so it seems. But some commentators give a much more damning condemnation of the state of the biotechnology industry. Gary Pisano views the biotechnology business model as fundamentally flawed, particularly because the small specialized firm model has created islands of expertise that impede the integration of critical knowledge (Pisano 2006). New biotechnology firms have mirrored the success of the early entrants, Genentech, Amgen and others. His view of the US biotechnology industry is that most companies are losing money, few new products are emerging, and venture capitalists are unwilling to commit to the time frame needed for drug development and approval (Pisano 2006). In Australia, biotechnology organizations are in a similar struggle to adequately resource their work (Vitale and Sparling 2003). Venture capital is difficult to obtain, government funds are small and any revenue (let alone profit) may be years away. Because the majority of Australian core biotechnology companies are investing significant amounts of their revenues in research and development, they are not profitable. A study of Australian biotechnology firms that listed between 1998 and 2002 found that, on average, the companies were approximately six years old when they listed. By 30 June 2003, when these firms were almost a decade old, less than half of them had received any revenue from product sales; only two had positive cash flow; and only one was profitable (Vitale and Sparling 2003). In such cases, companies have generally been sustained by government grants, interest and cash raised through public offerings. With these depressing statistics in mind, how does the industry move forward and fulfil its promise? The biomedical industry is now probably mature enough to start thinking about cooperative mechanisms to facilitate exchanges by bundling together of patents through such mechanisms as patent pooling, clearinghouses and other collective rights organizations. There has been debate about the role of such mechanisms in the literature for some time (Merges 1996; Heller and Eisenberg 1998). International agencies are turning their attention to these options (OECD 2004, p. 2), although preliminary data suggest that the industry itself remains to be convinced of their value (Hopkins et al. 2006). I am involved in a comparative study of industry attitudes with colleagues at KU Leuven in Belgium to test if there is any likelihood that such strategies might be more willingly embraced. If these views are negative or ambivalent, new ways need to be found to change these perceptions. New strategies are needed for the biomedical industry as a whole to achieve optimal innovative capacity, and they will not succeed if the relevant industry sectors are unwilling to participate. Other strategies also have a role to play, including open source, open access, prize funds and the like. Large pharmaceutical firms are playing an

76


important role in sponsoring important public sector initiatives, like the SNP Project and the HapMap Project, which are premised on open access. Without challenging the benevolence of these actions, the words of caution that I expressed at the start of this chapter need to be reiterated: if such actions threaten the survival of the small specialist biomedical firms that have grown up around strong patent rights, then this might ultimately impact detrimentally on innovation. In biomedicine, the small specialist bed has now been made and, to push Burk and McDonnell’s analogy one step further, if we don’t sleep in it, it will be broken and the innovation it promises will be gone!

NOTES 1. 2. 3.

4.

5. 6. 7. 8. 9. 10. 11.

As an example of these differences, Mann and Sager (2007) provide a comparison of the relationship between patent ownership and venture capital financing in the software industry and the biotechnology industry. 447 US 303, 309 (1980). As with other statements above on the changing nature of the modern university, it is recognized that this is an over-simplification. Scientists in the private sector can and do publish their results in peer-reviewed journals. The point is that it is likely that there will be greater requirements for them to have their work vetted prior to publication, which could impact on both timeliness and content. There are also likely to be increased restrictions on the transfer of tacit information. The post-Chandlerian school has emerged following the work of Alfred Chandler, a business historian from Harvard who described the traditional firm as large and vertically integrated. The new post-Chandlerian firm is modular and disaggregated, resulting in more inter-firm transactions (Merges 2005, p. 1514). For example, it was reported in 2007 that a number of small companies which had been ‘hanging on’ ceased operations in 2006 (Innovation Australia 2007, p. 3). Madey v Duke University, 307 F 3d 1351, 1360–1 (Fed. Cir. 2002). Utility Examination Guidelines, 66 Fed Reg 1092, 1095 (2001). Guidelines for Examination of Patent Applications under the 35 USC 112, para. 1, ‘Written Description’ Requirement, 66 Fed Reg 1099 (2001). 421 F 3d 1365 (Fed Cir. 2005). One other option could be to use competition law to regulate the conduct of large firms that patent for strategic anti-innovative purposes, particularly when this behaviour creates barriers to entry for others. I am indebted to Chris Arup for this suggestion. Marrakesh Agreement Establishing the World Trade Organization, opened for signature 15 April 1994, 1867 UNTS 3 (entered into force 1 January 1995), annex 1C (Agreement on Trade-Related Aspects of Intellectual Property Rights), 1869 UNTS 299.

REFERENCES Arora, A. and R. Merges (2004), ‘Specialized Supply Firms, Property Rights and Firm Boundaries’, Industrial and Corporate Change, 13, 451–75. Arora, A., M. Ceccagnoli and W. Cohen (2004), ‘R & D and the Patent Premium’, Mimeo, Carnegie Mellon University. Australian Government (2007), Australian Code for the Responsible Conduct of Research, Canberra: Australian Government.


77

Australian Law Reform Commission (2004), Report 99: Genes and Ingenuity, Canberra: Australian Government. Bar-Gill, O. and G. Parchomovsky (2005), ‘Intellectual Property and the Boundaries of the Firm’, New York University Law and Economics Working Papers, Paper 18. Benkler, Y. (2004), ‘Commons-Based Strategies and the Problems of Patents’, Science, 305, 1110–11. Biotechnology Industry Organization (2005), Biotechnology Industry Facts, http://www.bio.org/speeches/pubs/er/statistics.asp. Burk, D. and M. Lemley (2002), ‘Is Patent Law Technology-Specific?’, Berkeley Technology Law Journal, 17, 1155–1206. Burk, D. and B. McDonnell (2007), ‘The Goldilocks Hypothesis: Balancing Intellectual Property Rights at the Boundary of the Firm’, University of Illinois Law Review, 575–636. Commission of the European Communities (2005), Report of the Commission to the Council and the European Parliament: Development and Implications of Patent Law in the Field of Biotechnology and Genetic Engineering, http://www. europa.eu.int/comm/internal_market/en/indprop/invent/com_2005_312final_en.pdf. Diabetes Health (2008), ‘The History of Human Insulin, Problems, and the Loss of Choice, 1980 to 1993’ (compiled from various news sources, 1 January 2008), http://www.diabeteshealth.com/read/2008/01/01/5791.html. Eisenberg, R. (1997), ‘Genomic Patents and Product Incentives’, in B. Knoppers, C. Laberge and M. Hirtle (eds), Human DNA: Law and Policy, The Hague: Kluwer Law International, pp. 373–8. Gold, R., T. Bubela, F. Miller, D. Nicol and T. Piper (2007), ‘Gene Patents – More Evidence Needed, But Policymakers Must Act’, Nature Biotechnology, 25, 338–9. Hall, B. (2007), ‘Patents and Patent Policy’, Oxford Review of Economic Policy, 23, 568–87. Hall, B. and R. Ziedonis (2001), ‘The Patent Paradox Revisited: An Empirical Study of Patenting in the US Semiconductor Industry, 1979–1995’, RAND Journal of Economics, 32, 101–28. Heller, M. and R. Eisenberg (1998), ‘Can Patents Deter Innovation? The Anticommons in Biomedical Research’, Science, 280, 698–701. Hope, J. (2008), Biobazaar: The Open Source Revolution in Biotechnology, Cambridge, Mass: Harvard University Press. Hopkins, M., S. Mahdi, S. Thomas and P. Patel (2006), The Patenting of Human DNA: Global Trends in Public and Private Sector Activity (The PATGEN Project), Brighton: Science Policy Research Unit, Sussex University. Horn, L. (2003), ‘Alternative Approaches to IP Management: One-stop Technology Platform Licensing’, Journal of Commercial Biotechnology, 9, 119–27. Innovation Australia (2007), State of the Nation, http://www.innovationaustralia.net/ article/article.php?article=4,087. Intellectual Property Institute (2004), on behalf of the Department of Trade and Industry, Patents for Genetic Sequences: The Competitiveness of Current UK Law and Practice, http://www.dti.gov.uk/5397_DTi_Patent_Study.pdf. Jaffe, A. and J. Lerner (2004), ‘The Silent Revolution’, in A. Jaffe and J. Lerner (eds), Innovation and Its Discontents: How Our Broken Patent System Is Endangering Innovation and Progress, and What To Do About It, Princeton: Princeton University Press, pp. 96–126. Jensen, K. and F. Murray (2005), ‘Intellectual Property Landscape of the Human Genome’, Science, 310, 239–40.

78


Kaye, J., N. Hawkins and J. Taylor (2007), ‘Patents and Translational Research in Genomics: Issues Concerning Gene Patents May Be Impeding the Translation of Laboratory Research to Clinical Use’, Nature Biotechnology, 25, 739–41. Kingston, W. (2004), ‘Why Harmonisation is a Trojan Horse’, European Intellectual Property Review, 26, 447–60. Lemley, M. (2001), ‘Rational Ignorance at the Patent Office’, Northwestern University Law Review, 95, 1495–1532. Macdonald, S. (2003), ‘Bearing the Burden: Small Firms and the Patent System’, Journal of Information, Law and Technology, 1, http://www2/warwick.ac.uk/ fac/soc/law/elj/jilt. Mann, R. and T. Sager (2007), ‘Patents, Venture Capital and Software Start-Ups’, Research Policy, 36, 193–208. Merges, R. (1996), ‘Contracting into Liability Rules: Intellectual Property Rights and Collective Rights Organizations’, California Law Review, 84, 1293–1393. Merges, R. (2005), ‘A Transactional View of Property Rights’, Berkeley Technology Law Journal, 20, 1477–1525. Merges, R. and R. Nelson (1995), ‘On the Complex Economics of Patent Scope’, Columbia Law Review, 90, 839–916. Mowery, D., R. Nelson, B. Sampat and A. Ziedonis (2004), Ivory Tower and Industrial Innovation: University–Industry Technology Transfer Before and After the Bayh–Dole Act, Stanford: Stanford Business Books. National Research Council (1997), Intellectual Property Rights and Research Tools in Molecular Biology, Washington, DC: National Academy of Sciences. Nicol, D. (2005), ‘On the Legality of Gene Patents’, Melbourne University Law Review, 29, 809–42. Nicol, D. and J. Hope (2006), ‘Cooperative Strategies for Facilitating Use of Patented Inventions in Biotechnology’, Law in Context, 24, 85–112. Nicol, D. and J. Nielsen (2001), ‘The Australian Medical Biotechnology Industry and Access to Intellectual Property: Issues for Patent Law Development’, Sydney Law Review, 23, 347–74. Nicol, D. and J. Nielsen (2003), Patents and Medical Biotechnology: An Empirical Analysis of Issues Facing the Australian Industry, Hobart: Centre for Law and Genetics Occasional Paper No. 6, 2003, http://www.ipria.org/publications/ reports.html. Nicol, D. and J. Nielsen (2005), ‘Australian Medical Biotechnology: Navigating a Complex Patent Landscape’, European Intellectual Property Review, 27, 313–18. Nielsen, J. (2008), ‘Innovation in Medical Biotechnology: Mapping the Intersection of Intellectual Property and Competition Law’, Australian Business Law Review, 35(6), 407–33. OECD (2004), Patents and Innovation: Trends and Policy Changes, Paris: OECD. Pisano, G. (2006), ‘Can Science Be a Business? Lessons from Biotech’, Harvard Business Review, 84, 114–25. Pressman, L., R. Burgess, R. Cook-Deegan, S. McCormack, I. Nami-Wolk, M. Soucy and L. Walters (2006), ‘The Licensing of DNA Patents by US Academic Institutions: An Empirical Survey’, Nature Biotechnology, 24, 31–9. Rai, A. (2003), ‘Engaging Facts and Policy: A Multi-institutional Approach to Patent System Reform’, Columbia Law Review, 103, 1035–1135. Scherer, F. (1984), Innovation and Growth: Schumpeterian Perspectives, Cambridge, Mass: MIT Press. Scotchmer, S. (1991), ‘Standing on the Shoulders of Giants: Cumulative Research and the Patent Law’, Journal of Economic Perspectives, 5, 29–41.


79

Scotchmer, S. (2004), ‘Standing on the Shoulders of Giants: Protecting Cumulative Innovators’, in S. Scotchmer (ed.), Innovation and Incentives, Cambridge, Mass: MIT Press, pp. 127–59. Shapiro, C. (2001), ‘Navigating the Patent Thicket: Cross Licences, Patent Pools and Standard-Setting’, in A. Jaffe, J. Lerner and S. Stern (eds), Innovation Policy and the Economy, Cambridge, Mass: MIT Press, pp. 119–50. Stern, S. and F. Murray (2005), ‘Do Formal Intellectual Property Rights Hinder the Free Flow of Scientific Knowledge? An Empirical Test of the Anti-commons Hypothesis’, NBER Working Paper No. W11465, http://ssrn.com/abstract=755701. Stokes, D. (1997), Pasteur’s Quadrant: Basic Science and Technological Innovation, Washington, DC: Brookings Institution Press. Straus, J., H. Holzapfel and M. Lindenmeir (2002), Empirical Survey on Genetic Invention and Patent Law, Munich (copy on file with the author). van Caenegem, W. (2007), Intellectual Property Law and Innovation, Melbourne: Cambridge University Press. Verbeure, B., G. Matthijs and G. van Overwalle (2005), ‘Analysing DNA Patents in Relation with Diagnostic Genetic Testing’, European Journal of Human Genetics, 14, 26–33. Vitale, M. and D. Sparling (2003), New Biotechs Face Mixed Future, http://www.asx.com.au/investor/pdf/new_biotechs_face_mixed_future.pdf. Wadman, M. (2007), ‘Patent Crunch Pending’, Nature, 449, 651. Walsh, J., A. Arora and W. Cohen (2003), ‘Effects of Research Tool Patenting and Licensing on Biomedical Innovation’, in W. Cohen and S. Merrill (eds), Patents in the Knowledge-Based Economy, Washington: National Academies Press, http://books.nap.edu/books/0309086361/html/285.html#pagetop. Walsh, J., A. Ashish and W. Cohen (2003), ‘Working Through the Patent Problem’, Science, 299, 1021. Walsh, J., C. Cho and W. Cohen (2005), ‘View from the Bench: Patents and Material Transfers’, Science, 309, 2002–2003. Ziedonis, R. (2004), ‘Don’t Fence Me In: Fragmented Markets for Technology and the Patent Acquisition Strategies of Firms’, Management Science, 50, 804-820.

5. The jewel in the crown: India’s patent office and patent-based innovation Peter Drahos At the beginning of the 21st century Brazil, China and India are generally talked about as high growth economies, with China and India in particular set on a path that might potentially see them individually chug past the US economy in terms of size (see for example, Goldman Sachs 2003). Whatever the future holds these three countries are presently major economic powers using measures such as total GDP (see for example, World Bank 2006). All three countries have a patent law and have begun investing heavily in the creation of a large modern patent office. In this way they are following a historical pattern that held true for European powers of the late 19th century and early 20th century in which the modernization of their respective patent offices followed well after their enactment of patent law (Drahos 2007a). Of the three countries China is the most advanced down this path. Its patent office is an International Searching Authority (ISA) for the purposes of the Patent Cooperation Treaty (PCT). The patent offices of India and Brazil have only just been admitted to ISA status (PCT Committee for Technical Cooperation 2007). This chapter traces the evolution of the Indian Patent Office (PO) from its colonial origins to its modernization following India’s membership of the World Trade Organization (WTO) and therefore its obligations to comply with the Agreement on the Trade-Related Aspects of Intellectual Property Rights (TRIPs). The central argument of the chapter is that the modernization of India’s PO represents a change in thinking by Indian policy elites about the patent institution. After independence India re-designed its patents system on the basis that it was mainly a technology follower. The aim was to prevent foreigners from using patents to dominate domestic markets, especially pharmaceuticals. The modernization of the Indian PO is linked to an optimism about the strength of India’s research and development (R&D) system, a belief that Indian firms can use patents as a tool for the extraction of monopoly rents in the global technology markets. The analysis of the Indian PO presented here is part of a larger project being undertaken by the author that examines the regulatory role of patent offices in developing and developed countries. Patent offices are the neglected creatures 80

The jewel in the crown

81

of socio-legal regulatory scholarship. There is little literature on the actual workings of patent offices, something that is surprising given that very few patents will be tested by litigation. Despite the fact that the world’s patent offices grant hundreds of thousands of patents each year that shape the structure and evolution of global technology markets, we know comparatively little about the history of these offices (as opposed to the history of the patent law they administer), their administrative practices, organizational culture and levels of international cooperation and competition. This chapter provides an example of the kind of regulatory analysis of patent offices that needs to be developed.

I.

INDIA: RE-DESIGNING A COLONIAL INSTITUTION

India’s first patent law was passed in 1856 but repealed in 1857, the beginning of the Indian Mutiny, because it had not received the sanction of the Crown. After the end of the Mutiny in 1858, the British Crown assumed full power from the East India Company and patent law became part of general colonial administration. A new law was passed in 1859 (Fraser 1860, pp. 64–74). It required an inventor to petition the Governor-General of India in Council for the grant of a term of exclusive privilege (Fraser 1860, p. 65). In granting the petition the Governor-General authorized the petitioner to file a specification that described the invention and the manner of its performance. The term of the privilege was 14 years and it could be renewed for another 14. Novelty was confined to the UK and India and UK patentees could within 12 months of the grant of their UK patent petition the Governor-General of India for a grant of an exclusive privilege in India. A consolidated law for the protection of inventions and designs was passed in 1888 and administration of it shifted from the Home Department to the Department of Revenue and Agriculture. The increasing volume of petitions saw the appointment of part-time ‘patent secretaries’ to examine the specifications. In 1911, a fully revised patent law was passed that established a patent office, which began operating the following year, and a ‘Controller of Patents and Designs’. A former examiner of patents from the British Patent Office was appointed Controller. The examination of patents became mandatory but no examiners were appointed until 1919, when two were appointed. By necessity the task of examining fell to the Controller. From 1931 to 1944 there were 5 examiners, but by 1946 the number of examiners and assistant examiners had been raised to 26 (Mithal 1948, p. 64). In that year the office received 2610 applications. These were very low numbers compared with the UK, where from 1920 to 1930 the number of applications containing complete specifications had grown steadily (from 36 681 to 39 367), as had the number of examining

82


staff (from 231 to 303) (His Majesty’s Stationery Office 1931, p. 103). The low number of Indian patent applications referred to by Mithal in his 1948 article when compared with the number of applications going through the UK Patent Office suggests that pursuing patents in India was not a priority for many British patentees (Mithal 1948, p. 66). British India also formed part of a patent Empire network in which it entered into reciprocal arrangements for the protection of inventions with other Empire countries. Patent reform became a national priority soon after India became independent in 1947. Two reports on India’s patent system followed, one in 1950 which concentrated on the use of compulsory licences (the Chand Patents Enquiry Committee) and a second in 1959 (the Ayyangar Committee Report) which looked carefully at the broad question of how best to design a patent system that served India’s economic and social welfare goals (see Chand 1950; Ayyangar 1959). These two reports together triggered a long process of discussion in India about the design of a patent law that would truly serve India’s national interests and crucially the interests of its citizens. As the Indian generic industry was to point out in the early 1990s in its campaign against TRIPS, Indian consumers entered independence having to pay some of the highest drug prices in the world. For evidence the heads of influential generic companies like Dr Hamied of Cipla would point to the findings of a 1961 US Senate Committee headed by Senator Kefauver, which named India as a high price country for pharmaceuticals (see for example, Hamied 1993; Indian Drug Manufacturers’ Association 1996). The high prices in India were due to a combination of factors. During the first part of the 20th century a strong indigenous Indian pharmaceutical industry had not developed (Mueller 2007, p. 508). The Chand Patents Enquiry Committee of 1950 found that foreigners used India’s product patent regime to block local manufacture while not establishing local manufacturing facilities, preferring to rely on importation (Chaudhuri 2005, p. 29). The lack of manufacturing minimized learning by the local industry and confined it to mainly formulation of medicines. After independence India was faced with the problem that it had little in the way of indigenous pharmaceutical manufacturing capacity and at the same time the emerging Western pharmaceutical transnational corporations (TNCs) were forming cartels that raised prices in a number of developing countries, including India (Braithwaite 1984). The world-wide cartel over broad spectrum antibiotics in which companies like Pfizer and Bristol participated is a well-known example (Braithwaite 1984, pp. 175–90). Without indigenous manufacturing capacity, issuing compulsory licences, which would have been a possibility under the 1911 Indian Patents Act, was not really an option. The 1950 Chand Committee had also found that India’s compulsory licensing


83

provisions were too complicated to be able to be used by local companies to gain access to the product patents (Chaudhuri 2005, p. 29). Together the Ayyangar and Chand reports laid the foundation for the Indian Patents Act, 1970 (‘1970 Act’).1 The 1970 Act was also accompanied by a strategy that had been publicly advocated by a number of jurists, including former Chief Justices of the Supreme Court of India, of India not becoming a member of the Paris Convention (Keayla 1996, p. 7). The aim was to maximize India’s sovereignty over the design of its national patent law. The 1970 Act is a good example of the way in which a developing country can successfully re-design an institution that was imposed upon it during colonization. The monopoly privileges that the patentee obtained under the 1970 Act were nested amongst a set of regulatory levers, some of which could be pulled by government and some by private actors in order to overcome the anti-competitive effects of a monopoly. So, for example, the central government could, if it took the view that the invention was not available to the public at a reasonable price, apply to the Controller to have a patent endorsed with the words ‘licences of rights’, meaning that the patentee had to grant a licence to any person interested in working the patent (see Sub-section 86(1) and Sub-section 88(1) 1970 Act). If the parties could not agree terms the Controller had to settle the terms, but in the case of patents related to pharmaceutical process patents the royalty could not exceed 4 per cent of the ‘net-exfactory sale price in bulk’ (see Sub-sections 88(2) and (5), 1970 Act). There were also detailed provisions designed to encourage the working of inventions in India on a commercial scale, and to allow individuals to obtain compulsory licences (see Chapter XVI, 1970 Act). A separate Chapter of the Act regulated the right of the Indian government to use and acquire patented inventions (see Chapter XVII, 1970 Act). The scope of invention was carefully delineated with a considerable number of things excluded from the meaning of invention, including methods of agriculture and horticulture and processes for medicinal, surgical, curative and prophylactic treatment of human beings (see Section 3, 1970 Act). Patent claims for substances intended for use as a food, medicine or drug as well as substances produced from chemical processes could not be granted, but claims relating to the processes for manufacture of such substances were allowable (see Section 5, 1970 Act). The term ‘medicine or drug’ was defined in the widest possible way to include, for example, insecticides, germicides, intermediates in the preparation of medicines and substances intended to be used in the maintenance of public health (see Paragraph 2(l), 1970 Act). The patent term for process patents for food, medicines or drugs was reduced from 14 years to 5 years from the date of sealing of the patent or 7 years from the date of patent, whichever was the shorter period (see Paragraph 53(1)(a), 1970 Act). It is worth noting that the designers of the 1970 Act were not engaging in a

84


frolic of their own. Licences of right were a feature of UK patent law and an amendment in 1919 to the UK’s patent law had ruled out the granting of claims to substances produced by chemical processes or intended for food or medicine. The UK Patents and Designs Act, 1919 inserted Sub-section 38A(1) into the UK Patents and Designs Act, 1907. This sub-section excluded substances produced by chemical processes or intended for food or medicine from being claimed in a specification except when produced by a special method or process of manufacture. This sub-section was repealed in 1949. Licences of right remain a part of the UK system (UK Patents Act, 1977, Section 46). As Narayanan points out, the 1970 Act was ‘modelled substantially on the U.K. Patents Act of 1949’ (Narayanan 1997, p. 17). When it came to using a national patent system, India was treading along very much the same path that its former colonizer had used. A little more than 20 years after the 1970 Act, India found that path blocked. The Uruguay Trade Round of the General Agreement on Tariffs and Trade (GATT) in which India had been an active participant had concluded in December of 1993. Amongst the many agreements contained in its Final Act was TRIPs. TRIPs contained a number of important provisions on patents that led the members of the Indian generic industry to conclude that US and European pharmaceutical companies, which had been behind the drafting of these provisions, were trying to put them out of business (Drahos with Braithwaite 2000). Amongst other things, TRIPs obliged all members of the WTO to allow patents on pharmaceutical products and it prevented a country from treating one field of technology differently from other fields (TRIPs, Article 27(1)). India could not, for instance, continue with shorter patent terms for pharmaceutical processes. The Indian generic industry argued that the 1970 Act was the foundation stone of its success. Prior to its enactment the Indian pharmaceutical industry had been dominated by TNCs, with only two of the largest ten companies in terms of sales being Indian (Kettler and Modi 2001, pp. 742–3). This foreign domination of the industry was reflected in the patent figures: foreigners held 99 per cent of the 1704 pharmaceutical patents registered in India between 1947 and 1957 (Centre for Study of Global Trade System and Development 1996, p. 60). Aside from patent law the Indian generic industry also benefited from importation restrictions and taxes and price controls. Following the 1970 Act, the Indian industry grew quickly and was able to achieve export quality (Keayla 1996, p. 11). It was not just pharmaceuticals. The Indian pesticides industry, for example, was another industry that had flourished and become an export industry in the absence of patents on pesticides and chemical intermediates (Keayla 1996, pp. 11–12). Implementing TRIPs is a mandatory part of WTO membership (see Article


85

II.2 of the Agreement Establishing the World Trade Organization). India after the conclusion of the Uruguay Round faced a clear choice. It could become a member of the newly established WTO and comply with TRIPs or it could stay out of the WTO and retain its patent system. It chose the former. In the decade that followed the establishment of the WTO and the beginning of TRIPs on 1 January 1995, patent law and pharmaceuticals became a major political issue in the Indian Parliament. To some extent, the debate over patent law was a proxy for a bigger debate about India’s strategy for dealing with globalization. In the complex currents and eddies of Indian politics, neoliberal approaches to liberalization and deregulation gained much more traction in public life. The links that Nehru and others had forged in the Cold War era between Indian nationalism and socialism weakened as the deregulatory agendas of Regan and Thatcher began to influence developing country governments around the world. In 1991 the Congress Party government of P. Narasimha Rao started a period of economic liberalization in India. New interest group voices within the Congress Party and within Indian industry, such as the Confederation of Indian Industry, articulated the case for India to move towards a patent system that favoured the owners of patents, arguing that this would benefit India in terms of access to technology and foreign investment (Ramanna 2005, p. 161). Step by step India brought itself into full compliance with TRIPs. The reforms over pharmaceutical patents attracted most of the public attention and debate, but in many ways it is the changes to India’s patent office that signal most clearly the depth of change in India’s approach to patents.

II.

INDIA: THE PATENT RAJ

TRIPs contains transitional provisions for developing countries in relation to patents of which India took full advantage (TRIPs, Article 65). Part of the price of delaying the entry of patents on pharmaceuticals and chemicals under the TRIPs transitional provisions for a country was an obligation to establish a temporary system of protection. Basically this involved creating a filing system for patent applications during the transitional period and then providing the term of patent protection that remained once the patent was granted (commonly referred to as ‘pipeline protection’) (TRIPs, Article 70.8). In addition a country also had to create a temporary form of protection, known as an exclusive marketing right, for products that had been patented and had gained marketing approval in another WTO member country (TRIPs, Article 70.9). After a failed attempt to introduce implementing legislation along these lines India found itself on the receiving end of a WTO dispute resolution action brought by the US, an action that the US won (WTO 1998). India eventually

86


introduced this temporary system of protection in 1999 (Patents (Amendment) Act, 1999). In 2002 another Act amending India’s 1970 patent law was passed (Patents (Amendment) Act, 2002). This amending law brought India’s patent law into line with India’s membership of the Paris Convention and the PCT. The strategy of staying out of the international patent regime that had been forged by jurists of Nehru’s India was replaced by one that recognized in tiny detail the fact that India had joined the Paris Convention and PCT in 1998. There were some 64 amendments in total, including the lengthening of the patent term to 20 years. The strategy of shorter patent terms for process patents for medicines that had been built into the 1970 Act came to an end. A new chapter on compulsory licences was written into the 1970 Act, with the provisions on licences of right and the capping of the royalty rate for such licences in the case of pharmaceuticals being omitted. Under its TRIPs deadline India had until 1 January 2005 to bring its patent law into full compliance with the patent provisions of TRIPs. The final step meant introducing patent product protection for pharmaceuticals and chemicals. Such product protection had been a feature of India’s patent laws under the British Raj. By omitting such protection India, in a remarkably short space of time, had built a world class generics industry. Not surprisingly, the public debates in India over this last stage of TRIPs compliance were heated, even by Indian standards. In 2005 an amendment was passed that omits the restriction on the patenting of substances intended for use as a food, medicine or drug, or substances produced by chemical processes (Patents (Amendment) Act, 2005). With this law India secured its passage to the new global patent raj.

III. INDIA’S PATENT OFFICE: THE JEWEL IN THE CROWN Mithal’s description of the Indian PO at the beginning of the 20th century shows clearly that it was a small operation (Mithal 1948, p. 62). The Patent Controller who was appointed in 1912 did not have any examiners to assist him, despite examination being compulsory. In the year of his resignation (1919) the number of applications had reached 1000. If we assume that the processing, examining and granting of one application took five hours (obviously this would allow for only a cursory examination) and the Patent Controller worked every day of the year and did two applications a day, he would have been unable to process a 1000 applications in one year. Things seemed to have improved by 1946, when there were about 26 examining staff for the 2610 applications received that year, but Mithal gives no information


87

about the size of the backlog. The modernization of the Indian PO was probably not a high priority during the British Raj. After the 1970 Act came into operation in 1972 the amount of patenting by foreign companies in India dropped – dramatically so. Bagchi’s study in 1984 showed that in 1968 there were 1110 applications by Indians and 4248 applications by foreigners. By the financial year 1979–80 the number of Indian applications was not much different (1055) but the number of foreign applications had dropped to 1925. The number of foreign patents in force in India had dropped from 37 816 in 1968 to 14 476 in 1979–80 (Bagchi with Banerjee and Bhattacharya 1984, p. 293). We saw earlier just how widely the 1970 Act defined medicines and drugs for the purpose of excluding such things from patentability as products, as well as reducing the patent term for processes in these fields. The most obvious explanation for the drop in the number of foreign patent applications is that prior to the 1970 Act, the bulk of patenting going on in India by foreign interests was related to pharmaceuticals and chemical products in one way or another, something the wide exclusion under the 1970 Act largely stopped. In addition India had raised its standard of novelty to include documentary publication anywhere in the world and raised its patent fees (1970 Act, Sub-section 13(2); see also Bagchi with Banerjee and Bhattacharya 1984, p. 293). The drop in patent applications after the 1970 Act would have provided no incentive to modernize the Indian PO. Things changed with India’s membership of the WTO. Foreign companies in the pharmaceutical/chemical field began to show an interest in patenting in India, even before India had implemented the temporary pipeline protection that was required under TRIPs and well before it moved to product protection in 2005. In the WTO action that the US brought against India for failing to implement pipeline protection, the evidence showed that between 1 January 1995 and 15 February 1997 the Indian PO had received 1339 applications for pharmaceutical and agricultural products, despite the fact that it only had in place an obscure, unpublicized procedure for dealing with patent applications intended for pipeline protection (WTO 1998). These applications were an early signal that India could expect to receive many of these applications and that the PO would have to have proper procedures in place for their processing. Pipeline protection under TRIPs was only ever going to be a temporary source of work for the Indian PO. The real source of work was always going to be the PCT. When India joined the PCT foreigners began to use it to obtain patents in India. Foreigners could designate India in an international application under the PCT as from 7 December 1998. India could only act as a receiving office for PCT applications. Table 5.1 shows that foreigners prefer to use the PCT to file for patents in India rather than filing directly at the Indian PO or using Paris Convention priority (15 467 represents PCT foreign filings in

88


Table 5.1

The PCT route in India

Total number of applications Number of national phase PCT applications Residents Non-residents

1999–2000

2005–2006

4824 269

24 505 15 467

2206 2349

4521 4517

which a PCT application has entered the national phase and 4571 represents foreigners using other routes to the Indian PO) (Indian Patent Office 2006, p. 18). Applicants from the US, Germany and Japan are the biggest users of the PCT route to India (Indian Patent Office 2006, p. 10). Of the 24 505 patent applications that the Indian PO received in 2005–06, 33 per cent were for chemicals, drugs and food. For 2005–06, the number of patent applications was: chemical – 5810; drug – 2211; food – 101 (Indian Patent Office 2006, p. 10). If one adds biotechnology applications to this group this percentage climbs to 39 per cent (Indian Patent Office 2002, p. 3). Given the low number of patent applications by Indian residents, foreigners would be mainly responsible for most of the patent applications in these sectors. India’s re-integration into the international patent regime through TRIPs has, for the time being at least, produced a situation not unlike the one that prevailed under the British Raj in which much of the patenting was in the drug–chemicals area, with almost all the patents in foreign hands. Reversing this domination through patents was the purpose of the 1970 Act.

IV.

MODERNIZATION

The increasing number of patent applications that India began receiving as a result of its WTO membership made it clear that it would have to modernize its patent office. Modernization was approved in 1998 but work did not start until 2000 (Indian Patent Office 2002, p. 3). Amongst other things, this meant hiring more examiners, building new office space and investing heavily in information technology. The Indian PO is made up of the Head Office in Kolkata with branch offices in Chennai, Delhi and Mumbai. I visited the Mumbai Office in February 2005. At first I thought that the taxi driver had taken me to a derelict building and so an animated discussion ensued in which the taxi driver assured


89

me that the PO was indeed located in the building. Refusing to pay, I went to check this. Picking my way through the hanging wires and pigeons in the dimly lit stairwell I came upon the sign of the Mumbai PO on the third floor. Hurrying back to the taxi driver I offered an apology and a very healthy tip, both of which were politely accepted. The rest of the day was spent talking to examiners in rooms that had not seen paint for a while. Today, if the photo in the 2005–06 annual report of the Indian PO is accurate, the Mumbai branch has all the glass and concrete that one would wish upon a modern building. That same annual report announces that India has completed the modernization of all its intellectual property (IP) administration. Keeping in mind that the Indian PO has achieved ISA status, the changes in the Mumbai Office have been dramatic since the time of my visit, when in essence I saw a branch office of the Indian PO in a pre-modern state. In 2005 the Mumbai branch had about 30 examiners and was planning to recruit another 50. Two years prior it had only three examiners. Recruiting examiners has not proved a problem, but retaining them, an Assistant Controller suggested, is more of an issue because of poor public sector salaries. A recurring theme amongst the examiners was the lack of access to databases for the purpose of doing proper searches. They were using free patent searching tools available on the European Patent Office (EPO) website (esp@cenet was expressly mentioned) and the US Patent and Trademark Office’s website. The lack of databases, they pointed out, made it difficult to check the work of other offices (‘our databases are of no use at the moment’, as one of them put it). This in turn meant that they had to give credence to the examination reports being produced by ISAs. Inferior databases heighten the chances that the examiner will adopt a play-it-safe strategy and grant the patent: ‘If you reject [the] patent you must be correct. Easier to grant patent’ (Indian examiner, Mumbai branch). The examination reports of large offices like the EPO and the USPTO were described as a useful starting point. As one Assistant Controller put it: ‘we are getting expertized by reading those reports’. But at the same time the Indian examiners were reading the reports of other offices critically, even if they were in an inferior position when it came to searching. Why, they asked me, did the Australian IP office grant claims that other offices did not? Caught by surprise I said I didn’t know, adding as an afterthought that perhaps they want to attract the work (this was followed by nods of agreement). Reaching ISA status was a clear priority for the Indian PO because it increases its capacity to influence other offices and provides it with more autonomy within the patent regime. Given that India is now an ISA, the situation with respect to databases must have improved since the time of my interview. India has a Memorandum of Understanding (MOU) with the EPO,

90


signed in 2006, in which the EPO agrees that it may give India access to its search engine EPOQUE (India’s MOU with EPO 2006). This would give Indian examiners access to the EPO’s subscription databases. These databases are regarded by patent aficionados as being the best in the world. The Indian PO returns the fees it collects to general revenue. Fees have been raised in the last couple of years in order to cover its running costs. As one interviewee put it: ‘we are moving you can say into profit like a company’. It is also facing a growing backlog of patent applications. In 1992–93, there were almost 8000 patent applications to be carried forward to the next financial year and in 2001–02 that number had grown to more than 44 000 (Indian Patent Office 2002, p. 39). The strategy for dealing with the backlog is based on a system of examination upon request and increasing the number of patent examiners. A person has four years from the date of filing or priority of the application within which to request an examination. Moving to a model of deferred examination in order to deal with a growing number of applications is a strategy that has been tried in other countries, including Germany and Japan, but ultimately it only defers the problem of processing large numbers of applications. The other strategy is to increase the number of examiners. During the interview in Mumbai, India’s target for the number of examiners was said to be 270. According to the 2005–2006 annual report the Indian PO has a working strength of 179 made up of 140 examiners of patents (and designs) and another 39 staff occupying more senior positions of Controller and Assistant Controller of patents. The success of this strategy obviously depends on recruitment and retention of examiners.

V. ‘HOW CAN WE WIN?’ – INDIA’S STRATEGY ON PATENTS The reform of India’s patent law and the subsequent modernization of its patent office were triggered by TRIPs. But in implementing these changes new policy voices within India have pushed for a more intimate embrace of the patent system. So, by way of example, Dr Rakesh Mohan, Secretary, Department of Economic Affairs in the Union Finance Ministry, has suggested that India’s new product patent rules will encourage the outsourcing of chemical and pharmaceutical R&D to India.2 Senior people in the Indian PO take a similar view: ‘Indian researchers do research abroad for US companies. They can easily do that research here in India’ (interview with Assistant Controller of Patents, Indian PO, 8 February 2005). Underpinning this thinking is India’s large pool of low cost but highly trained scientific R&D workers (UNDP 2005, p. 264). Chemical and pharmaceutical multinationals will be encouraged to


91

establish their R&D laboratories in India in order to take advantage of this cheap labour force. R&D centres will spring up like call centres. There is also the belief that India’s existing R&D system is strong enough to be able to support the extraction of patent rents from other countries. At a meeting in 1996 organized by the Department of Scientific and Industrial Research (DSIR), ministers and officials drew attention to the fact that at that time India had 325 national laboratories and R&D institutions, 450 scientific and industrial R&D organizations, and 1250 in-house R&D units collectively employing around 50 000 scientific and technical personnel. The problem lay in the lack of patents. One survey by DSIR showed in the previous four years these 1250 in-house units had only filed 743 patents, a ‘miniscule’ effort compared with other powers, including China (Department of Scientific and Industrial Research 1996a, p. 25). The Minister of State made a similar argument pointing out that India had changed its position on patents because it was time for India ‘to use the industrial base built up mainly for the domestic market over the preceding 30 years, to move out into world markets’ (Department of Scientific and Industrial Research 1996b, p. 34). These kinds of views are not surprising. After all, one of the lessons of colonization that India would have learnt under the British Raj is that the patent system is a rent extraction machine, which used correctly allows domestic firms to dominate foreign markets. The reference to China is telling. Once China had signalled that it was reforming its patent system in order to play the patent-based innovation game, Indian policy elites would have concluded that they had little choice but to be in a patent arms race. The optimism of India’s policy elites about India’s bright patent-based future had not made it down to the 30 or so patent examiners in the Mumbai branch at the time of my interview in 2005. In summary, their views were that the globalization of patent standards put developing countries like India in a difficult position. The local pharmaceutical industry was not in a position, they thought, to do sufficient R&D to take advantage of the new system. Many Indian companies could not afford the cost of patent applications. Inequalities of national incomes meant that richer countries were in a better position to afford the costs of patents in India. The system did not benefit small to medium sized enterprises and even the Indian pharmaceutical industry could not invent new molecules. ‘How can we win?’, they kept asking.

VI.

FROM BRIGHT LINE RULES TO SHADES OF GREY

One of the features of the amended 1970 Patents Act is that it contains more complex rules. Complexity is the happy hunting ground of lawyers. A rule that simply excludes a drug from patentability is readily understandable and leaves

92


comparatively little scope for argument. Similarly, a licence of right system, of the kind that India had in its 1970 Act for pharmaceuticals, automates the process of obtaining a licence from the patentee and sets a cap on the royalty. This leaves very little room for litigation strategies to defeat the grant of a licence. The amendments to the 1970 Act are more complex rule formulations that invite litigation. For example, under the 1970 Act, Section 5 simply stated that ‘in the case of inventions claiming substances intended for use … as medicine … no patent shall be granted in respect of claims for the substances themselves’. Moreover, the definition of medicines was cast in the widest possible terms, leaving little possible doubt about the scope of its operation. The 2005 amendment that India passed to the 1970 Act omits Section 5 but inserts a revised Paragraph 3(d), the section that deals with exclusions from the meaning of invention under the Act. This paragraph excludes: ‘the mere discovery of a new form of known substance which does not result in the enhancement of the known efficacy of that substance’. The aim is to prevent pharmaceutical companies from recycling for patentability known molecules in a new form that deliver no real therapeutic gain by excluding such a practice from the meaning of invention. The logic in essence is to stop the same invention from being patented again. All jurisdictions have a rule against double patenting.3 What is open to interpretation and left open to interpretation by TRIPs is what counts as the same invention and this is the gap that Paragraph 3(d) aims to fill in the case of pharmaceuticals. An explanatory note adds that for the purposes of Paragraph 3(d) salts, esters, polymorphs, isomers and other things will be considered to be the same substance ‘unless they differ significantly in properties with regard to efficacy’. The rule in Paragraph 3(d) is a more complex rule of exclusion because it depends on the meaning of ‘enhancement of the known efficacy of that substance’ and its operation is also affected by an explanatory note that sets out a test for chemical sameness. In the first instance an examiner will have to interpret this exclusion since a patent can only be granted for inventions within the meaning of the Act. The Indian generic industry, like other national generic industries around the world, has long complained of the practice of evergreening in which pharmaceutical multinationals find ways to patent slightly revised versions of old molecules.4 Indian generic companies will be looking to the Indian PO to interpret the exclusion widely since this favours market entry and competition. Pharmaceutical multinationals will be looking to the Indian PO to read this exclusion in the narrowest possible way since this favours exclusivity and monopoly rents. The decision of the Indian PO in which it rejected Novartis’ patent application for its brand name drug Gleevec on the basis that it was caught by Paragraph 3(d) is precisely the decision that brand name manufacturers fear. Novartis’ price for Gleevec in India was around the US$3000 mark.5


93

Indian generic manufacturers were selling a generic version for US$200 (Bidwai 2007). Both groups will in essence want to create a supportive interpretive community that reads patent applications in a way that more often than not will produce the correct result (as defined by their business model). The patent interpretive community is bigger than just patent examiners. It includes judges and patent attorneys. Moreover, examiners are at the bottom of the hierarchy of authoritative interpretation. It is courts that have the final say as to patentability: decisions of the Controller of Patents can be appealed to a High Court of India, including a decision of the Controller to refuse an application on the ground that it is not for an invention under the Act (Patents Act 1970). But in practical terms patent examiners are, as it were, the ‘grantkeepers’. It is their decisions that often end up determining whether or not a patent is to be granted. Very few decisions of the patent office end up being contested in the courts. One estimate suggests that between 1996 and 2000 less than 100 decisions of the Indian PO were appealed to the Indian courts (Basheer 2005, p. 320). The interpretive examination culture of a patent office matters, crucially so, because in almost all cases the final decision about patentability will be made by an examiner and not a court. The vast patent portfolios that TNCs build globally are built on the backs of examiners’ decisions. Developing countries like India that are integrating into the international patent regime present opportunities for the creation of new interpretive communities within patent offices. To begin with, developing country offices themselves create a demand for this training in interpretation, especially in areas such as pharmaceutical product patent protection where they lack experience. During the interview in the Mumbai Office many of the examiners kept coming back to the point that they needed training, especially in the area of pharmaceuticals. Offices like the European Patent Office and the USPTO offer this training (Drahos 2008). But at the same time as they do, they build an interpretive community. They instruct developing country examiners in the interpretation of claims and the making of judgements about the level of inventive step required to satisfy the inventiveness criterion. In the case of India, the USPTO has been prepared to help out by training Indian examiners in the nuanced art of pharmaceutical product examination. At an interview in the USPTO, US officials after drawing attention to this capacity-building exercise noted that it had an element of controversy about it (Drahos 2007b). They were hinting at the fact that Indian novices, after exposure to US patent examination culture, which itself is based on decades of experience with pharmaceutical product patents, may come to read Paragraph 3(d) through the lens of that US experience. And as we saw earlier, there is also the fact that the US is the single biggest source of patent applications in India and pharmaceutical–chemical

94


patenting represents, for the time being, the single biggest category of patenting in India. The USPTO, much like the EPO, builds capacity in developing countries that favours the patenting strategies of its respective companies. It is a form of encoded capacity building in which the aim is to form an interpretive community through patient years of assistance and training. The fact that brand companies lose some cases as Novartis did with Gleevec, although no doubt heart-wrenching for Novartis, is beside the point. Rather, the point is to create a dense technical jurisprudence of rules in which a community of examiners deeply familiar with those rules ultimately comes to ‘see’ that, for example, changing the particle size of a known drug so as to improve its bioavailability is sufficient to satisfy both the efficacy requirement of Paragraph 3(d) and the inventive step requirement.6 Examiners are not, in the process of technical capacity building, where they are turned into automatons of interpretation in which the right outcome is implanted in them. Rather, the technical training opens up for them a world of characterization possibilities in which inventive steps, which from a scientific viewpoint are minor or even trivial, can been seen to be inventive for the purposes of the patent system. Wittgenstein’s philosophical aphorism, ‘the limits of my language, mean the limits of my world’ has real-world grip. By deepening an examiner’s knowledge of patent language, as it were, the examiner can begin to see the meaning of invention in a different way. Individual examiners will have professional disagreements about how these characterization possibilities are to be drawn in the case of a given application, but the aim of encoded capacity building is not to stop this. These kinds of debates over the application of technical rules help to create and maintain a technical rulesbased discourse. Instead the aim is to foster the growth of an interpretive culture that (following Wittgenstein’s idea) breaks the ordinary language limits on invention and creates a new world in which scientifically trivial innovation becomes patentable invention.

VII. CONCLUDING OBSERVATIONS India’s entry into TRIPs and the modernization of its patent office since 2001 suggests that it sees the possibility of much greater gains for itself from the system than in the past. Classified as a mega diverse country, it wants a system that helps it to track patent applications that makes use of its genetic resources. It was a key mover in submitting draft amending text on TRIPs in the WTO’s Doha Round that would require patent applicants to disclose the origin of genetic resources over which they were seeking a patent (the proposed amendment takes the form of a new article in TRIPs – Article 29bis) (WTO, 2006).


95

Similarly, it has been an advocate of the digitization of traditional knowledge resources for the purposes of ensuring that such knowledge forms part of the prior art for the purposes of patent law. The creation in India of a Traditional Knowledge Digital Library led in turn to the development of a Traditional Knowledge Resources Classification system so that traditional knowledge resources could be incorporated into the International Patent Classification (IPC) system.7 India’s democracy is also vigorous enough to allow passionate public debates to influence patent reform. Patent reform is not a quiet technocratic exercise carried on behind closed doors. My interviews at the Mumbai PO suggested that senior officials were not in favour of retaining India’s pre-grant opposition system, preferring instead to rely on post-grant opposition. The move to abolish pre-grant opposition is favoured by multinationals. It was a feature of India’s Patent Act in 1970. It had been a part of English law going back to the 1850s, the UK eventually removing it when it joined the European Patent Convention (see Clause XII, UK Patent Law Amendment Act, 1852: this clause allowed any person with an interest in opposing the patent to object in writing). Despite a trend for countries to move away from pre-grant opposition, India has retained its pre-grant opposition and added post-grant opposition. Some of India’s non-governmental organizations have started using pre-grant opposition in the pharmaceuticals area and the Indian generic industry has launched many such oppositions.8 Since 2005 domestic pharma companies have filed nearly 150 pre-grant oppositions to patent applications in India (Economic Times 2006). It is also clear that India is paving the way for the entry of many more patents into its business culture. It has, for example, introduced a fast-track procedure that will see a patent issue in India in less than six months (this fasttrack process is started by a request for early publication under Section 11A(2) of the amended 1970 Act. A request for examination has to be dealt with within three months. Not all companies will necessarily want to fast-track their applications, but the option exists. For TNCs, the cost of patent fees in India would represent a small proportion of the overall cost of their global patenting fees. At the time of the interview in 2005, Indian patent officials put the patent fees for a standard patent at around US$300 and estimated patent attorney fees to be about US$2000 for a standard application. It is also reasonably clear that the more complex rule orientation of the 1970 Act will simply attract more litigation. Companies like Novartis are unlikely to sit meekly by while the Indian PO issues its rejections of their patent applications in pre-grant oppositions. These TNCs know that the meek shall not inherit the riches of the pharmaceutical kingdom. Indian patent examiners will be given more training by the USPTO and Indian courts will find

96


themselves hearing more cases until a more favourable line of interpretation emerges, unless of course the Indian Parliament is persuaded to fix the operation of Paragraph 3(d) by the usual army of lobbyists. Similarly, India’s amendment to Section 3 in 2002 to exclude from the meaning of invention ‘a mathematical or business method or computer program per se or algorithms’ will see a voluminous patent theology develop around the meaning of ‘per se’, much as it has around the words ‘as such’ in the context of the exclusion of computer software programs in the European Patent Convention (EPC).9 More patenting in India and more patent litigation mean that Indian companies will need access to legal services, including firms doing patent searches, patent drafting work, patent advising, opposition and litigation work. Patent-based innovation requires high levels of legal servicing. After its entry into the WTO and TRIPs some Indian policy makers drew attention to the fact that India lacked sufficient numbers of people skilled in these areas (Department of Scientific and Industrial Research, 1996c, p. 20). Ten or so years down the track it is hard to say to what extent the situation has changed. People within the Mumbai PO were not flattering about the quality of work of the local patent profession, the exception being those legal firms that worked for foreign TNCs. It is a reasonable conjecture that the best patent attorneys in India are likely to be those that work for TNCs since those companies pay the most. Local inventors are, for the most part, left with those that are less able to work the system in favour of their clients. This is a problem for local firms because so much of what ends up in a granted patent is dependent upon the drafting skill of the patent attorney. The great skill of good patent attorneys is to be able to listen to inventors tell their stories of discoveries, take that raw narrative and turn it into something that after two or so years will be accepted by a patent office and sealed as a patent. One examiner put it thus: ‘The [patent] agent is inventing for the inventor.’ Developing countries like India, which want to play in the patent-based global economy, need armies of patent attorneys, just as much as they need scientists, since what counts is not invention but patentable invention. As the pharmaceutical industry has routinely demonstrated, one does not need genuine invention to obtain a lucrative patentable invention. The small step of changing the particle size of a known drug in order for it to be turned into a patentable invention requires some clever drafting of patent claims, a compliant patent office that understands that trivial innovation is patentable innovation and doctors willing to prescribe the patentable version of the drug rather than the generic. India in raising a patent attorney army has the advantage of English and a large enough market and population in which such differentiated legal services can evolve. Few other developing countries can raise the patent armies needed


97

to set up, defend and extract patent monopoly rents. India will probably have a much greater depth in these kinds of services in the next decade. Can then the patent-based innovation strategy that India is developing work? Bagchi’s 1984 study of India’s 1970 Patents Act came to the conclusion that it had not really delivered in terms of innovation in India, but that this was not the fault of the law (Bagchi with Banerjee and Bhattacharya 1984). Rather, it had to do with R&D investment by India in unproductive sectors like nuclear power and space, as well as failure to create networks between public research laboratories and India’s industrial base. Probably by the early 1990s, prior to TRIPs coming into operation in India, one could say that the Indian generic industry was a success story attributable in part to the operation of the 1970 Act. But here one has to keep in mind Bagchi’s point that India was primarily a follower of technology and so the purpose of its 1970 patent law was to minimize foreign patenting restrictions on Indian firms (Bagchi with Banerjee and Bhattacharya 1984, p. 300). By the time of India’s entry into TRIPs, India’s policy elites wanted an Indian patent system that would, in the words of Dr Mashelkar, turn India’s ‘intellectual prowess into knowledge and wealth’ (Department of Scientific and Industrial Research 1996d). It would, in other words, be Indian firms owning the patent monopolies and extracting the global rents. India’s reform of its patent office has been driven by this kind of thinking. To date all that has happened is that foreign patenting levels have risen dramatically, especially in the areas of pharmaceuticals and chemicals, the areas which the 1970 Act had excluded from patentability. India’s policy elites would no doubt reply that this is a long-run strategy. But the only certain thing about the long run, as John Maynard Keynes once famously observed, is that ‘we’re all dead’.

NOTES 1. 1970 Act: references to the Indian Patents Act 1970 are references to the Patents Act, prior to its amendments as a result of TRIPs. 2. See http://www.ipai.in/product-patent.htm. 3. See for example, Eli Lilly and Co v. Barr Laboratories Inc., 251 F.3d 955 (2001). 4. On the views of the Indian Drug Manufacturers’ Association see Report of the Technical Expert Group on Patent Law Issues (2006, p. 23). Complaints about evergreening have been raised in other countries too (Hore 2004; Somogyi, Bochner and Foster 2004; Kesselheim 2007). 5. Novartis AG Switzerland v Union of India, through the Secretary of Department of Industry, Ministry of Industry and Commerce, New Delhi and others (2007) 4 MLJ 1153 (Mad), 1177. 6. For an unsuccessful attempt by Novartis to argue this in relation to the known drug oxacarbazepine see the pre-grant opposition filed against Novartis by Ranbaxy Laboratories and Torrent Pharma held on 16 November 2006 and decided on 4 January 2007. 7. The eighth edition of the IPC contains classification A61K 36/00, which is described as ‘Medicinal preparations of undetermined constitution containing material from algae, lichens, fungi or plants, or derivatives thereof, e.g. traditional herbal medicines’ (see

98

8.

9.

Intellectual property policy reform http://www.wipo.int/classifications/ipc/ipc8/?lang=en); on the history of the creation of the Traditional Knowledge Digital Library see Gupta (2005). For a description of India’s and WIPO’s roles in the IPC process related to traditional knowledge, see World Intellectual Property Organization (2004). A pre-grant opposition has been filed against the anti-HIV drug, Combivir, manufactured by GlaxoSmithKline, a leading pharmaceutical company. The Manipur Network of Positive People (MNP+) and the Indian Network of People Living with HIV/AIDS have lodged the complaint at the Kolkata Patent Office; see http://www.medindia.net/news/view_news_ main.asp?x=8969&t=1. On the different approaches in the context of the EPC, see Aerotel Ltd. v Telco Holdings Ltd & Ors, Rev. [2006] EWCA Civ 1371.

REFERENCES Ayyangar, Rajagopala (chair) (1959), Report on the Revision of the Patent Law (Information provided by the Indian PO, Mumbai branch, 8 February 2005). Bagchi, Amiya Kumar with Parthasarathi Banerjee and Utlam Kumar Bhattacharya (1984), ‘Indian Patents Act and Its Relation to Technological Development in India: A Preliminary Investigation’, Economic and Political Weekly, 19(7), 18 February 287–302. Basheer, Shamnad (2005), ‘“Policy Style” Reasoning at the Indian Patent Office’, Intellectual Property Quarterly, 3, 309–320. Bidwai, Praful (2007), ‘Health-India: Novartis Patents Case Far From Dead’, at http://ipsnews.net/news.asp?idnews=38840, 9 August 2007. Braithwaite, John (1984), Corporate Crime in the Pharmaceutical Industry, London, Boston, Melbourne and Henley: Routledge and Kegan Paul. Centre for Study of Global Trade System and Development (1996), ‘Peoples’ Commission on GATT: On the Constitutional Implications of the Final Act Embodying the Results of The Uruguay Round of Multilateral Trade Negotiations’, New Delhi. Chand, Bakshi Tek (chair) (1950), Report of the Patents Enquiry Committee (Information provided by the Indian PO, Mumbai branch, 8 February 2005). Chaudhuri, Sudip (2005), TRIPs and Changes in Pharmaceutical Patent Regime in India, Calcutta: Indian Institute of Management, Working Paper No. 535, January. Department of Scientific and Industrial Research, Government of India (1996a), ‘Welcome Address by Shri Ashok Parthasarathi, Additional Secretary, DSIR’, in Proceedings of the 9th National Conference on In-House R&D Industry: The New Intellectual Property Regime: Implications for Industry and R&D, New Delhi. Department of Scientific and Industrial Research, Government of India (1996b), ‘Inaugural Address by Shri Bhuvnesh Chaturvedi, Minister of State (Prime Minister’s Office and Science & Technology’, in Proceedings of the 9th National Conference on In-House R&D Industry: The New Intellectual Property Regime: Implications for Industry and R&D, New Delhi. Department of Scientific and Industrial Research, Government of India (1996c), Proceedings of the 9th National Conference on In-House R&D Industry: The New Intellectual Property Regime: Implications for Industry and R&D, New Delhi. Department of Scientific and Industrial Research, Government of India (1996d), ‘Text of Address by Dr R.A. Maskelkar, Secretary, DSIR and DG, CSIR’, in Proceedings of the 9th National Conference on In-House R&D Industry: The New Intellectual Property Regime: Implications for Industry and R&D, New Delhi.


99

Drahos, Peter (2007a), ‘Patent Reform for Innovation and Risk Management: A Separation of Powers Approach’, Knowledge Ecology Studies, 1, 1–11. Drahos, Peter (2007b), Interview at the USPTO, Virginia, USA, 19 March 2007. Drahos, Peter (2008), ‘“Trust Me”: Patent Offices in Developing Countries’, American Journal of Law and Medicine, 34, 151–74. Drahos, Peter with John Braithwaite (2000), Information Feudalism, London: Earthscan. Economic Times (2006), see http://blog.ipfactor.co.il/2006/10/25/indian-genericpharma-companies-delay-patent-protection-by-pre-grant-opposition-proceedings. Fraser, James (1860), A Handy-Book of Patent and Copyright Law: English and Foreign, London: Sampson Low, Son and Co. Goldman Sachs (2003), ‘Dreaming with BRICS: The Path to 2050’, Global Economics Paper No. 99, available at www2.goldmansachs.com/ideas/brics/book/99dreaming.pdf. Gupta, V.K. (2005), ‘Traditional Knowledge Digital Library’, Sub-Regional Experts Meeting in Asia on Intangible Cultural Heritage: Safeguarding and InventoryMaking Methodologies, Bangkok, Thailand, 13–16 December. Hamied, Y.K. (1993), ‘Patents and the Pharmaceutical Industry: A Review’, paper for the International Conference on Patent Regime Proposed in the Uruguay Round, 2–4 September, New Delhi. His Majesty’s Stationery Office (1931), Report of the Department Committee on the Patents and Designs Acts and Practice of the Patent Office, London: HMSO. Hore, Edward (2004), ‘Patently Absurd: Evergreening of Pharmaceutical Patent Protection under the Patented Medicines (Notice of Compliance) Regulations of Canada’s Patent Act’, available from the Canadian Generic Pharmaceuticals’ Association, Suite 409, 4120 Yonge Street, Toronto, Ontario, Canada. Indian Drug Manufacturers’ Association (1996), Submission by IDMA before International Conference organized by Forum of Parliamentarians on Intellectual Property and National Working Group on Patent Laws at New Delhi, 14–15 November. Indian Drug Manufacturers’ Association (2006), Report of the Technical Expert Group on Patent Law Issues, New Delhi, December. Indian Patent Office (2002), Annual Report 2001–2002, India: Office of the ControllerGeneral of Patents, Designs, Trade Marks and Register of Geographical Indications. Indian Patent Office (2006), Annual Report, 2005–2006, India: Office of the Controller-General of Patents, Designs, Trade Marks and Register of Geographical Indications. India’s MOU with EPO (2006), Indian Patent Office and European Patent Office, signed 29 November, http://ipindia.nic.in/ipr/patent/patents.htm. Keayla, B.K. (1996), ‘New Patent Regime: Implications for Domestic Industry, Research and Development and Consumers’, National Working Group on Patent Laws, New Delhi, http://www.patentoffice.nic.in/ipr/patents.htm. Kesselheim, A. (2007), ‘Intellectual Property Policy in the Pharmaceutical Sciences: The Effect of Inappropriate Patents and Market Exclusivity Extensions on the Health Care System’, American Association of Pharmaceutical Scientists Journal, 9(3), E306. Kettler, Hannah and Rajiv Modi (2001), ‘Building Local Research and Development Capacity for the Prevention and Cure of Neglected Diseases: The Case of India’, Bulletin of the World Health Organization, 2001, 742–3. Mithal, Ved P. (1948), ‘Patents in India’, Journal of Patent Office Society, 30, 62.

100


Mueller, Janice M. (2007), ‘The Tiger Awakens: The Tumultuous Transformation of India’s Patent System and the Rise of Indian Pharmaceutical Innovation’, University of Pittsburgh Law Review, 68, 491–508. Narayanan, P. (1997), Intellectual Property Law, second edition, Calcutta: Eastern Law House. PCT Committee for Technical Cooperation (2007), PCT/CTC/23/5, 12 November 2007, http://www.wipo.int/edocs/mdocs/govebody/en/pc-ctc-23-5.doc. Ramanna, Ranitha (2005), ‘Shifts in India’s Policy on Intellectual Property: The Role of Ideas, Coercion and Changing Interests’, in P. Drahos (ed.), Death of Patents, London: Lawtext Publishing, pp.150–61. Somogyi, A., Bochner, F. and Foster, D. (2004) ‘Inside the Isomers: The Tale of Chiral Switches’, Australian Prescriber, 27(2), 24. UNDP (2005), Human Development Report, New York: UNDP. World Bank (2006), GDP (adjusted using purchasing power parity factors) for 2006 (China ranked 2, India 4 and Brazil 10), http://web.worldbank.org/WBSITE/ EXTERNAL/DATASTATISTICS/0,,contentMDK:20399244~menuPK:1504474~p agePK:64133150~piPK:64133175~theSitePK:239419,00.html. World Intellectual Property Organization (2004), Classification Tools Relating To Traditional Knowledge and Biodiversity, Committee of Experts, Special Union of the International Patent Classification, IPC/CE/35/7, Geneva: WIPO, 21 September. WTO (1998), ‘Report of the Panel, India – Patent Protection for Pharmaceutical and Agricultural Chemical Products, WT/DS50/R, adopted 16 January 1998, as Modified by the Appellate Body Report’, WT/DS50/AB/R, http://www.wto.org/ English/tratop-e/dispu-e/cases-e/ds50-e.htm WTO (2006), World Trade Organization Trade Negotiations Committee, a New Article in TRIPs – Article 29bis, see document WT/GC/W/564/TN/C/W/41 submitted to the Trade Negotiations Committee on 29 May by Peru, Brazil, India, Pakistan, Thailand and Tanzania, Geneva: WTO.

6. The first steps in remedying the relationship between patents and competition Charles Lawson I.

INTRODUCTION

Modern statutory competition laws such as the Australian Trade Practices Act 1974 (Cth) (‘Trade Practices Act’) evolved out of the common law, probably because of a failure of the private right of action to sustain the broader public interests in vibrant competition and free trade (see Donald and Haydon 1978, pp. 2–4; Letwin 1954). Together with this evolution has been the development of the idea that markets, and the competitive forces operating independently in those markets, are best placed to make decisions about how societies’ scarce resources might be allocated for the benefit and desires of the majority of consumers (see Hilmer 1993, pp. 1–6). As a cornerstone of modern market economies like Australia, competition laws are now considered necessary to drive innovation delivering better quality products and services to consumers at lower prices by improving productivity and economic efficiency (being technical or productive efficiency, allocative efficiency and dynamic efficiency: see, for example, Hilmer 1993, pp. 3–5). The rationale of government regulation in implementing competition laws is to keep modern market economies open for competition by limiting conduct that is detrimental to competition (Hilmer 1993, pp. 1–16). Importantly though, these regulations should be directed to protecting the competitive process rather than favouring particular sectors of the economy (Hilmer 1993, p. 26). However, a tension remains in some dealings with patents under the Patents Act 1990 (Cth) (‘Patents Act’) where the statutory patent privilege results in monopoly pricing or restricts access to the patent protected product or process more than the social advantage secured by the property right (see Landes and Posner 2003, pp. 372–85; Scotchmer 1991, pp. 30–32; see also, for examples, Gifford 2003, pp. 1695–718; Carrier 2002). Significantly, the privileges granted by patents under the Patents Act are of a different character to other forms of property (see Ergas 2000, p. 11), essentially being a right to the sale of the idea in the patent protected product or 101

102


process, together with a statutory privilege to control how the patent protected product or process is used (Boldrin and Levine 2004, p. 328). It is this control of downstream uses of the patent protected product or process, and its leverage to achieve other business purposes, that is the concern of competition law. But how far does that concern extend? Commonly, the view is that it is only when a dealing (or a refusal to deal) with the patent protected product or process lessens competition in the ‘market’ that competition laws become relevant. That is, where a patent holder refuses to deal, imposes anti-competitive conditions in proposed license agreements, or otherwise acts (anti-competitively) in a way that would not be possible without some form of market power. Australia provides a case study of the interaction between patent laws and competition regulations, illustrating the ‘exceptionalism’ granted to patents over broader competition concerns and the failure to grapple with critical competition issues at the interface between patents and competition laws. This chapter takes the opportunity offered by a recent review of the Australian law to highlight the analysis that ought to be applied to competition law if it is to serve the purpose of promoting innovation. The current Australian consensus appears to be that the purposes of patents and competition laws are ‘largely complementary’, with patents promoting innovation and competition laws ‘keeping markets open and effective’ in order to ‘preserve the primary source of the pressure to innovate and to diffuse innovations’ (Ergas 2000, p. 6; see also United States Department of Justice and the Federal Trade Commission 1995, p. 7). However, the actual boundaries between patents and competition law are a lot more problematic, reflecting the different ways they seek to achieve their purposes – there are ‘serious problems of assessing effects, but not conflicting purposes’ (Bowman 1973, p. ix; see also Stonier 1984, pp. 22–5). As the laws currently stand, the Trade Practices Act seeks to promote the process of competition through rivalry between competitors for the custom of consumers, theoretically increasing consumer access to goods and services and reducing prices (Hilmer 1993, p. 4). In contrast, patents under the Patents Act limit this competitive process by granting exclusivity to only those innovative ideas that satisfy the minimum legislated requirements for patentability. While the optimal scope and duration of statutory patent privileges is unknown (and unknowable) the current protections of the Australian patent law (the Patents Act) reflects: 1.

Australia’s commitments to international minimum patent standards as part of the World Trade Organisation’s Agreement on Trade Related Aspects of Intellectual Property Rights and the Australia–United States Free Trade Agreement;

Patents and competition

2.

3.

103

the Australian Government’s standards as set out in the Patents Act and Trade Practices Act, the way the standards are applied by IP Australia and the courts, and a general policy drive for productivity and economic growth (for a recent overview of the Australian Government’s patent policy see Lawson and Pickering 2004, pp. 363–6; see also Lawson 2008); and the application and exercise of patent privileges within the boundaries of other statutory regulation schemes, such as the requirement for ‘the quality, safety, efficacy and timely availability of therapeutic goods’ under the Therapeutic Goods Act 1989 (Cth) before a patent protected therapeutic agent may be marketed (see s. 4(1)).

There remains debate in Australia about the most appropriate form of competition regulation to balance the objectives of competition law against the necessary incentive for innovation addressed by the patent. Several inquiries have been mounted to examine the interface between patent privileges and competition law. They include Weisbrot 2004; Dawson 2003; Ergas 2000; and Samuel 1999. Despite the inquiries, two of the central unresolved questions are: 1.

the content and application of regulation that will improve business investment in innovation, stimulate growth of innovative firms, strengthen commercial linkages between publicly funded research institutions and industry, and take promising research to the stage of commercial viability (see, for example, Commonwealth of Australia 2001a, p. 18). The place given to legislated patent privileges, and in particular ‘stronger’ patent privileges, has failed to connect the issue of the performance of the regulation required by the high level Competition Principles Agreement (CPA) that forms part of the National Competition Policy (see National Competition Council 1997). That is, demonstrating that the benefits to the community of restricting competition through patents outweighs the costs (see Competition Principles Agreement cl. 5(1)(a)), and that the objectives of the patent privileges can only be achieved by restricting competition (see Competition Principles Agreement cl. 5(1)(b)); and 2. the nature and scope of exemption that the long-standing provision, s. 51(3) of the Trade Practices Act, should allow for licence and assignment dealings with patent protected products and processes. In other words, how should patent licence and assignment conditions be exempt from Pt IV of the Trade Practices Act (the competition laws)? Pt IV is generally concerned to restrict contractual practices that involve fixing or controlling prices, exclusionary provisions, exclusive dealings, and mergers and acquisitions that might restrict competition. As a generalisation, these

104


restrictions are against contractual practices that adversely affect prices, quantities and qualities of presently or potentially available varieties of goods and services (Explanatory Memorandum 1974, pp. 4–8). On this basis, this chapter examines these two areas of competition policy that might deliver better outcomes: first the benchmark for regulatory quality and performance according to the CPA, and secondly the evolution and application of the existing pro-competition measures in s. 51(3) of the Trade Practices Act. The chapter concludes with the finding that a detailed competition analysis of the patent privileges set out in the Patents Act, and their exercise according to the Trade Practices Act, has been avoided. It should be pursued in order to deliver better quality and performing laws. Such laws are more likely to promote innovation without the detrimental effects of inappropriately restricted competition.

II. REGULATORY QUALITY AND PERFORMANCE OF THE PATENTS ACT ACCORDING TO THE CPA The Australian Government, as part of the agreements made by the Council of Australian Governments (CoAG), has undertaken an extensive review of its regulations and government actions in order to remove anti-competitive arrangements that cannot be justified to achieve an identifiable benefit or ‘public interest’ (see National Competition Council 2003, pp. 13.1–13.7; Rimmer 2006, pp. 3–6). The result has been some institutional changes, related reforms to the electricity, gas, water and road transport industries, and various reforms to regulation (see National Competition Council 1997). The regulatory reforms have primarily been implemented through the CPA (see Productivity Commission 2000, pp. 1–10). Essentially, this process involves examining existing (and proposed) regulation against a pro-competition template (Competition Principles Agreement cl. 5(3)) with a view to establishing that the regulations in force across the Commonwealth are ‘efficient’ in terms of ‘maximising the benefits to the community taking account of the costs’, and ‘effective’ in ‘addressing an identified problem’ (Office of Best Practice Regulation 2007, p. 1; see also Productivity Commission 2003, p. 1). The approach to conducting, and the substance of, these regulation reviews is primarily addressed in the Terms of Reference, although there may be additional considerations (see, for example, Competition Principles Agreement cl. 5(9)), mandatory procedures (see, for example, Office of Best Practice Regulation 2007, p. 1) and guidance from other sources (see, for example, Centre for International Economics 1999; see generally National Competition


105

Council 2004). Essentially, the objectives are to assess whether the arrangements restrict competition; whether the benefits to the community as a whole outweigh the costs (including the broader assessment of the ‘public interest’) and whether it can clearly be demonstrated that the benefits exceed the costs; and whether the same objects could be achieved by other better (more procompetitive) means (Centre for International Economics 1999, p. 7; see the Competition Principles Agreement cl. 5(1)). Following the CPA programme proposed by the Australian Government, the ‘Ergas inquiry’ (the Intellectual Property and Competition Review Committee) undertook a review of intellectual property legislation (excluding the Plant Breeder’s Rights Act 1994 (Cth)) (see Ergas 2000). I shall pursue the work of this Committee to exemplify Australian policy analysis. The Terms of Reference of the Committee provided, in part, that the review ‘shall have regard to: (a) the determination, in the CPA, that legislation which restricts competition should be retained only if the benefits to the community as a whole outweigh the costs, and if the objectives of the legislation can only be achieved by restricting competition’ (Ergas 2000, p. 217). However, the Terms of Reference also included specific matters that the review ‘shall inquire into and report … on’, including ‘the objectives of, including the nature and magnitude of the problems sought to be addressed by … the Patents Act 1990’, ‘the nature of the restrictions in the legislation on competition’, ‘the likely effect of those restrictions on competition’, alternative means of achieving the same objectives, and the ‘costs and benefits’ and ‘appropriateness, effectiveness and efficiency’ of the legislation, restrictions on competition and alternatives (at p. 217). These requirements of analysis are consistent with the CPA (Competition Principles Agreement cl. 5(9)). Ergas set out its view of the impact of intellectual property privileges on competition, including patents: it is important to recognise that competition occurs in a number of dimensions. More specifically, firms do not only compete in the prices they set but also in their ability to develop new processes and to design and market new products. This dynamic competition is of special importance. In effect, rather than simply reallocating existing resources, it expands the resources on which society can draw and allows for sustainable increases in living standards. It is also important because in practice it is the main way established market positions are over-turned, and the threat of competition made into an ever-present constraint on the conduct of firms. An effective system to define and enforce intellectual property rights is critical for this type of dynamic competition to occur on a material scale. (Ergas 2000, p. 5)

Importantly, Ergas expressed the view that the interaction between intellectual property and competition was ‘largely complementary’, with intellectual property promoting innovation and competition policy ‘keeping markets open and effective, preserv[ing] the primary source of the pressure to innovate and

106


to diffuse innovations’ (p. 6). However, recognising that intellectual property privileges do have social costs, Ergas conceded: Intellectual property laws must … involve some balance between the incentives to invest in creative effort and the incentives for disseminating material that is the subject of intellectual property protection. This balance turns on determining the appropriate scope of protection, in terms of the conditions under which protection is granted, the scope and effectiveness of the exclusive privileges provided by protection, and the duration of the protection given. Balancing between providing incentives to invest in innovation on one hand, and for efficient diffusion of innovation on the other, is a central, and perhaps the crucial, element in the design of intellectual property laws. In the Committee’s view, it is essential that the terms of this balance be clearly set out in the intellectual property laws themselves, so that rights owners and users can be certain about the scope and content of the grants being made. (p. 6)

In addressing patents specifically, Ergas rejected the notion that Australia might apply a higher threshold standard to non-resident patent applicants (p. 139). In doing so, given that Australia is overwhelmingly an importer of patents, it presented a particular perspective on the benefits of patents in Australia: effective patent protection facilitates trade in technology, both domestically and internationally. An effective patent system, accessible to foreign technology suppliers, allows Australian firms to import technology that would otherwise be unavailable, or would only be available at higher cost. This increases productivity and enhances competition in the Australian economy. The importance of technological imports is illustrated by the more than 90 per cent of patents registered in Australia, which are owned by foreigners. In addition, there are more indirect cross-border spill-overs through importing of goods which embody innovations and which may be used as intermediate inputs or sold directly to end-users. (p. 139)

Ergas did present some assertions in support of its perspective about the benefits of patent privileges. It argued that the private value of research and development was much less than the social value (p. 137), and that patent privileges were the best system yet devised to balance the trade-off between maintaining incentives to invest and fostering the diffusion of new technology (p. 143). Unfortunately, these assertions, while not contentious as a generalisation, gloss over, for example, the hotly contested and disparate debate about the appropriate scope and allocation of patent privileges that Ergas itself had identified in discussing the challenge of balancing incentives and exploiting intellectual property generally (p. 6). Interestingly, Ergas did cite ‘uncertainty as to which of several contending parties will receive patent protection and how much protection patents will afford’ as an ‘imperfection’ in the existing patent privilege scheme (p. 143). Further, Ergas’s analysis and conclusions were not based on Australia’s expe-


107

rience with patent privileges, but rather relied on international comparisons that were then assumed to be applicable to Australia. Ergas then concluded that patent privileges can lead to ‘losses in allocative and productive efficiency’ but ‘[i]n practice … a patent holder can rarely act as a pure monopoly, because of the availability of alternative and substitute products and processes, and also because some scope for imitation almost always exists’ (p. 138). The loss of some ‘dynamic efficiency’ in the development of derivative innovations was also acknowledged, but again, ‘[t]o some extent dynamic losses are counteracted by the disclosure of ideas as part of the quid pro quo of granting a patent and … the patent system itself … facilitates the use of licensing’ (p. 139). Ergas then reached an ‘overall’ conclusion: Overall, the Committee agrees with Scherer that ‘the patenting system is recognised to be an imperfect instrument. Nevertheless, it may be the best solution policy man can devise to the difficult trade-off between, on the one hand, maintaining incentives for investment and, on the other hand, fostering the diffusion of new technology’s benefits to consumers and to those who might make leapfrogging inventions’. (p. 143)

Having adopted this conclusion, the view that compliance with international patent standards was beneficial to Australia (pp. 27 and 139–41) and a part of Government policy (pp. 216–17), glossing the debates about appropriate patent scope and allocation, Ergas accepted the existing legislated scheme for patent privileges. Nonetheless, it identified a number of improvements that might promote more competition in the application of the threshold tests for patentability and the duration of the patent term (p. 144). However, we should note that these issues were examined from the perspective of Ergas’s particular concern about the economic effects of the certainty of the patent grant (pp. 143–4), both in granting patents that should not be granted and in not granting patents that should be granted (p. 153). From this perspective Ergas considered threshold test improvements. They included requiring a specific, substantial and credible use to be defined (pp. 151–4) and that the scope of prior art be expanded for assessing inventive step (pp. 154–6 and 168–70). It was suggested that other requirements be restricted, including prior use (pp. 157–9) and compulsory licensing (pp. 162–3). On patent term, Ergas ‘believed’ there was not enough evidence to extend the patent term, although it did suggest that raising renewal fees might be applied to ‘extract a lower economic rent’ (pp. 144 and 156). While these assessments and recommendations certainly affect competition policy, the Ergas approach avoided assessing the contentions about the appropriate balance of patent scope and allocation and how this control might be countered when the social costs were judged to be too high (such as the appropriate threshold of ‘public interest’ before a compulsory licence is to be granted).

108


The major flaw in Ergas’s approach was to avoid any analysis of the controversy about the most appropriate threshold requirements in the Patents Act. For example, different theories about the objectives of patent privileges propose very different threshold standards depending on what the patent scheme is intended to achieve. Ergas failed to clearly identify what patent privileges in Australia are intended to achieve (pp. 136–8) and to consider the most appropriate test in achieving this objective (pp. 154–6). Comparing the ‘reward theory’ and the ‘prospect theory’ illustrates this contention. The ‘reward theory’ views a patent as an incentive to undertake uncertain invention with an opportunity to appropriate greater commercial returns. This is considered to foster socially beneficial inventions, but with significant social costs for short-term inefficiencies in the market, resulting from the anti-competitive effects of the patent (primarily being restricted output and higher prices) appropriating public goods (ideas) that would otherwise be used (see, for example, Machlup 1958). In contrast, the ‘prospect theory’ views patents as promoting the commercial development of inventions, with patents granted to early stage inventions facilitating the bringing of a usable invention to the market and acting as an incentive to maximise the commercial value from exploiting the invention with relief from free-riders (see, for example, Kitch 1977; Merges 1992). These different theories pose significantly different consequences for shortterm competition. The ‘reward theory’ imposes high thresholds for patentability, seeking to limit patents to only those inventions that would not have been made with significant concerns about the effects on competition. In contrast, the ‘prospect theory’ imposes lower thresholds, giving the patent holder control over the development process and possibly increasing the efficiency of commercialisation (that otherwise may not occur) with less concern about the effects on competition. Perhaps the major flaw in the Ergas report was that it adopted different approaches (or CPA methodologies) to different intellectual property statutes, a more rigorous approach being applied to the Ergas majority’s consideration of parallel importing under the Copyright Act 1968 (Cth) than to the Patents Act patents. Significantly, on copyright, the Ergas majority objected to many of the very same impacts that it glossed over in its analyses of the Patents Act (pp. 134–78). In addressing parallel importing under the Copyright Act 1968 (Cth) the Ergas majority was able to structure its analysis of the issues very differently, and reach a very different conclusion, suggesting that the benefits of parallel import restrictions did not outweigh the detrimental anti-competitive effects, and that the restrictions should be repealed entirely (Ergas 2000, p. 5). My contention here is that if Ergas had applied the same CPA methodology to Patents Act patents, the outcome would have been a more rigorous and informed analysis of the patent scheme in the context of promoting competition.


109

The Ergas majority accepted that copyright had a ‘utilitarian justification of protecting and promoting investment in creative effort to secure, for the Australian community, gains associated with investment’ (p. 61), so that the privileges granted needed to be ‘assessed in terms of whether the benefits they may bring, in improved investment in, and access to the results of, creative efforts, outweigh the costs they impose’ (p. 62). Further, ‘[t]his assessment of the impact of the restrictions needs to include analysis of the wider costs and benefits associated with those impacts’ (p. 62). The majority’s key concern about parallel import restrictions appeared to be market segmentation with the ability to then charge higher prices (and possibly restrict availability) for materials subject to copyright (p. 62). In effect, this was an assessment about international exhaustion of copyright. From this basis the majority was able to reject arguments about economic incentives to create (pp. 49–51 and 66–9), prices and availability (pp. 51–3 and 64–9), remainder books (pp. 54–5 and 64), marketing and services (pp. 55–6 and 66–9), censorship (pp. 56–7), piracy (pp. 57–60) and economic analysis that favoured maintaining the existing restrictions (p. 65), because they failed to satisfy the CPA criteria (p. 73). The most significant difference between the majority’s dealing with parallel imports and patent privileges was the detailed approach to addressing the analysis of whether a restriction on competition was justified: The Committee started from the premise that restrictions on competition need to be justified. In other words, the Committee, consistent with the NCP [National Competition Policy] and the CPA, accepts that the onus of making a case lies with those who would prevent, limit, or in other ways restrict, competitive forces from operating. More specifically, we accept that those who would restrict competition should establish the restrictions are in the public interest, rather than merely serving the interests of particular producers. The Committee believes that this well-established principle – requiring those who would restrict competition to demonstrate the need to do so – appears to be fully justifiable. However, experience and analysis amply demonstrate the importance of competition in promoting efficiency and underpinning prosperous, open economies. It also demonstrates the frequency with which restrictions on competition, though claimed to serve wider interests, have been used to confer above normal profits on narrow groups at the expense of the community. A presumption, albeit a rebuttable one, in favour of competition, is consequently clearly reasonable. Such a presumption also places the evidentiary burden on those best placed to demonstrate the position. The reality is that the benefits from restrictions on competition generally accrue to concentrated groups, while the costs of these restrictions are spread widely throughout the community. Given this spreading of costs, it is far more difficult for those adversely affected by restrictions to organise themselves and present their case, than it is for the direct beneficiaries to support the restrictions.

110


As a result, the Committee believes that it is reasonable to expect those who would introduce or perpetuate restrictions to provide convincing evidence of why the restrictions are in the public interest. It follows that the relevant test is whether the material made available to the Committee establishes that the restrictions these provisions impose on competition confer benefits on the community that outweigh their costs. In cases where arguments put to us appear weak, the Committee actively sought further information and tried to analyse the arguments in the best light. As a result, we are convinced that we have provided the differing points of view with a fair and thorough hearing. (p. 61)

The different approach (or CPA methodology) of the Ergas majority, in directly addressing the arguments about theoretical benefits of particular policy settings for parallel importing, and the absence of this analysis for patent privileges, is perplexing. Significantly, the majority questioned the assertions of benefit that copyright privileges under the Copyright Act 1968 (Cth) were protecting and promoting investment. Had Ergas applied a similar critical analysis of patent privileges, then the debates about appropriate patent scope and allocation would probably have been more closely examined and the requirements of the CPA more properly addressed. Further, broader issues such as the high costs of patented pharmaceuticals, non-tariff trade barriers, ethical considerations about patenting life, and so on, may have required consideration in a broader assessment of the ‘public interest’. This approach appears to more closely fit with the CPA and the principles articulated in the Hilmer report directed to promoting competition. Further, such an analysis is more likely to deliver some insight into the effects of patent privileges and their likely benefits for the Australian community. Similar to the Ergas majority’s approach to patent privileges, the Ergas minority view accepted the assumptions and assertions of benefit, and therefore concluded that parallel import restrictions in the Copyright Act 1968 (Cth) were justified: It is true that the ability to restrict parallel imports gives rise to an economic rent in favour of the copyright owner. However this rent encourages innovation and investment, and is precisely the foundation on which copyright is based. Allowing parallel imports reduces the incentives to innovate or invest. It is submitted that the costs incurred in removing the restriction will exceed the costs (in economic terms) of retaining that power. (p. 74)

The consequence of the Ergas minority accepting this approach was to avoid the broader assessment of the anti-competitive effects of copyright and a proper assessment of the criteria set out in the CPA. Significantly, these are exactly the flaws in Ergas’s assessment of the Patents Act (and also the relevant parts of the Trade Practices Act considered below).


111

As I have noted, Ergas did make some recommendations for patent law reform. In response to Ergas, the Australian Government has amended the Patents Act and adopted new practices as follows: 1. expanded the publicly available information (the prior art) to which an invention is compared to determine whether it is novel and involves an inventive (or innovative) step (see Patents Amendment Act 2001 (Cth) sch. 1 (and sch. 2); Intellectual Property Laws Amendment Act 2003 (Cth) sch. 2); 2. introduced the ‘balance of probabilities’ evidence standard for patent examinations (Commonwealth of Australia 2001b, pp. 26974–5; 3. applied a standard that each aspect of use be ‘specific, substantial and credible’ (see IP Australia 2007, para. [1.3.8.6.3]; following Australia–United States Free Trade Agreement, Art. 17.9.13); 4. implemented more steeply rising renewal fees over the term of a patent (Patents Regulations 1991 (Cth) sch. 7 (Part 2, item 211)); 5. required various patent searches to be disclosed with the application (Patents Amendment Act 2001 (Cth) sch. 1); 6. introduced a ‘grace period’ for any information made publicly available by or with the consent of the applicant by publication or use within 12 months before the filing date of the complete application (Patents Regulations 1991 (Cth) rr. 2.2 and 2.3); 7. clarified the ‘prior use’ defence to patent infringement (Intellectual Property Laws Amendment Act 2006 (Cth) sch. 5); and 8. added a new ground of compulsory licensing, although in a different form from that recommended by Ergas (Intellectual Property Laws Amendment Act 2006 (Cth) sch. 8; see also Ergas 2000, p. 163). Significantly, the Australian Government has not repealed ss 144–6 of the Patents Act dealing with contracting conduct that is also addressed by s. 51(1) of the Trade Practices Act, despite accepting this recommendation (see IP Australia 2001, p. 8). Further, the legislation amending the Patents Act appears to have been assessed by the Office of Best Practice Regulation (formerly the Office of Regulation Review) as adequately addressing the CPA requirements for proposed regulations. This perhaps reflects the inability of the Australian Government to determine so far the appropriate content of competition, and instead favour the process of analysis over the content of the analysis (see CoAG 2006, attachment B, decision 1.2(a); see also Lawson and Hindmarsh 2007).

112


III. THE JUSTIFICATION FOR THE PARTIAL TRADE PRACTICES ACT EXEMPTION OF THE PATENTS ACT The objective of the current Trade Practices Act is to ‘enhance the welfare of Australians through the promotion of competition and fair trading and provision for consumer protection’ (s. 2). This is achieved by restricting some forms of market conduct and market structures considered to be anti-competitive (Pts IIIA, IV and XIB), while allowing that prohibited conduct and structure if it can be justified to be in the ‘public interest’ (Pt VII). This balances competition as an important means to achieving economic efficiency, rather than competition as an end in itself (Dawson 2003, p. 32; Skehill 1992, pp. 94–6), and targets particular horizontal, vertical and economic monopolisation practices that are likely to undermine that efficiency. Thus, the Trade Practices Act ‘is conducive to greater economic efficiency by the prohibition of anticompetitive conduct, but it is not creative of it’ (Skehill 1992, p. 95). The Trade Practices Act, however, does not generally define ‘competition’. Instead, it relies either on prohibiting some conduct always considered to be anti-competitive (per se prohibitions), or prohibiting some other conduct and some market structures only when they have passed a threshold of ‘substantially lessening competition’ or a ‘substantial degree of power in a market’ (the misuse of market power) (Pt IV). These prohibitions are then relaxed for some dealings with patent protected products and processes, or the prohibitions are avoided by seeking an authorisation, notification or clearance. The Trade Practices Act also deals with some conduct and market structures, including patent related products and processes that are an integral but subsidiary service or facility, under the rubric of access to essential facilities and the telecommunications market. They are not addressed in this chapter (see for a further articulation, Ergas 2000, pp. 214–15). Over time, the Trade Practices Act has been subjected to considerable amendment. However, the measures specifically addressing intellectual property have only been subjected to limited change. The effect of the current s. 51 of the Trade Practices Act provision is to provide an exemption from Pt IV of the Trade Practices Act for certain conditions in a licence or assignment contract that ‘relates to’ the patent protected product or process, except misuse of market power (see ss 46 and 46A) and resale price maintenance (see s. 48). Thus, the current s. 51 of the Trade Practices Act (convolutedly) provides, in part: (1) In deciding whether a person has contravened this Part [IV], the following must be disregarded: (a) anything specified in, and specifically authorised by: (i) an Act (not including an Act relating to patents, …); or (ii) regulations made under such an Act. …


113

(3) A contravention of a provision of this Part [IV] other than section 46, 46A or 48 shall not be taken to have been committed by reason of: (a) the imposing of, or giving effect to, a condition of: (i) a licence granted by the proprietor, licensee or owner of a patent, … or by a person who has applied for a patent …; or (ii) an assignment of a patent, … or of the right to apply for a patent …; to the extent that the condition relates to: (iii) the invention to which the patent or application for a patent relates or articles made by the use of that invention.

The exemption of some patent licence and assignment contract conditions was reflected in earlier incarnations in the Trade Practices Act 1965 (Cth) and Restrictive Trade Practices Act 1971 (Cth), that in turn trace their origins to the Restrictive Trade Practices Act 1956 (UK). The purpose of retaining a specific exemption for some patent licence and assignment conditions in the Trade Practices Act is uncertain, other than it ‘removes doubts that the exemptions provided in relation to patents and other forms of industrial property are those provided in clause 51(3) and are not to be found under clause 51(1)(a)’ (Explanatory Memorandum 1974, p. 13). Perhaps the provision was ‘to prevent a perceived clash between the interests of intellectual property owners and competition law’ because at the time of the Trade Practices Act enactment ‘intellectual property laws were widely believed to confer on the owners of intellectual property a limited economic monopoly’ (Ergas 2000, p. 206; see also Samuel 1999, p. 149). Its present purpose, however, appears to be: providing businesses with greater certainty when engaging in licensing and assignment activity. This greater certainty can help reduce the costs associated with compliance with trade practices law and encourage more licensing [and assignment] activity. (Samuel 1999, p. 150; see also Ergas 2000, p. 211)

Unfortunately, the application of ss 51(1) and (3) of the Trade Practices Act remains uncertain. This in part reflects its complex construction and the paucity of extrinsic materials and court decisions examining its possible meanings. Perhaps significantly, a literal construction means that (from Ergas Interim 2000, attachment 1): 1.

2. 3.

the parenthesis in s. 51(1)(a) clarifies that ‘there is no general exemption to the operation of pt IV for things specified in, and authorised by, Commonwealth laws relating to patents’; section 51(3) does not apply to conduct that is a misuse of market power (ss 46 and 46A) and resale price maintenance (s. 48); section 51(3) does not include the license or assignment of future rights;

114

4. 5. 6.


section 51(3) does not deal with know-how or confidential information relating to patent protected products and processes; it is not clear whether s. 51(3) applies to an entire assignment or a part only of the assignment; and the s. 51(3) exemption ‘probably does not apply in respect of the intellectual property rights but it applies to the subject of the rights’ so that it is the patent protected product or process that is being exempted by the provision.

The High Court’s one interpretation of this provision clarifies that some licence and assignment activity is exempted from Pt IV of the Trade Practices Act, although overall the scope of the exemption remains uncertain. In Transfield Pty Ltd v Arlo International Ltd (1980) the High Court considered a ‘best endeavours’ clause that provided: The Licensee covenants during the period of the Power Transmission Line Licence at all times to use its best endeavours in and towards the design fabrication installation and selling of the ARLO PTL pole throughout the licensed territory and to energetically promote and develop the greatest possible market for the ARLO PTL pole. (at p. 87 of the report of the judgment)

The licence agreement concerned the use of a patent protected process for the manufacture and erection of steel poles. Following Transfield Pty Ltd’s (‘Transfield’) contract with Arlo International Ltd (‘Arlo’), Transfield had used Arlo’s pole to construct a transmission line and encountered some technical difficulties (p. 88). In a subsequent tender for a transmission line Transfield had calculated its costs based on using the Arlo poles, but had not expressly provided that it would use those poles (p. 88). During the tender discussions some doubts were expressed about the suitability of the Arlo pole and Transfield had ‘overcome the apprehensions’ by not pressing the use of the Arlo pole, but rather offering its own pole developed in reliance on some information provided by Arlo about its pole (p. 88). The tender was finally awarded to Transfield on the basis of using Transfield’s pole (p. 88). Arlo subsequently brought an action against Transfield claiming that it had not used its ‘best endeavours’ to promote the Arlo pole as required by the contract (p. 89). It was common ground that ‘as the result of the use of the Transfield pole, the potential market for the Arlo pole in Australia was effectively destroyed’. The issue before the High Court was whether the ‘best endeavours’ clause confined Transfield to using only Arlo’s pole. In its defence, Transfield argued the clause was either unlawful because s. 45(1) of the Trade Practices Act prohibited a clause with the purpose, effect or likely effect of substantially lessening competition, or a prohibited term


115

under s. 112 of the Patents Act 1952 (Cth) (p. 91). The trial judge and Court of Appeal had found that Transfield had not used its best endeavours as it was required to do according to the contract (p. 90), and that Arlo had lost a substantial share of the market as a result of its pole not being used as it had intended by its contract (pp. 90 and 91). The High Court majority decided in favour of Arlo (see p. 92 per Barwick CJ, 97 per Stephen J, 103 per Mason J, 109 per Wilson J and 105 per Murphy J, dissenting). In essence, the decision reflects difficult facts and circumstances, and finally the judges’ preference that Transfield should not escape its contractual obligations with significant consequences for Arlo by taking advantage of a defence based on the legalistic construction of a statute. Transfield argued that the ‘best endeavours’ clause was unenforceable because it was outside the patent exemption in the Trade Practices Act with the effect that the exemption would allow Transfield to escape its earlier contractual obligations: In our opinion there is no warrant for the introduction of words of the kind contended for by the appellant to the plain unambiguous words of cl. 7. It would be destructive of the whole purpose of the licensing agreement if the licensee were at liberty to promote some other pole, if it or somebody else formed the opinion that the other pole was more suitable for a projected use. (p. 90)

In the High Court’s decision, Justice Mason supported a wide construction of the term ‘relates to’ in s. 51(3) of the Trade Practices Act recognising the provision allowed a patentee to impose conditions on the license or assignment of the patent ‘to protect the patentee’s legal monopoly’ (p. 103), but then said, ‘conditions which seek to gain advantages collateral to the patent are not covered by s. 51(3)’ of the Trade Practices Act (p. 103). Chief Justice Barwick considered the condition in question was entirely within s. 51(1) of the Trade Practices Act and attracted the exemption but did not give detailed reasons for this conclusion (p. 92). Justice Wilson considered the clause fell ‘squarely within’ s. 51(3) of the Trade Practices Act exemption but did not give detailed reasons for this conclusion (p. 108). Justice Stephen considered there was no evidence that the condition, read either alone or in combination with other provisions of the licence, had either the purpose or effect of substantially lessening competition (p. 97). Justice Murphy decided the matter on the basis of the Patents Act 1952 (Cth) (p. 105). The key question in considering s. 51(3) of the Trade Practices Act is, if the limited patent exemption were repealed, what conduct would be subject to Pt IV of the Trade Practices Act that was previously exempted? This in part will depend on the interpretation given to the term ‘relates to’ as it appears in the Trade Practices Act. Several views are possible:

116

1.

2.

3.


narrow view – a licence or assignment condition relates to the patent protected product if it relates directly to the goods produced (Samuel 1999, p. 184; Ergas 2000, p. 207); intermediate view – a licence or assignment condition relates to the patent protected product if it seeks to protect the patentee’s exclusive rights or secure an advantage that is not collateral to the patentee’s exclusive rights (Ergas 2000, p. 207); or broad view – a licence or assignment condition relates to the patent protected product unless it seeks to apply to an almost entirely unrelated transaction or arrangement (Samuel 1999, p. 184; Ergas 2000, p. 207).

Despite the broad ranging reviews of the Trade Practices Act, only very few reviews have actually examined the operation of s. 51(3) of the Trade Practices Act (see Weisbrot 2004, pp. 555–80; Dawson 2003, p. 86; Ergas 2000, pp. 202–15; Samuel 1999, pp. 149–246; Hilmer 1993, pp. 149–51; Stonier 1984, pp. 22–6). Of particular significance, Ergas, as part of its broader assessment of the Patents Act under the CPA (discussed above), examined the recommendations of the earlier Samuel (1999) report (Ergas 2000, pp. 202–15). Its Terms of Reference only required Ergas to ‘have regard to … the conclusions and recommendations’ of the Samuel (1999) report. In addressing the Terms of Reference, Ergas carefully confined its comments to the existing legislative scheme, ‘considering the effects that (given the [Trade Practices Act] as it stands) would flow from different approaches to the coverage by the Act of conduct relating to the exercise of [intellectual property] rights’ (p. 210). Further, Ergas carefully noted that the Trade Practices Act was fashioned in a different economic era and probably should be subjected to its own independent review, whereupon the place of patent privileges might be more substantially addressed (pp. 209–10). With these riders in place, Ergas recommended that the Trade Practices Act should be amended to further narrow the existing exemption by applying a test of whether the relevant conditions in licences and assignments substantially lessened competition. This would adopt a test that applied in other parts of that Trade Practices Act (pp. 11 and 215; including a refusal to deal, p. 213). Yet, Ergas’s view appears to have been that the Trade Practices Act ‘should come into play when intellectual property rights are used in ways that go beyond the scope of the right being granted’ (p. 24) – the so-called ‘scope of grant’ doctrine derived from the ‘inherency’ doctrine in the United States (see generally Carrier 2002, pp. 788–91; Stedman 1973, p. 595; Buxbaum 1965, pp. 641–5). The consequence of this view appears to have been that the exercise of patent privileges was therefore an acceptable restriction on competition so long as exercising the patent privileges was not ‘going beyond market power’ (Ergas 2000, p. 211):


117

The Committee recognises that the IP legislation confers upon the intellectual property right holder a series of exclusive privileges designed to promote innovation. Given that these rights are conferred by legislation, they should be able to be effectively exercised even when this involves (as it generally must) the exclusion of others. However, these rights should not be capable of being used to go beyond the market power those rights directly confer. That is, the right holder should not be allowed to extend the statutory right into a wider right of exclusion with the effect of substantially lessening competition.

Ergas received a number of submissions arguing that the existing Trade Practices Act provision should be retained because of ‘the importance which Australian industry and research institutes attach to the operation of s. 51(3)’ of the Trade Practices Act, ‘the need to protect exclusive licensing arrangements, such as those underpinning the operation of CRCs [government funded research institutions]’ and ‘the need for certainty in intellectual property licensing’ (pp. 207–8). Perhaps surprisingly, Ergas noted that ‘[s]omewhat paradoxically, submitters typically admitted that the precise meaning of s 51(3) [of the Trade Practices Act] was highly uncertain, and yet they asserted that it had provided a firm basis for major investment decisions’ (p. 208). Despite this ‘paradox’, and the carefully phrased riders about the Trade Practices Act being a competition law reflecting the times of its origins (p. 210), Ergas appeared to consider that transaction costs pose a special barrier to the ‘efficient use of intellectual property’ compared with ‘other property or assets’ (p. 210), and as a consequence: this means that it is essential that firms have the scope to enter into efficient contracts that involve intellectual property rights, free of onerous and ultimately counter-productive regulatory burdens. Great caution is therefore needed in imposing on transactions in IP rights constraints that may be less costly when applied in other areas of property. (p. 211)

Despite its reservations about the drafting of s. 51(3) of the Trade Practices Act, Ergas rejected simply repealing the provision because of theoretical efficiency gains from some of the conduct prohibited by Pt IV of the Trade Practices Act and the uncertainty about the effects of a repeal on licensing and assignment decisions (p. 212). The ‘central question’ for Ergas was: ‘whether the current s 51(3) [of the Trade Practices Act] strikes an appropriate balance in allowing the exercise of the rights but not their use in ways that go beyond the purposes underlying the rights?’ (p. 212). While Ergas was uncertain about ‘what content would be left in [s. 51(3) of the Trade Practices Act], if (in line with the [Samuel (1999)] recommendations), all horizontal arrangements and all price and quantity restrictions were removed from its scope’ (Ergas 2000, p. 213), it ‘believ[ed] that adoption of the [Samuel (1999)] proposal would amount to a repeal of the section’

118


(p. 213). The consequence was to ‘impose unnecessary costs on the innovation process’ (p. 213). The solution that balanced the needs of the intellectual property system and the wider goals of competition policy, in Ergas’s ‘view’, ‘requires a careful re-framing of the section’ (p. 213): we believe this balance can best be addressed through the following changes: • repealing s 51(3) and related provisions from the [Trade Practices Act]; • …; and • including amendments in the [Trade Practices Act] to ensure that a contravention of Pt IV of the [Trade Practices Act], or of s 4D of that Act, shall not be taken to have been committed by reason of the imposing of conditions in a licence, or the inclusion of conditions in a contract, arrangement or understanding, that relate to the subject matter of that intellectual property statute, so long as those conditions do not result, or are not likely to result, in a substantial lessening of competition. (pp. 213 and 215)

Ergas also considered that: The Committee also notes that the imposing of conditions in a licence, or the inclusion of conditions in a contract, arrangement or understanding, should also clearly mean the refusal by the owner of an [intellectual property] right to enter into a licence, contract, arrangement or understanding. This means that the intellectual property owner still has the benefit of the relevant provisions even in the absence of a contract. However, this will not over-ride separate provisions made in the IP legislation that bear on compulsory licensing. (p. 213)

To assist the ‘users of the [intellectual property] system to better understand the proposed changes’ Ergas considered the Australian Competition and Consumer Commission (ACCC) should issue guidelines ‘on how it will implement enforcement activities related to the provision’ (pp. 19 and 215). The purpose of these guidelines would be to ‘provide sufficient guidance to owners of intellectual property rights as to the types of behaviour likely to result in a substantial lessening of competition’ and ‘make provisions for potential contractors to seek written clearance from the ACCC on whether the proposed behaviour is likely to result in a substantial lessening of competition’ (pp. 19 and 215). The results of these changes would: ensure that conduct relating to the subject matter of IP rights, which does not result in a substantial lessening of competition, would be exempt from the relevant provisions of the [Trade Practices Act]. It would accept that conduct from the Act’s per se prohibitions, thus minimising the range of instances in which administrative procedures such as authorisation or notification were required. (p. 214)

Ergas’s conclusion – that the Trade Practices Act ‘should come into play when [patents] are used in ways that go beyond the scope of the right being granted’ (p. 24) – failed to assess whether a patent privilege is an acceptable


119

restriction on competition, even though there has been no substantial lessening of competition (such as the per se prohibitions that do not have the purpose, effect or likely effect of substantially lessening competition). Its justification for this conclusion appears to be: it seems reasonable to suppose that the per se prohibitions embodied in the [Trade Practices Act], and the potentially burdensome requirements for administrative review, would catch many license conditions that are usually socially beneficial – for example, tying and exclusive dealing arrangements in patent licenses. Over the longer term, this could both reduce innovation and distort competition between those (typically smaller and more specialised) firms that depended on licenses and assignments and those that did not. (p. 212)

The major criticism of this ‘scope of grant’ approach is that it resolves the interface between patents and competition law by favouring patents, with an inherent assumption that the welfare increases from patents will always outperform the welfare increases from enhanced competition (see, for example, Carrier 2002, p. 764). Unfortunately, Ergas provided no further analysis of what exempted per se prohibited conduct was to be regarded as socially detrimental (or its consequences). At the same time, it assumed that any detrimental conduct would be outweighed by beneficial conduct because of the vulnerability of licensing practices to uncertainty and transaction costs (see Ergas 2000, pp. 210–11; also Eagles and Longdin 2003, p. 35). If, however, Ergas was really only concerned about certainty and transaction costs, then there are arguably better ways of addressing this than an exemption from competition laws (see, for example, Hilmer 1993, pp. 88–9 suggesting direct subsidies, tax breaks, and so on), but none of these was addressed. Following these various reviews (including Ergas), the Australian Government has responded with its legislative proposals (see IP Australia 2001; Treasury 2003). The responses, in dealing with the Trade Practices Act exemptions for patents, are to: 1.

continue to accord distinctive treatment to patents (and some other forms of intellectual property) under the Trade Practices Act (IP Australia 2001, p. 11), but not know-how, confidential information or data protection; 2. include transitional arrangements to preserve the effect of the existing licences and assignments entered into before the proposed amendments (IP Australia 2001, p. 12); 3. maintain the separate treatment of ss 46, 46A or 48 of the Trade Practices Act (misuse of market power and resale price maintenance) in dealing with any exemptions to the Trade Practices Act (IP Australia 2001, p. 12); 4. limit patent license (and presumably assignment) so that ‘IP licensing would be subject to the provisions of Pt IV [of the Trade Practices Act],

120


but a contravention of the per se prohibitions (ss 45, 45A and 47, or of s 4D), would instead be subject to a substantial lessening of competition test’ (IP Australia 2001, p. 12); and 5. require the ACCC to issue guidelines ‘outlining its enforcement approach to these provisions’ so that the guidelines ‘outline when IP licensing and assignment conditions might be exempted under s 51(3) [of the Trade Practices Act], when IP licences and assignments might breach Pt IV …, and when conduct that is likely to breach Pt IV … might be authorised’ (IP Australia 2001, p. 12; see also Treasury 2003, p. 5). In dealing with the Trade Practices Act generally, the Australian Government proposes to: 1.

2.

3.

4.

5.

support the development of guidelines by the ACCC about the application of Pt IV of the Trade Practices Act to intellectual property (Treasury 2003, p. 5) and joint ventures (often involving intellectual property) (Treasury 2003, p. 9); restrict the per se prohibition against exclusionary provisions by establishing a competition defence that the provision does not have the purpose, effect or likely effect of substantially lessening competition, and confine the reach of the prohibition ‘to those agreements that target competitors, actual or potential, of the parties to the agreement’ (Treasury 2003, pp. 7–8); remove the prohibition per se against third-line forcing, and instead prohibit third-line forcing only if it has the purpose, effect or likely effect of substantially lessening competition (Treasury 2003, p. 8); modify by amendment the limits on exemptions for joint ventures from the per se prohibition against exclusionary provisions and price fixing (Treasury 2003, p. 8); and introduce by amendment a time limit of six months for the consideration of non-merger applications for authorisation by the ACCC, and include discretion to waive, in whole or in part, the fee for filing a non-merger application for authorisation (Treasury 2003, p. 6).

The various Australian Government responses indicated that the uncertain s. 51(3) of the Trade Practices Act exemption will be limited further, although the precise formulation of the exemption remains unclear. It appears that the exempted conduct that s. 51(3) of the Trade Practices Act will continue to allow will be those conditions in licences and assignments that ‘relate to’ the patent protected products and processes, but would otherwise be characterised as per se prohibitions, or that have the purpose, effect or likely effect of substantially lessening competition. These include for example:


121

1.

price fixing (per se) – s. 45A(1) of the Trade Practices Act deems price fixing (price, discount, allowance, rebate or credit) contracts, arrangements or understandings to have the purpose, effect or likely effect of substantially lessening competition. The consequence is that price fixing is prohibited per se under ss 45(2)(a)(ii) and (2)(b)(ii) of the Trade Practices Act – which proscribe the making or giving effect to a provision in a contract, arrangement or understanding that has the purpose, effect or likely effect of substantially lessening competition; 2. exclusionary conduct (per se) – ss 45(2)(a)(i) and (2)(b)(i) of the Trade Practices Act proscribe the making or giving effect to an exclusionary provision in a contract, arrangement or understanding, being defined in s. 4D of the Trade Practices Act as a contract, arrangement or understanding made between competitors which has the purpose of preventing, restricting or limiting the supply or acquisition of goods or services to or from particular persons or classes of persons either altogether or in particular circumstances or on particular conditions; 3. covenants that fix prices (per se) – s. 45C(1) of the Trade Practices Act provides that, in applying the restrictions on covenants (as defined in s. 4) that fix prices as per s. 45B(1) of the Trade Practices Act, the substantially lessening competition requirement is ‘omitted’, so that a covenant is unenforceable ‘in so far as it confers rights or benefits or imposes duties or obligations on a corporation or on a person associated with a corporation’; and 4. third-line forcing (per se) – s. 47(6) and (7) of the Trade Practices Act provides that the supply, offer to supply or price (including a discount, allowance, rebate or credit) of goods or services tied to acquiring goods or services of a particular kind or description directly or indirectly from another person is prohibited exclusive dealing according to s. 47(1) of the Trade Practices Act.

These prohibitions may, however, be allowed where they have been authorised (s. 88), cleared (s. 95AC) or notified (in the case of some exclusive dealings) (s. 93(1)). While the Australian Government has promised guidelines on the operation of the Trade Practices Act in respect of intellectual property (Treasury 2003, pp. 5 and 9), these remain outstanding, with the ‘in force’ guidelines still being the guidelines issued in 1991 (Trade Practices Commission 1991). Meanwhile, the issues of certainty and transaction costs have not been addressed further by the Australian Government.

IV.

CONCLUSIONS

The policy objective set out in the CPA is to promote competition by removing

122


unjustified restrictions on competition in Australia. For statute-based intellectual property laws, the Hilmer report expressed clear concern that these regulations potentially created barriers to entry that might restrict competition (see Hilmer 1993, p. 195), and that the need for exemptions for certain licence and assignment conditions from the Trade Practices Act were uncertain (Hilmer 1993, p. 150). This chapter shows that important controversial issues have been glossed over in both Ergas and Samuel. Thus, a detailed competition analysis – of the patent privileges set out in the Patents Act and their exercise according to the Trade Practices Act – would appear to have been avoided. While there are practical difficulties in applying the CPA in determining exactly what standard to apply in assessing costs and benefits, it is undertaking the process of analysis that is likely to deliver better regulation by ‘questioning, understanding real world impacts, [and] exploring assumptions’ (Industry Commission 1996, p. 11; Productivity Commission 2005, p. 140). The Ergas’ majority’s approach to parallel importing under the Copyright Act 1968 (Cth) merely highlights an inadequate analysis of the Patents Act 1990 (Cth). Meanwhile, Ergas alluded to concerns about the role of Pt IV of the Trade Practices Act and its design in an ‘intellectual climate … [where] … most participants in the policy debate would have reasonably assumed that many practices that involved tying, bundling, price discrimination or resale price maintenance’ were anti-competitive without setting out its reasoned justification (Ergas 2000, p. 209). This concern might explain Ergas’s reluctance to recommend amendment to the wider provisions of Pt IV of the Trade Practices Act (see Eagles and Longdin 2003, pp. 34–5). However, important conclusions about the hotly contested theoretical preferences for balancing intellectual property, and patents in particular, and competition law were left un-stated, perpetuating the potential for regulatory inefficiencies. Perhaps most disappointingly, the two key issues that require resolution remain unanswered: 1.

the so-called ‘paradox’: why the current s. 51(3) of the Trade Practices Act exemption remains so apparently important when its exact meaning remains so uncertain (see Ergas 2000, p. 208); and 2. given the global nature of intellectual property licensing and assignments, and patent licensing and assignments in particular, why is there a need for favourable treatment in Australia when favourable treatment in one jurisdiction may not apply in another jurisdiction? (see Samuel 1999, p. 200).

While there is unlikely ever to be a complete resolution to the tension between patents and competition law, the failure of recent reviews, and particularly the Ergas review, together with the weak legislative responses, comprise another lost opportunity to develop better quality regulation in Australia (see


123

Stonier 1984, p. 80). Perhaps Australia’s contribution to this debate is to illustrate the ‘exceptionalism’ granted to patents over broader competition concerns, and the ongoing failure to grapple with critical competition issues at the interface between patents and competition laws. The first steps in remedying the relationship between patents and competition, therefore, must be a genuine and critical analysis of patenting (under the Patents Act and Trade Practices Act) according to the CPA’s methodology.

REFERENCES Boldrin, M. and D. Levine (2004), ‘The Case Against Intellectual Monopoly’, International Economic Review, 45(2), 327–50. Bowman, W. (1973), Patent and Antitrust Law: A Legal and Economic Appraisal, Chicago: University of Chicago Press. Buxbaum, R. (1965), ‘Restrictions Inherent in the Patent Monopoly: A Comparative Critique’, University of Pennsylvania Law Review, 113(5), 633–67. Carrier, M. (2002), ‘Unraveling the Patent–Antitrust Paradox’, University of Pennsylvania Law Review, 150(3), 761–854. Centre for International Economics (1999), Guidelines for NCP Legislation Reviews, Canberra: Centre for International Economics. CoAG (2006), Communiqué: 10 February 2006, Canberra: Department of the Prime Minister and Cabinet. Commonwealth of Australia (2001a), Backing Australia’s Ability: An Innovation Action Plan for the Future, Canberra: Big Island Graphics and Corporate Communications ISR. Commonwealth of Australia (2001b), Parliamentary Debates, House of Representatives, 24 May, pp. 26974–5 (Parliamentary Secretary to the Minister for Industry, Science and Resources). Dawson [Trade Practices Act Review Committee] (2003), Trade Practices Act Review, Canberra: CanPrint Communications Pty Ltd. Donald, B. and J. Haydon (1978), Trade Practices Law, Volume 1, Sydney: Law Book Co. Eagles, I. and L. Longdin (2003), ‘Competition in Information and Computer Technology Markets: Intellectual Property Licensing and Section 51(3) of the Trade Practices Act 1974’, Queensland University of Technology Law and Justice Journal, 3(1), 31–50. Ergas [Intellectual Property and Competition Review Committee] (2000), Review of Intellectual Property Legislation under the Competition Principles Agreement, Canberra: IP Australia. Ergas Interim [Intellectual Property and Competition Review Committee] (2000), Interim Report, Canberra: IP Australia. Explanatory Memorandum (1974), Trade Practices Bill 1974 (Cth), Canberra: Australian Government Publishing Service. Gifford, D. (2003), ‘Antitrusts’ Troubled Relations with Intellectual Property’, Minnesota Law Review, 87(6), 1695–718. Hilmer [Independent Committee of Inquiry into Competition Policy in Australia] (1993), National Competition Policy, Canberra: Australian Government Printing Service.

124


Industry Commission (1996), Regulation and its Review 1995–6, Annual Report Series, Canberra: Australian Government Printing Service. IP Australia (2001), Government Response to Intellectual Property and Competition Review Recommendations, Canberra: IP Australia. IP Australia (2007), Australian Patent Office Manual of Practice and Procedures, Canberra: IP Australia. Kitch, E. (1977), ‘The Nature and Function of the Patent System’, Journal of Law and Economics, 20(2), 265–90. Landes, W. and R. Posner (2003), The Economic Structure of Intellectual Property Law, Cambridge, Mass.: Harvard University Press. Lawson, C. (2008), ‘Managerialist Influences on Granting Patents in Australia’, Australian Journal of Administrative Law, 15(1), 1–30. Lawson, C. and R. Hindmarsh (2007), ‘Regulating Genetically Modified Organisms: A Case Study Assessing Regulatory Quality and Performance’, Australian Business Law Review, 35(3), 181–203. Lawson, C. and C. Pickering (2004), ‘“TRIPs-Plus” Patent Privileges – An Intellectual Property “Cargo Cult” in Australia’, Prometheus, 22(4), 355–77. Letwin, W. (1954), ‘The English Common Law Concerning Monopolies’, University of Chicago Law Review, 21(3), 355–85. Machlup [Subcommittee on Patents, Trademarks and Copyrights of the Senate Committee on the Judiciary] (1958), An Economic Review of the Patent System, 85th Congress, 2nd Session, Study No. 15, Washington, DC: Government Printing Office. Merges, R. (1992), ‘Uncertainty and the Standard of Patentability’, High Technology Law Journal, 7(1), 1–70. National Competition Council (1997), Compendium of National Competition Policy Agreements, Canberra: Australian Government Printing Service. National Competition Council (2003), Assessment of Governments’ Progress in Implementing the National Competition Policy and Related Reforms, Vol 2: Legislation Review and Reform, Canberra: AusInfo. National Competition Council (2004), National Competition Council Legislation Review, Compendium, fifth edition, Canberra: AusInfo. Office of Best Practice Regulation (2007), Best Practice Regulation Handbook, Canberra: Productivity Commission. Productivity Commission (2000), Regulation and Its Review 1999–2000, Annual Report Series, Canberra: AusInfo. Productivity Commission (2003), Regulation and Its Review 2002–2003, Annual Report Series, Canberra: AusInfo. Productivity Commission (2005), Review of National Competition Policy Reforms, Report No 33, Canberra: Productivity Commission. Rimmer, S. (2006), ‘Best Practice Regulations and Licensing as a Form of Regulation’, Australian Journal of Public Administration, 65(2), 3–15. Samuel [National Competition Council] (1999), Review of Sections 51(2) and 51(3) of the Trade Practices Act 1974, Final Report, Canberra: CanPrint Communications. Scotchmer, S. (1991), ‘Standing on the Shoulders of Giants: Cumulative Research and the Patent Law’, Journal of Economic Perspectives, 5(1), 29–41. Skehill, S. (1992), ‘Competition Policy’, in Economic Planning Advisory Council, Competition and Economic Efficiency, Background Paper No. 19, Canberra: Australian Government Printing Service. Stedman, J. (1973), ‘Patents and Antitrust: The Impact of Varying Legal Doctrines’, Utah Law Review, 588–617.


125

Stonier [Industrial Property Advisory Committee] (1984), Patents, Innovation and Competition in Australia, Canberra: Australian Government Printing Service. Trade Practices Act Review Committee (1976), Report to the Minister for Business and Consumer Affairs, Canberra: Australian Government Printing Service. Trade Practices Commission (1991), Application of the Trade Practices Act to Intellectual Property, Background Paper, Canberra: Trade Practices Commission. Transfield Pty Ltd v Arlo International Ltd (1980), 144 CLR 83 (Barwick CJ, Stephen, Mason, Wilson and Murphy JJ). Treasury (2003), Commonwealth Government Response to the Review of the Competition Provisions of the Trade Practices Act 1974, Canberra: Treasury. United States Department of Justice and the Federal Trade Commission (1995), Antitrust Guidelines for the Licensing of Intellectual Property, Washington, DC: United States Department of Justice and the Federal Trade Commission. Weisbrot [Australian Law Reform Commission] (2004), Genes and Ingenuity: Gene Patenting and Human Health, Report No. 99, Sydney: SOS Printing Group Pty Ltd.

PART II

Open intellectual property systems

7. An introduction to open source biotechnology Janet Hope I.

INTRODUCTION

This chapter provides an introduction to the emerging phenomenon of ‘open source biotechnology’ (OSB). It explores a number of questions commonly raised by innovators, intellectual property (IP) practitioners and end users about the desirability and feasibility of translating the open source model of software development into the life sciences. For example: What is OSB? What specific problems is it intended to solve? What are the key elements of an open source approach outside the software context? In the absence of proprietary control over their contributions, why would self-interested actors choose to participate in open sourcestyle collaboration? How does OSB address problems of excessive proprietary control over knowledge inputs? Are there any real-world examples of OSB? And finally, what are the major challenges in implementing this approach? The responses provided to these questions are not intended to be conclusive. Rather, they are offered as a conversation starter for anyone interested in exploring possible applications of the open source model in a new context. For this reason, the discussion assumes no more than a basic knowledge of the relevant fields.

II.

WHAT IS OPEN SOURCE BIOTECHNOLOGY?

OSB is a nascent movement comprising scholars in the disciplines of law, economics and innovation management, scientific researchers and managers in non-profit and for-profit institutions, and social entrepreneurs. Although it is possible to identify a number of individuals who are actively working to make OSB a reality (see VIII below, ‘Are there any real-world examples of OSB?’), this is not a movement defined by its leaders. Instead, it takes the form of a diverse community of interest centred on the translation of open 129

130


source principles from software to the life sciences. Similarly, while proponents may receive government, philanthropic and/or corporate funding to develop their own versions of OSB, the concept itself is not associated with any particular organisation. At the theoretical or conceptual level, OSB has its roots in a tradition of sceptical IP scholarship that pre-dates the umbrella term ‘intellectual property’ itself. This tradition is currently enjoying a revival in the form of a concern with protecting the ‘intellectual commons’ from encroachment by various forms of privatisation across a range of technology and content categories (Boyle 2003). In the life sciences, dramatic recent expansion of IP protection for biotechnology inventions has sparked concerns about a potential ‘tragedy of the anticommons’ (Heller and Eisenberg 1998), in which the owners of multiple complementary intellectual assets are unable to strike a mutually satisfactory deal. The concerns extend to the apparent anti-competitive effects of overly broad patents and/or consolidated patent portfolios (for example, Walsh, Arora and Cohen 2003; Graff, Rausser and Small 2003). OSB represents one of a number of proposed strategies for pre-empting or counteracting some of the adverse effects of increasing proprietary control over this broad enabling technology (see III below, ‘What problem is open source biotechnology intended to solve?’). At a practical level, OSB is a novel approach to managing research collaborations, licensing IP and commercialising or otherwise exploiting new technologies. The OSB name and concept are derived from the free/libre and open source software (FLOSS) movement that has emerged over the past two decades in the field of information and communications technology. In its original context, ‘open source’ refers to software source code – that is, the version of a software program that can be read, understood and modified by human beings. ‘Open source’ means that source code is (1) readily accessible to anyone who receives a copy of the binary or executable version of the relevant software program, and (2) free from legal encumbrances commonly found in software copyright licence agreements. The term ‘open source’ when applied to biotechnology research and development is not straightforwardly descriptive because there is no direct technical equivalent of software source code in the biotechnology context. Instead, use of the term ‘open source’ is intended to signal a conscious modelling process in which the institutional design elements of open source technology development, licensing and commercialisation are translated into a new setting.

An introduction to open source biotechnology

131

III. WHAT PROBLEM IS OPEN SOURCE BIOTECHNOLOGY INTENDED TO SOLVE? OSB is sometimes thought of primarily as a solution to the tragedy of the anticommons. While the preconditions for anticommons tragedy – namely, excessively fragmented property rights and high transaction costs – are certainly present in many areas of biotechnology research and development, the empirical question of whether such tragedy has actually occurred has no clear or universal answer (Murray and Stern 2005; Epstein and Kuhlik 2004; Nicol and Nielsen 2003; Walsh, Arora and Cohen 2003; Straus, Holzapfel and Lindenmeir 2002). If OSB were narrowly focused on preventing or relieving a tragedy of the anticommons, this uncertainty might cast doubt on the need for OSB. However, it is more accurate to think of OSB as a broad-spectrum treatment for a disease of the innovation system caused by excessive proprietary control over information goods. This disease disrupts the flow of the technologyrelated information among people and institutions which is central to the innovative process. And anticommons tragedy is but one of a variety of possible symptoms whose presentation depends on the structure and dynamics of the affected industry.

IV. HOW, THEN, DOES THIS IP-MEDICATED DISEASE MANIFEST ITSELF IN BIOTECHNOLOGY AND RELATED INDUSTRIES? In the field of biomedicine, less than 5 per cent of global expenditure on biomedical research and development is directed towards developing treatments for serious diseases that affect millions of the world’s poorest people every year, demonstrating that biotech and other drug patents do not generate a sufficient incentive to invest in the most socially valuable health-related innovation (Commission on Intellectual Property Rights 2002, pp. 30–34). While some drugs developed primarily for rich country markets can also be used to treat diseases prevalent in poor countries, IP rights keep prices high, both directly and indirectly (by introducing inefficiencies of the kind that may contribute to anticommons tragedy). IP rights also suppress competition, thereby negatively affecting both affordability and availability of existing treatments (Commission on Intellectual Property Rights 2002, pp. 30–34). In the field of agricultural biotechnology, the use of molecular biotechnologies to facilitate conventional plant breeding and introduce new farmerfriendly crop traits has been touted as a solution to problems of poverty and

132


food insecurity (for example, see Conway and Toennissen 2003). In reality, however, most developing countries have little capacity for agbiotech research; meanwhile, international public sector funding for agricultural research has been stagnant or declining for several decades, leaving developed country researchers narrowly focused on domestic priorities (Pray and Naseem 2003, pp. 4–10; Commission on Intellectual Property Rights 2002, p. 60; Anon. 2001). As in the case of biomedicine, private sector research and development priorities in both developed and developing countries tend to reflect the needs of large commercial operations that target mass markets in rich countries. IP rights have not made any positive difference to these incentives, yet anticommons tragedy is reportedly imminent in the public sector, while in the private sector stark industry consolidation (driven at least in part by the need to avoid such a tragedy) has drastically reduced competition (Atkinson et al. 2003; Graff et al. 2003). Not only have new traits that might be useful to poor farmers and consumers not been forthcoming, but also traditional agricultural practices that save farmers money and help them conserve genetic diversity have been reframed as IP infringement (Commission on Intellectual Property Rights 2002, pp. 68–9). OSB is not the only possible treatment for this IP right-mediated disease. In biotechnology, other proposed solutions include patent law reforms (for example, Barton 2000), changes to the patent administration system (Drahos 2007), establishment of intellectual property clearinghouse mechanisms and patent pools,1 a renewed commitment to open access policies on the part of public sector institutions and increased public funding for socially valuable research and development. Efforts to promote greater information sharing in biotechnology also form part of a larger reaction against what legal scholar James Boyle has termed the ‘second enclosure movement’ (Boyle 2003), in which IP rights are used to bring art, music, film, literature, science, technology and other information and cultural goods under private, often corporate, control. In this sense, attempts to implement OSB form part of a broad network of initiatives, including the original FLOSS movement, Creative Commons (www.creativecommons.org) and its sister initiative Science Commons (www.sciencecommons. org), Wikipedia (en.wikipedia.org), Public Library of Science (www.plos.org), Electronic Frontier Foundation (www.eff.org), P2P Foundation (p2pfoundation.ning.com), Open Hardware and many others.

V. WHAT ARE THE KEY ELEMENTS OF AN OSB APPROACH? By analogy with FLOSS, OSB incorporates a distinctive approach to (1)


133

developing new biotechnologies; (2) managing intellectual and other property rights relating to those technologies, specifically through open source licensing; and (3) commercialisation (or, more broadly, technology delivery). A.

Technology Development

The characteristic mode of development of new biotechnologies in OSB is bazaar governance (Demil and Lecocq 2006), also termed ‘commons-based peer production’ (Benkler 2002; see also Benkler 2006). Defining features of this mode of production include: 1. 2. 3.

4.

5.

integration of contributions from a broad range of participants (individuals, commercial firms and state or private non-profit organisations); diverse motivations to contribute, none of which rely on exclusionary property rights (see below ‘B. Open source licensing’); coordination of contributions through self-selection for tasks on the basis of freely accessible information about the object of production itself, as distinct from direction by superiors in a hierarchy or self-selection on the basis of price signals; low barriers to engagement: technology users are free to become developers according to their own incentives and capacities, irrespective of formal qualifications and without the necessity of making any long-term commitment; and sharing of contributions on terms that permit broad (though not necessarily unlimited) freedom of access and freedom to operate (see below ‘B. Open source licensing’).

In real-life OSB projects, as in FLOSS, bazaar governance would predominate but would coexist with firm, market and network structures enforcing a mix of other (private) incentives. These private incentives could be further supplemented by collective action-style incentives and public subsidies of the kind that already support biotechnology research and development (Hope 2008, pp. 106–41). B.

Open Source Licensing

OSB is distinct from ‘open access’ or ‘public domain’ regimes in which innovators refrain from obtaining intellectual property rights or other forms of proprietary or quasi-proprietary protection for their innovations. In OSB, as in FLOSS, contributions are legally owned by innovators or their assignees but are licensed to all comers on terms that, at a minimum, offer a credible commitment to provide ongoing access to the licensed technology and guarantee a

134


level playing field between licensor and licensee with respect to freedom to operate with the licensed technology. The second point is crucial: open source licences are intrinsically – and radically – pro-competitive. Optionally, OSB licences may also impose an obligation on licencees to onlicence improvements to the technology on the same terms as the original licence. Such an obligation would normally apply only to a defined subset of improvements and would not apply to any improvements that are reserved by the licensee/developer solely for in-house use. This mechanism, inspired by a subset of FLOSS licences known as ‘copyleft’ or ‘reciprocal’ licences, is designed to give both licensors and licensees (in their capacity as licensors of any improvement technologies) ongoing access to a dynamic, legally protected technology commons incorporating updated versions of the original technology as well as any new technologies based upon it. C.

Commercialisation

A key feature of the OSB approach that distinguishes it from pre-existing forms of bazaar governance (typified by the practices underpinning traditional publicly funded scientific research) is the nature of the relationship it fosters between commercial and non-commercial contributors. Traditional bazaar-style production has frequently involved the transfer of commercially valuable technologies from the public and non-profit sectors without fair compensation and/or adequate recognition of the substantial public subsidy thereby indirectly provided to private/corporate players. In OSB, the relationship between commercial and non-commercial participants would be synergistic and symbiotic, not parasitic. In other words, noncommercial contributions would provide opportunities for commercial players to enhance their profits while commercial players would bring private resources to the production of public knowledge. (See the next section for information on how commercial players can profit by participating in OSB initiatives.)

VI. WHAT INCENTIVES ARE THERE FOR SELFINTERESTED ACTORS TO PARTICIPATE IN OSB PROJECTS? The strongly pro-competitive terms of OSB licences effectively limit the income an innovator can derive from licence fees and royalties. However, there are other ways for a self-interested actor, such as a commercial firm, to capture private rewards from its contribution to an OSB project. By analogy


135

with FLOSS, any contributor to an OSB project may be motivated by the anticipation of (1) process benefits; (2) user benefits; (3) non-proprietary, but still potentially lucrative, business opportunities; and/or (4) restructuring competition within a market. A.

Process Benefits

Some private research and development investments are motivated by expectations, not of a particular outcome, but of benefits to be derived from engaging in the innovative process. In the FLOSS context, process benefits include fun, learning and social interaction with like-minded peers (Ghosh and Prakash 2000). These factors may intuitively seem less relevant in biotechnology; but in fact, biotechnology industry participants do sometimes invest in research and development activity solely or primarily for the sake of analogous process benefits. For example, some commercial firms encourage staff to engage in discretionary collaborations in which they can acquire and maintain skills, build their professional reputations and connect with a community of peers (Drews 1999, pp. 229–32). From the individual staff member’s perspective, these motivations are essentially the same as those that drive individual academic or hobbyist contributions to FLOSS development. From the firm’s perspective, the point is to attract and retain high quality staff, a matter of commercial significance in any knowledge industry. At the corporate level, potential process benefits of participating in OSB development include organisational learning, developing and maintaining a wide range of inter-organisational linkages, and establishing a corporate reputation as a competent and trustworthy collaborator. All are commercially relevant in biotechnology and related industries, where the ability to engage in productive collaborations across firm boundaries is crucial to retaining a competitive edge (Powell 2001, pp. 252ff.). B.

User Benefits

Companies and other institutions (such as universities) that hope to use a technology in-house – in a research programme, for example, or as a component of a production process – may contribute to its development for the sake of gaining access to a useful tool, irrespective of the technology’s exchange value in the marketplace. The primary rationale for choosing an OSB approach in this context is that it permits prospective users to share the cost and risk of developing a new technology – and, depending on the chosen licence terms, potentially gain free access to valuable downstream technologies. Further, an OSB approach may

136


enhance the usefulness of the technology by expanding the number of people who can help eliminate design flaws and introduce new features. In some cases, the mere fact that a technology is available on open source terms and is therefore accessible to a larger number of users can enhance its usefulness even in the absence of quality improvements. This is true for any technology whose use is subject to positive network externalities (‘network effects’). In the FLOSS context, technical interoperability is the main driver of network effects. Interoperability is also important in relation to some biotechnologies, but it is not the only possible source of network effects. Another is the need to comply with regulatory requirements – for example, in clinical trials for new biological drugs or in field trials for genetically modified organisms. C.

Non-Proprietary Business Opportunities

Mainstream or conventional biotechnology business strategies typically rely on property rights as a means of excluding non-paying users from accessing or commercialising the technology. By contrast, non-proprietary business strategies are designed to exploit the enhanced value to a firm’s clients or customers of technologies whose use, modification and redistribution are relatively unrestricted – and which are therefore likely to be cheaper, more readily available, better quality (for example, in the sense of being more reliable due to extensive peer review) and/or more transparent than proprietary technologies. Analyses of successful FLOSS businesses have identified a number of viable non-proprietary business strategies in which the non-market transfer of open source technologies is coupled with other, market-based transactions in order to boost the value of those transactions (Perens 2005). Arguably, the most successful businesses are those that combine non-proprietary strategies with other proprietary and/or non-proprietary strategies as part of a mixed overall business model. Non-proprietary business strategies that have parallels in biotechnology and related industries include (Perens 2005; Hecker 2000): 1.

2.

Service provision: a technology is distributed on an open source basis in order to expand the market for services such as training, consulting, custom development and after-sales support. An example in biotechnology would be the service provision aspects of the DArT business model (see VIII below, ‘Are there any real-world examples of OSB?’). Market positioning: an open source product that itself generates little or no revenue helps build the firm’s overall brand and reputation and increases the number of technology developers and users who are famil-


3.

137

iar with and loyal to the product line as a whole. In biotechnology, for example, a contract research organisation might choose to ‘open source’ a research tool developed in the course of its work with a client in order to establish a reputation for creative problem solving that will help it to attract new clients. Other applications of this strategy include brand licensing and franchising, in which a fee is charged for the right to use brand names and trademarks associated with technology that is not itself proprietary. Leveraging OSB technologies to add to the appeal of complementary proprietary products: a common approach in FLOSS is to ‘open source’ an enabling or platform technology and then distribute complementary niche or application technologies on proprietary terms. For example, a manufacturer of biotechnology assay kits might make the assay protocol available on open source terms in order to increase the number of users of the assay, thereby boosting demand for its kits.

It might be assumed that non-proprietary strategies are inherently less lucrative than proprietary strategies, but it depends on the context. Both proprietary and non-proprietary strategies carry actual costs and opportunity costs that should be weighed against one another in any rational strategic decision-making process. IBM’s experience with FLOSS is one example of a commercial firm deriving greater benefit from a non-proprietary versus a proprietary approach. Despite having amassed tens of thousands of patents over the past decade, as of 2003 IBM’s non-proprietary ‘Linux-related services’ revenue was reportedly double that from patent-related sources – of the order of billions of dollars (Benkler 2006, p. 46). D. Restructuring Competition A further possible motivation for commercial firms to invest private resources in generating public knowledge is the potential for non-proprietary technologies to be used as a competitive weapon. A crucial feature of open source is that it promotes free and open competition with respect to the technology in question. Although commercial players generally prefer to protect themselves from competition wherever possible, they always want to ensure competition among their suppliers. In the pharmaceutical context, for example, cutthroat rivals were prepared to collaborate in the creation of a public database of human genetic markers (via the SNP Consortium) in order to ensure that the data did not fall under the proprietary control of smaller biotechnology firms (or each other) (Holden 2002; see also Pisano 2006). Other ways an OSB approach might be used to restructure the competitive

138


landscape in an industry sector to contributors’ advantage include creating the opportunity for several smaller firms to combine resources against a larger competitor or attracting customers away from established technology providers (Hecker 2000). In considering the possible uses of OSB as a competitive weapon, it is important to recognise that exclusionary uses of property rights always create both winners and losers. Ex post, the winner of a technology race will naturally favour a winner-takes-all rule. But everyone else, from the immediate runners-up right down to the wooden spoon recipient, would do better to share both costs and benefits. If it is not clear ex ante who the winner will be, all competitors have something to gain from OSB as an alternative to the higher risk proprietary strategy. Even if an OSB approach is not adopted ex ante, all those who do not turn out to be winners will then have a common interest in inventing around or otherwise undermining the commercial value of the winner’s technology. OSB offers a way for them to collaborate in achieving this goal.

VII. HOW DOES OSB ADDRESS PROBLEMS OF EXCESSIVE PROPRIETY CONTROL OVER R&B INPUTS? When anyone who wants to contribute to technology development has both the legal freedom to do so and the ability to exercise that freedom (that is, access to the technology in a transparent, readily modified form), then would-be users can adapt the technology to their own needs, irrespective of the size of the market those needs represent; the technology may evolve more quickly, becoming progressively more robust and reliable; and because distribution is subject to free competition, the technology is likely to be both more affordable and more readily available from a range of sources. Further, because the process of technology development in OSB is transparent and there is no proprietary restriction on the number of people who are able to familiarise themselves with the technology, inherently dangerous technologies can be more easily and effectively regulated. (Note that lifting proprietary restrictions on participation in biotechnology research and development does not entail lifting biosafety or other regulatory restrictions.) An earlier section (III above, ‘What problem is open source biotechnology intended to solve?’) sought to clarify the relationship between OSB and the problem of anticommons tragedy. While OSB is not merely a response to anticommons tragedy, it is of interest to consider the extent to which OSB could relieve or reverse an incipient tragedy of the anticommons.


139

Anticommons tragedy in biotechnology is triggered by high transaction costs associated with assembling a ‘tool kit’ composed of multiple proprietary elements. If the owner of one of these proprietary elements were to adopt an OSB strategy, then the overall transaction costs would, in principle, be proportionally lowered because anyone, anywhere would then be sure of obtaining permission to use, modify and distribute the relevant technology element for any purpose without ongoing obligations to the owner. (Any copyleft-style obligation is owed to all prospective licensees of an improvement technology, not just the original licensor.) Note that although OSB entails fewer transaction costs than a conventional proprietary approach, it does not eliminate IP-related transaction costs altogether. Unlike a fully ‘public domain’ or ‘open access’ approach, OSB does involve IP ownership and does require technology users to enter into a licence contract. In addition, copyleft-style open source licences place conditions on the distribution of downstream technologies that, despite their non-proprietary purpose, do limit freedom to operate with respect to those technologies. Whether open source licensing of just one element of an otherwise proprietary tool kit could lower overall transaction costs enough to reverse an imminent tragedy of the anticommons depends on the circumstances. Non-proprietary distribution of a single tool kit element does not automatically dissolve proprietary barriers with respect to other elements, but it does change the environment in which other technology owners and users make their own strategic decisions. For example, FLOSS pioneer Richard Stallman’s early efforts to generate a non-proprietary tool kit quickly attracted new investments as other users recalculated the trade-off between accepting proprietary restrictions and collaborating to invent around proprietary tools (Weber 2004, p. 49). If anticommons tragedy is not yet imminent in a particular field, OSB offers a way to pre-empt such a tragedy by establishing a robust commons with respect to basic or fundamental technologies whose value is likely to be enhanced by cumulative innovation. What might OSB do to address the distinct but related issue of overly broad patents on basic (platform or infrastructure) technologies? If the platform is non-substitutable, as in the case of genomic sequence data, OSB can do little once a proprietary choke-hold has already been established. However, it is an excellent way for actors dedicated to maintaining ongoing public access and freedom to operate to prevent this outcome. If the platform is not impossible to invent around, then OSB has something to offer even after the platform has fallen under proprietary control for the reasons given earlier (see above ‘D. Restructuring competition’).

140


VIII. ARE THERE ANY REAL-WORLD EXAMPLES OF OSB? OSB is still at a relatively early stage of its evolutionary development. Realworld initiatives do exist, but none constitutes a mature working example of OSB in the sense of having developed, licensed and commercialised a viable new biotechnology in accordance with fully open source principles. This section provides some examples of past and present biotechnology initiatives that are consciously modelled on aspects of FLOSS. A.

General Public Licence for Plant Germplasm

Proposed by plant scientist Professor Tom Michaels in 1999, the General Public Licence for Plant Germplasm (GPLPG) is a fairly straightforward adaptation of the archetypal ‘copyleft’ software licence, the GPL, to germplasm (the indispensable raw material of plant breeding). It comprises four key elements (Michaels 1999). First, anyone can release germplasm under the terms of proposed material transfer agreement, unless the material is already encumbered. Second, a sample of all germplasm released under the GPLPG must be deposited in an existing publicly accessible collection of plant gene resources. The owner is permitted to charge for the physical act of reproducing and distributing the germplasm, and may also charge a royalty to a third party for the right to reproduce and distribute the unmodified germplasm. Third, any germplasm descended from a GPLPG strain may be released and distributed freely, provided that the resulting material is also designated as GPLPG and accompanied by full pedigree information. Fourth, release and distribution of novel germplasm based wholly or in part on GPLPG plant germplasm are prohibited, except as expressly provided under the GPLPG. Despite having been well received within Michaels’ immediate research community, the GPLPG proposal languished for lack of persistent advocacy. Nevertheless, Michaels believes its spirit has been partially adopted by International Treaty on Plant Genetic Resources for Food and Agriculture (ITPGRFA) provisions establishing a multilateral system of access and benefit sharing.2 Commercial products based on plant genetic resources within the system may be developed and sold without compulsory sharing of profits with the developers of those resources provided the resulting product is made available without restriction to others for further research and breeding, an arrangement that is reminiscent of the GPLPG. B.

Public Sector Human Genome Project

Not publicised until after it had already been abandoned, 1999 saw another


141

proposal for OSB-style licensing in the context of the public–private race to complete a draft sequence of the human genome (Cukier 2003). Concerned that their commitment to make their own data publicly available gave private sector competitors an opportunity to free ride on their efforts, public sector researchers at the UK’s Sanger Institute explored the possibility of licensing the data on GPL-like terms. A draft licence was produced, but it was ultimately decided that a copyleft-style constraint on data use would not be acceptable to the Sanger Centre’s collaborators (Sulston and Ferry 2002, pp. 211–12). C.

International HapMap Project ‘Click-Wrap’ Database Licence

Before December 2004, researchers wishing to access the online database of the International HapMap Project, a private–public collaboration established to create a haplotype map of the human genome, had to agree to a set of ‘clickwrap’ conditions intended to discourage database users from filing patent applications that would block other users’ access to the data (for a discussion of the policy, see Gitter 2007). According to a notice on the site, the software GPL directly inspired the terms of the click-wrap agreement. Unfortunately, in this case the implementation of OSB terms was flawed (Gitter 2007; Eisenberg 2006, p. 1027). In the absence of clear intellectual property rights in the data, only those who had agreed to the contractual terms of the click-wrap agreement were obliged to observe restrictions on patenting. To overcome this problem, the agreement incorporated provisions seeking to prevent users of the data from releasing it to anyone not bound by the same terms. The undesirable consequence was that publications based on the data could not be properly peer reviewed. At the same time, users were permitted to file for patent protection on specific haplotypes perceived to be critical for future product development, thereby compromising the goal of creating a protected commons as a resource for downstream research and development. D.

Tropical Diseases Initiative/The Synaptic Leap

The Tropical Diseases Initiative (TDI) is an open source-style drug discovery scheme proposed in 2004 by lawyers Stephen Maurer and Arti Rai and computational biologist Andrej Sali (Maurer, Sali and Rai 2004). The scheme aims to reproduce the advantages of FLOSS technology development with respect to the identification of new drug targets and candidates by designing research projects to employ computational methods wherever possible so as to minimise the need for wet lab experiments. The TDI framework does not directly address later stages of drug development. Rather, it is anticipated that the selection and further development of promising drug candidates will be

142


coordinated by non-profit virtual pharmaceutical organisations such as the Drugs for Neglected Diseases Initiative (www.dndi.org). The TDI proposal does not include a specific OSB licence; instead, researchers are encouraged to develop their own licences on a case-by-case basis. TDI’s publicised efforts have so far taken place mainly through its partnership with The Synaptic Leap (TSL), a non-profit organisation founded in 2005 by former commercial software industry executive Ginger Taylor (Taylor 2006). TSL’s goal is to extend the range of scientific collaborations beyond researchers’ personal networks by establishing a usable online open information and communications infrastructure. Researchers making use of this resource have tended to go down the ‘open access’ or ‘public domain’ route rather than seeking intellectual property protection for their contributions (Taylor, personal communication 19 November 2006). E.

BioBricks Foundation

The BioBricks Foundation is a non-profit organisation based at the Massachusetts Institute of Technology and dedicated to promoting and coordinating the production of standardised modules or ‘biobricks’ that can be assembled into functional synthetic biological systems (openwetware.org/ wiki/The_BioBricks_Foundation). Some of the earliest writings on OSB (for example, Carlson 2001) emanated from synthetic biologists seeking to ensure the viability of the enormous collaborative effort needed to realise their engineering goals; as of 2008, key questions of intellectual property management in this growing field remain unsettled and an open source solution is still on the table (Maurer 2008). One challenge in this case is to identify a property right in the biobrick modules that can be effectively harnessed to create a protected commons (Rai and Boyle 2007). F.

Diversity Arrays Technology Pty Ltd

Diversity Arrays Technology (DArT) is a privately held Australian company established in 2001 to commercialise a patented molecular marker technology invented by its founder and Director, Andrzej Kilian. The company’s nonproprietary, service-based business model has been inspired by that of Red Hat, the successful Linux distribution company whose initial public offering (IPO) made Wall Street history in August 1999 (Kilian, personal communication 15 September 2006; see also Goldman-Rohm 1999). Instead of relying on injections of funds from capital markets supplemented by licensing revenue, the company offers genotyping services for a range of crops, custom development of DArT arrays for new species, a technology-specific laboratory information management system, and technology packages that can be used by


143

others to ‘port’ the DArT technique to new crops or use environments. Its business model is designed to exploit the synergy between service provision and ongoing development of the technology: progressive deployment in a range of settings is seen as the key to expanding the market and enhancing the technology’s use value through a process of iterative learning. Technology-specific know-how and a strong scientific reputation are crucial to maintaining a competitive edge. DArT offers a concrete illustration of the distinctive commercial logic underpinning OSB, described above. The company has something to gain by increasing the number of users of its technology, even if those users do not pay licence fees, because a larger user base expands the market for its related offerings. As a company that devotes significant resources to high risk research, DArT benefits by sharing costs and risks with a group of external user-developers: hence, it has established a ‘DArT network’ (www.diversityarrays.com/ dartnetwork.html) for the exchange of innovation-related information and materials among developers and service providers working in diverse locations on diverse crops. The company’s stated aim is to use any intellectual property rights arising from ongoing research and development to attract partners to the network rather than to restrict access to the relevant technology. Building a user community helps build a reputation for cutting-edge innovation that makes it easier to attract investors, collaborators and quality staff. As a company that uses its own technology in the provision of goods or services, DArT also benefits directly from any improvements in the use value of the technology. The longer the firm stays in the field that it has helped create through its own innovation, the more it builds on its initial competitive edge. The result is a thriving small business with a growing income stream and international customer base. G.

Biological Innovation for Open Society

Biological Innovation for Open Society (BIOS) is an initiative of CAMBIA (Center for Application of Molecular Biology in International Agriculture), an independent non-profit research institute that combines wet lab development of agricultural biotechnology research tools with intellectual property informatics and policy development. The BIOS initiative, launched in 2005, incorporates the ‘Patent Lens’, a searchable patent database containing ancillary information and tutorials; ‘BioForge’, a portal for protocol sharing, comments on patents, and public and private online discussion modelled on the FLOSS site Sourceforge.net; and ‘Biological Open Source’ (‘BiOS’) licensing of some elements of CAMBIA’s own IP portfolio (www.bios.net/daisy/bios/ home.html). Products derived from BiOS-licensed technology may be patented and commercialised subject to a number of constraints, including the

144


requirement that a broadly defined class of improvements be non-exclusively granted back to CAMBIA.3 Although CAMBIA’s BiOS licences explicitly invoke the language of open source, its CEO and Chairman Richard Jefferson cautions against treating open source as anything more than a metaphor in this case. In fact, were they software licences, the BiOS licences would not be approved by the relevant FLOSS standard setting and certification bodies, principally because the terms of the grant-back give CAMBIA too much control over downstream research and development (Hope 2006). Perhaps this is one reason why there is no evidence of significant uptake (Maurer 2007, p. 11 fn. 73). H.

Network for Open Scientific Innovation

Network for Open Scientific Innovation (NOSI) is a recently formed nonprofit corporation based in Brazil. Inspired in part by the example of CAMBIA and led by American social entrepreneur Joseph Jackson III and Brazilian lawyer Mauricio Guaragna, NOSI aims to provide support for legal and scientific development and training in Brazil, as well as advocating for its own (planned) programmes. These include ‘Open Source Biotechnology’ and ‘Capacity Building in Technology Management’. At present, the former programme looks set to revolve around the implementation of open source approaches to stem cell research and development. I.

Equitable Access and Neglected Disease Licensing

‘Equitable Access’ and ‘Neglected Disease’ licensing are twin elements of another OSB proposal put forward by an interdisciplinary group of Yale University scholars led by postdoctoral researcher Amy Kapczynski (Kapczynski et al. 2005). These licensing models explicitly envisage universities as licensors but could be adopted, for example, by other researchers and institutions operating within the TDI framework described above. Equitable Access licensing aims to improve access to biomedical innovations in low income and middle income countries. Under an Equitable Access licence, a university could, for a fair royalty payment, grant a nonexclusive licence to use its patented technology to a commercial firm to produce a research tool for sale in poorer countries. In return, the licensee would agree to grant back to the university any improvements it might make to the research tool and to cross-licence any other rights the licensee holds that might be used to block its production. These rights would be extended automatically on the same terms to any third party notifying an intention to produce the research tool for sale in low to middle income countries. The licence would not set a specific price for such sale, but the price would be kept low by competition


145

from other licensees. The inducement for prospective licensees to agree to a non-exclusive licence on these terms would be the opportunity to market the licensed technology in rich countries. Under a Neglected Disease licence, a university licensing out its technology would retain the right to license the technology for the purpose of research on neglected diseases anywhere in the world, and for commercial purposes in poorer countries. There are various possible formulations: the licensee might be required to grant back and cross-license its own intellectual property; the research exemption might be confined to non-commercial institutions, or to a definite list of neglected diseases, or to diseases that meet a general standard (for example, as set out in US legislation defining a rare condition). In all cases the purpose would be to facilitate research on orphan or neglected diseases. J.

Slow Food Project on Open Source Biodiversity Conservation

In Tuscany, University of Florence geneticist Marcello Buiatti is working with the Italian Slow Food organisation to develop ways to protect traditional Italian crop varieties and derivatives from the threat of biopiracy (Buiatti, personal communication 11 June 2007). One proposal, still in its infancy, is to develop an open source community for crop biodiversity conservation. The initiative would be designed to dovetail with Tuscan regional government moves to implement laws for the support of ‘conserver farmers’ and Italian government approval for the establishment of a ‘conserved varieties register’; it would be centred around local farmers’ co-operatives or individual farmers interested in the conservation and valorisation of forgotten varieties. Buiatti reports that the project’s supporters include activist Vandana Shiva and Jose Esqunas Alcazar, an author of the ITPGRFA treaty (Buiatti, personal communication 11 June 2007). It is hoped that an OSB approach, if successful, could serve as a model for similar efforts in developing countries, as well as helping to promote a move away from environmentally unsustainable monoculture in developed countries. K.

FLOSS Programs for Use in Biotechnology Settings

In the post-genomic age, software programs have become key research tools in biotechnology and related industries. Some of the most powerful and widely used software tools for handling the large and complex data sets produced by molecular biologists are open source in the FLOSS sense. These include the National Center for Biotechnology Information (NCBI) tool kit (incorporating BLAST), scripting language projects such as BioPerl and advanced projects such as BioLisp and hidden Markov model libraries. Other well-known open

146


source bioinformatics software projects include BioJava, BioPipe, BioRuby and BioPython, DAS, MOBY, OBDA and EMBOSS. The FLOSS/OSB approach is particularly suitable for bioinformatics software because bioinformatics applications generally incorporate similar fundamental features but often require extensive customisation before they can be applied to specific data sets. Unlike proprietary programs, FLOSS/OSB bioinformatics tools can be re-used in other contexts but also freely modified to suit unique use environments. It might be supposed that the popularity of open source bioinformatics software is best regarded merely as an aspect of the broader FLOSS phenomenon and should not be taken as any indication of the future prospects of OSB. However, open source bioinformatics may turn out to be significant as a nonproprietary wedge driven into the culture of an industry that has historically been strongly committed to proprietary strategies. If biotechnology industry participants are already willing to embrace non-proprietary tools in specific contexts, that willingness may gradually extend to other contexts – especially as biotechnology research and development comes to rely more and more heavily on computer modelling and data analysis and the industry continues to experience an influx of researchers with ‘hard’ science backgrounds who are familiar with FLOSS and carry with them FLOSS-influenced expectations regarding their ability to access and use the tools of their trade.

IX. WHAT ARE THE MAJOR CHALLENGES IN IMPLEMENTING OSB? As noted earlier (V above, ‘What are the key elements of an OSB approach?’), OSB is an instance of bazaar governance or commons-based peer production. A variety of objections to the feasibility of OSB – for example, relating to the technical differences between software and biotechnology production – are readily answered by reference to the fact that traditional publicly funded research and development is also an instance of bazaar governance that continues to underpin the present-day life sciences innovation system as a whole – the ‘traditional’ or ‘academic biobazaar’ (Hope 2008, p. 205). Since the traditional biobazaar is clearly viable as a means of producing biotechnology research and development, only aspects of the OSB-style biobazaar that differ from the traditional biobazaar require proof of feasibility. One obvious difference between the traditional biobazaar and FLOSS – and hence, perhaps, OSB – is the extent to which each relies on, and makes use of, the Internet. In fact, web-based collaboration tools are increasingly used in mainstream biotechnology research and development, and OSB initiatives such as TDI/TSL (see VIII above, ‘Are there any real-world examples of


147

OSB?’) are exploring the potential for enhancing Internet-mediated science and technology production still further. In any case, however, the success of OSB as a distinctive mode of technology production does not depend on its becoming as thoroughly digitised and widely distributed as FLOSS. The point is rather that restrictions on widespread participation that are by-products of the conventional proprietary approach to both software and biotechnology production should be lifted. A second difference is the use of open source licensing in FLOSS/OSB, compared with the primarily ‘open access’ or ‘public domain’ regime that has characterised the traditional biobazaar. It might be argued that the attempt to import FLOSS licensing practices into biotechnology research and development is unnecessary, on the ground that, unlike copyright, patent protection does not arise automatically. The point is that FLOSS licences could be regarded as a necessary evil, since software programmers own copyright in their work whether or not they intend to adopt a non-proprietary strategy, whereas biotechnology innovators can simply refrain from obtaining property rights in their inventions in the first place. It is true that an open access approach may often be appropriate in biotechnology where it would not have been an option in an otherwise equivalent software case. However, there are a number of reasons why biotechnology innovators who wish to pursue a non-proprietary strategy may also want or need to use an open source-style licence (Hope 2008, pp. 158ff): 1. Ownership rights do arise automatically with respect to some components of many biotechnologies. Biological materials such as cell lines and germplasm constitute personal property. Software code, data, written protocols or other components of a biological innovation may be subject to copyright protection. In such cases, a licence clarifies the owner’s intention to make the technology available on non-proprietary terms. 2. A patent owner may wish to open source his or her innovation at the end of its life cycle, after its value as a generator of proprietary licensing revenue begins to decline. Granting an open source licence is then an alternative to simply abandoning the patent. 3. The owner of a biotechnology innovation may intend to adopt a copyleftstyle licence. In FLOSS copyleft licensing, copyright is used as a mechanism for compelling follow-on innovators to contribute to a protected commons. Any similar mechanism in the biotechnology context would have to rely on the existence of a prior IP right: one cannot impose conditions on the distribution of a technology one does not own. 4. In a highly proprietary environment, such as that which prevails in many areas of biotechnology, failure to assert ownership over a technology before making it available for public use leaves open the possibility that

148

5.

6.


someone else will patent the technology and pursue a proprietary exploitation strategy, to the detriment of the original inventor as well as other users. A technology owner may wish to retain the option of making the technology available under more than one licence – an approach known in the FLOSS context as ‘dual licensing’ (Hope 2008, pp. 125–6). While everyone must be free to access open source technology under an open source licence, some licensees prefer more tailored terms, for example because they wish to avoid incurring copyleft obligations. Only an intellectual property owner has the ability to offer prospective licensees a choice between non-proprietary and proprietary alternatives. IP rights are frequently treated as a signal that an innovation has substance and is worth supporting; in consequence, some development pathways may be facilitated by the existence of an IP right even if the owner has no intention of pursuing either a proprietary or a copyleft-style licensing strategy. For example, access to seed funding to develop a prototype technology may be conditional on having filed a patent application even if the grants body does not concern itself with commercialisation.

Assuming that licensing is an integral aspect of OSB, there are two major challenges associated with developing OSB licences (Hope 2008, pp. 142–87). The first is to interpret the principles of FLOSS licensing in a way that makes sense in a different legal, technical and commercial environment, yet preserves functionally important institutional design elements. Some experimentation has already taken place along these lines, but in the absence of any licensing standard specific to OSB, it is difficult for prospective users (whether licensors or licensees) to judge the quality of licences intended or described as OSB licences. The second challenge is to develop a body of licensing practice and a set of standards that could address this problem. In the FLOSS context, although the term ‘open source’ itself is not protected, the Open Source Initiative (OSI: a non-profit corporation established in 1998) has registered the words ‘Open Source Initiative Approved License’ as a service mark. The OSI administers a certification programme for software licences based on compliance with a written standard: the Open Source Definition (OSD), the latest version of which is available at www.opensource.org/docs/osd. The OSD is widely accepted among free and open source software developers as a codification of many years’ trial-and-error learning about which distribution terms best permit developers to harness ‘the power of distributed peer review and transparency of process’ (OSI homepage, www.opensource.org). In developing an equivalent standard for OSB, some adaptation would obviously be required, but arguably changes should be kept to the minimum


149

necessary to create a meaningful translation of open source principles from one context to another. The reason for adhering as closely as possible to open source software principles in the articulation of OSB as a concept is not that the open source approach is assumed to be superior to all others, nor that dilution or modification of these principles in the implementation of real-life open source-style biotechnology collaborations is assumed to be undesirable. The rationale for taking a purist stance at the conceptual level is that prospective participants in such collaborations will be better able to identify and discriminate among a variety of institutional design options if those options are coherent and clearly distinguished and their respective merits and demerits separately enunciated. A third difference between the traditional biobazaar and OSB, as noted in Part V above (‘What are the key elements of an OSB approach?’), is the integration of both non-commercial and commercial contributions on the same non-proprietary terms. In principle, there is no reason why commercial players would not be motivated to contribute to OSB development in the same way that they have helped to drive FLOSS. However, until a substantial number of profit-motivated corporations or individuals come forward as OSB sponsors or participants, it is not clear under what circumstances the trade-off between proprietary and non-proprietary strategies, and between open access/public domain and OSB strategies, will come out in favour of OSB. Even more of an obstacle, at least for the time being, is the uncertainty associated with OSB as an untried and untested approach. This uncertainty undoubtedly has the potential to deter innovators who might otherwise perceive advantages in OSB. One goal of scholarly research and writing in this area is to break down the barrier of unfamiliarity and encourage those who might have something to gain from this novel approach to give it a try, learning by doing what does and does not work in their own environments. Other actors who wish to encourage such experimentation could do so in a range of ways. For example, government agencies and other institutions that sponsor research and development could consider giving overall priority and/or targeted funds (to meet the costs of establishing collaborative infrastructure and drafting new open source-style licences) to biotechnology projects that incorporate aspects of the open source model. Ideally, such funding would be provided on terms that encourage project leaders to reflect on their experience with the open source approach and communicate that experience to others, creating a learning loop. Similarly, patent offices could adopt a policy of expediting the patent application process in relation to applicants who undertake to make their IP available on open source terms (Drahos, personal communication 30 November 2007). Intellectual property experts could provide pro bono advice to would-be OSB licensors to help them develop appropriate licence provisions.4 Industry associations whose

150


members would benefit from access to cheap, high quality technologies could offer financial or in-kind support to new OSB businesses (Hope, Nicol and Braithwaite 2008). End users of biotechnology-related products and their agents, including government procurers and health insurance companies, could signal a market preference for products developed using open source methods.

X.

CONCLUSION

Proponents of OSB seek to realise an approach to biotechnology research and development that is efficient, participatory and firmly oriented towards the needs of users. While the implementation of OSB still entails significant uncertainty, this approach has already attracted innovators and investors willing to experiment with different versions of the open source model. Together, these initiatives form the kernel of an open source movement in biotechnology. If the challenges outlined above can be overcome, such a movement has the potential to revolutionise the conduct of life sciences research and development just as open source software has revolutionised information technology.

NOTES 1.

2.

3. 4.

On patent pools, see for example Clark et al. (2000); Gaulé (2006); Verbeure et al. (2006); Ebersole et al. (2005); Goldstein et al. (2005); and Grassler and Capria (2003). On intellectual property clearinghouses, see for example Graff et al. (2001); Graff and Zilberman (2001a); Graff and Zilberman (2001b); Atkinson et al. (2003); van Zimmeren et al. (2006); and Horn (2003). Tom Michaels, personal communication via email, 23 November 2006. The International Treaty on Plant Genetic Resources for Food and Agriculture (the treaty) was adopted by the thirty-first session of the Conference of the Food and Agriculture Organisation of the United Nations on 3 November 2001 and entered into force on 29 June 2004. Articles 10 to 13 (Part IV) of the treaty establish a Multilateral System to facilitate access to Plant Genetic Resources for Food and Agriculture and to share, fairly and equitably, the benefits arising from the use of these resources. Article 12.4 provides that facilitated access under the Multilateral System shall be provided pursuant to a Standard Material Transfer Agreement, which was adopted by the Governing Body of the treaty in its Resolution 1/2006 of 16 June 2006. The terms of the agreement that resemble the reciprocal terms of a copyleft-style licence are contained in Article 6 (‘Rights and Obligations of the Recipient’). Updated provisions are available online at www.bios.net/daisy/PELicense/751.html; the present analysis refers to CAMBIA’s BiOS License for Plant Enabling Technology Version 1.3, Recitals, first paragraph. For example, Public Interest Intellectual Property Advisors Inc. (http://www.piipa.org/) is a non-profit organisation dedicated to making intellectual property counsel available to developing countries and public interest organisations seeking to promote health, agriculture, biodiversity and science.


151

REFERENCES Anon. (2001), ‘Last Word: Grassroots Innovator’, Economist Technology Quarterly, 2001(4), 27–8. Atkinson, R.C. et al. (2003), ‘Public Sector Collaboration for Agricultural IP Management’, Science, 301, 174–5. Barton, J.H. (2000), ‘Reforming the Patent System’, Science, 287, 1933–4. Benkler, Y. (2002), ‘Coase’s Penguin, or, Linux and the Nature of the Firm’, Yale Law Journal, 112, 369–446. Benkler, Y. (2006), The Wealth of Networks: How Social Production Transforms Markets and Freedom, New Haven: Yale University Press. Boyle, J. (2003), ‘The Second Enclosure Movement and the Construction of the Public Domain’, Law and Contemporary Problems, 66, 33–74. Carlson, R. (2001), ‘Open Source Biology and Its Impact on Industry’, IEEE Spectrum, www.kurzweilai.net/meme/frame.html?main=/articles/art0613.html. Clark, J., J. Piccolo, B. Stanton and K. Tyson (2000), ‘Patent Pools: A Solution to the Problem of Access of Biotechnology Patents?’, US Patent and Trademark Office, Washington, DC. Commission on Intellectual Property Rights (2002), ‘Integrating Intellectual Property Rights and Development Policy’, Final Report, Commission on Intellectual Property Rights, London. Conway, G. and G. Toennissen (2003), ‘Science for African Food Security’, Science, 299, 1187–8. Cukier, K. (2003), ‘Open Source Biotech: Can a Non-Proprietary Approach to Intellectual Property Work in the Life Sciences?’, www.cukier.com/writings/ opensourcebiotech.html. Demil, B. and X. Lecocq (2006), ‘Neither Market nor Hierarchy nor Network: The Emergence of Bazaar Governance’, Organization Studies, 27(10), 1447–66. Drahos, P. (2007), ‘Trust Me: Patent Offices in Developing Countries’, CGKD Working Paper Series, Centre for Governance of Knowledge and Development, Australian National University, www.anu.edu.au/fellows/pdrahos/workingpapers/index.php (forthcoming in American Journal of Law and Medicine). Drews, J. (1999), In Quest of Tomorrow’s Medicines: An Eminent Scientist Talks about the Pharmaceutical Industry, Biotechnology and the Future of Drug Research, New York, NY: Springer-Verlag. Ebersole, T., Guthrie, M. and J. Goldstein (2005), ‘Patent Pools and Standard Setting in Diagnostic Genetics’, Nature Biotechnology, 23, 937–8. Eisenberg, R. (2006), ‘Patents and Data Sharing in Public Science’, Industrial and Corporate Change, 15(6), 1013–31. Epstein, R.A. and B.N. Kuhlik (2004), ‘Is There a Biomedical Anticommons?’, Regulation, 27(2), 54–8. Gaulé, P. (2006), ‘Towards Patent Pools in Biotechnology?’, Innovation Strategy Today, 2, 123–4. Ghosh, R.A. and V.V. Prakash (2000), ‘The Orbiten Free Software Survey’, First Monday 5(7), www.firstmonday.org/issues/issue5_7/ghosh/. Gitter, D. (2007), ‘Resolving the Open Source Paradox in Biotechnology: A Proposal for a Revised Open Source Policy for Publicly Funded Genomic Databases’, Houston Law Review, 43(5), 1475–1521. Goldman-Rohm, W. (1999), ‘Inside the Red Hat IPO’, Linux Magazine, www.linuxmag.com/id/348/.

152


Goldstein, J., Ebersole, T. and M. Guthrie (2005), ‘Patent Pools as a Solution to the Licensing Problem of Diagnostic Genetics’, Drug Discovery World, Spring, 86–90. Graff, G. and D. Zilberman (2001a), ‘An Intellectual Property Clearinghouse for Agricultural Biotechnology’, Nature Biotechnology, 19, 1179–80. Graff, G. and D. Zilberman (2001b), ‘Towards an Intellectual Property Clearinghouse for Agricultural Biotechnology’, Intellectual Property Strategy Today, 2001(3), 1–12. Graff, G.D., G. Rausser and A.A. Small (2003), ‘Agricultural Biotechnology’s Complementary Intellectual Assets’, Review of Economics and Statistics, 85, 349–63. Graff, G., A. Bennett, B. Wright and D. Zilberman (2001), ‘Intellectual Property Clearinghouse Mechanisms for Agriculture: Summary of an Industry, Academia, and International Development Round Table’, IP Strategy Today, 2001(3), 12–30. Graff, G., A. Heiman, C. Yarkin and D. Zilberman (2003), ‘Privatization and Innovation in Agricultural Biotechnology’, Agriculture and Resource Economics Update, 6(3), 5–7 (Giannini Foundation of Agricultural Economics, Berkeley). Grassler, F. and M. Capria (2003), ‘Patent Pooling: Uncorking a Technology Transfer Bottleneck and Creating Value in the Biomedical Research Field’, Journal of Commercial Biotechnology, 9(2), 111–18. Hecker, F. (2000), ‘Setting Up Shop: The Business of Open-Source Software’, hecker.org/writings/setting-up-shop. Heller, M.A. and R.S. Eisenberg (1998), ‘Can Patents Deter Innovation? The Anticommons in Biomedical Research’, Science, 280, 698–701. Holden, A.L. (2002), ‘The SNP Consortium: Summary of a Private Consortium Effort to Develop an Applied Map of the Human Genome’, Biotechniques, June 2002 Supplement, 22–4. Hope, J. (2006), ‘Some Questions about BiOS Licences’, message posted to CAMBIA BioForge BIOS-compliant licenses Discussion Forum, www.bioforge.net/forge/ thread.jspa?threadID=533&tstart=0, 16 November. Hope, J. (2008), Biobazaar: The Open Source Revolution and Biotechnology, Cambridge, Mass: Harvard University Press. Hope, J., D. Nicol and J. Braithwaite (2008), ‘Regulatory Capitalism, Business Models and the Knowledge Economy’, in J. Braithwaite, Regulatory Capitalism, Cheltenham, UK and Northampton MA, USA: Edward Elgar. Horn, L. (2003), ‘Alternative Approaches to IP Management: One-Stop Technology Platform Licensing’, Journal of Commercial Biotechnology, 9, 119–27. Kapczynski, A., S. Chaifetz, Z. Katz and Y. Benkler (2005), ‘Addressing Global Health Inequities: An Open Licensing Approach for University Innovations’, Berkeley Technology Law Journal, 20, 1031–114. Maurer, S.M. (2007), ‘Open Source Drug Discovery: Finding a Niche (or Maybe Several)’, University of Missouri at Kansas City Law Review, 76, 2–33. Maurer, S.M. (2008), ‘Open Source’, paper presented at the Microsoft-Hitotsubashi University Symposium on Law and Economics of Intellectual Property, Tokyo, 18 February. Maurer, S.M., A. Sali and A. Rai (2004), ‘Finding Cures for Tropical Diseases: Is Open Source the Answer?’, Public Library of Science: Medicine, 1(3), e56, www.plosmedicine.org. Michaels, T. (1999), ‘General Public Release for Plant Germplasm: A Proposal. Version 1.1’, www.oac.uoguelph.ca/www/CRSC/pltag/1998-99/gnucrop2.htm, 26 February (discontinued; version 1.2 of the proposal, dated October 1999, is on file with the author).


153

Murray, F. and S. Stern (2005), ‘Do Formal Intellectual Property Rights Hinder the Free Flow of Scientific Knowledge? An Empirical Test of the Anti-Commons Hypothesis’, NBER Working Paper Series No. 11465, National Bureau of Economic Research, Cambridge MA, http://www.nber.org/papers/W11465. Nicol, D. and J. Nielsen (2003), ‘Patents and Medical Biotechnology: An Empirical Analysis of Issues Facing the Australian Industry’, Centre for Law and Genetics Occasional Paper Series No. 6, Centre for Law and Genetics, University of Tasmania, Hobart. Perens, B. (2005), ‘The Emerging Economic Paradigm of Open Source’, First Monday, Special Issue 2 (Open Source), www.firstmonday.org/issues/special10_10/perens/ index.html. Pisano, G. (2006), ‘Profiting from Innovation and the Intellectual Property Revolution’, Research Policy, 35, 1122–30. Powell, W. (2001), ‘Networks of Learning in Biotechnology: Opportunities and Constraints Associated with Relational Contracting in a Knowledge-Intensive Field’, in R.C. Dreyfuss, D. Zimmerman and H. First (eds), Expanding the Boundaries of Intellectual Property: Innovation Policy for the Knowledge Society, Oxford: Oxford University Press, pp. 251–66. Pray, C.E. and A. Naseem (2003), ‘The Economics of Agricultural Biotechnology Research’, ESA Working Paper Series No. 03-07, Economic and Social Department, Agricultural Economics Division, Food and Agriculture Organisation, Rome. Rai, A. and J. Boyle (2007), ‘Synthetic Biology: Caught between Property Rights, the Public Domain, and the Commons’, Public Library of Science: Biology, 5(3), e58. Straus, J., H. Holzapfel and M. Lindenmeir (2002), ‘Empirical Survey on Genetic Inventions and Patent Law’, presented at the OECD Expert Workshop on Genetic Inventions, Intellectual Property Rights and Licensing Practices, Paris, OECD. Sulston, J. and G. Ferry (2002), The Common Thread, London: Random House. Taylor, G. (2006), ‘The Synaptic Leap Open Source Biomedical Resource: An Introduction’, paper presented at Science Food Camp, 11–13 August 2006, www.thesynapticleap.org/?q=node/122. van Zimmeren, E., B. Verbeure, G. Matthijs and G. van Overwalle (2006), ‘A Clearing House for Diagnostic Testing: The Solution to Ensure Access to and Use of Patented Genetic Inventions?’, Bulletin of the World Health Organization, 84, 337–424. Verbeure, B., E. van Zimmeren, G. Matthijs and G. van Overwalle (2006), ‘Patent Pools and Diagnostic Testing’, Trends in Biotechnology, 24, 115–20. Walsh, J.P., A. Arora and W.M. Cohen (2003), ‘Effects of Research Tool Patents and Licensing on Biomedical Innovation’, in W.M. Cohen and S.A. Merrill (eds), Patents in the Knowledge-Based Economy, Washington, DC: National Academies Press, pp. 285–340. Weber, S. (2004), The Success of Open Source, Cambridge, Mass: Harvard University Press.

8. Intellectual property, innovation and openness Ulf Petrusson and Caroline Pamp I. INTELLECTUAL PROPERTY TO PROMOTE INNOVATION AND DEVELOPMENT: SETTING THE SCENE When we discuss whether intellectual property has a positive or negative impact on innovation and development, we tend to focus on intellectual property rights (IPRs) regulation. We ask ourselves if patent regulation has an economically positive impact or not. Those of us who have a positive attitude tend to frame the discussion as a discussion of IPR protections as incentives for creativity and innovative activities. We tend to discuss how IPR regulations attract foreign investment and foreign out-sourcing. A conclusion is often that the protection against piracy is a cornerstone of a globalized economy. If we instead are more sceptical towards IPRs, it is more relevant to frame the discussion as a question of whether patent regulation and the other IPRs only work in the interests of the financially strong actors in the industrial world. We tend to discuss the risks that the TRIPs provisions will result in blocking: (1) knowledge sharing and technology transfer, (2) usage of traditional knowledge, (3) research efforts, (4) indigenous innovation, (5) diagnoses, treatments and cures, and (6) ultimately development overall. That we focus on regulation means that if we are critical we will question the IPR regime per se as well as whether the IPR provisions are well balanced. On the level of provisions, we will question the criteria of protectability, the scope, the time period possible to uphold the IPR, the exemptions and the possible remedies (see Petrusson 1999). However, if we limit the scope of the question and for example ask ourselves whether intellectual property is important in promoting start-up activities in bioscience research, it becomes obvious that we cannot really answer the question by discussing the regulation and its overall impact.1 We need to focus on how intellectual property concepts are used and can be used in research and innovation (see Petrusson 2003). Then, it starts to become clear that the usage of intellectual property can either promote or block devel154

Intellectual property, innovation and openness

155

opment. In a concrete situation, a claim of a patentable invention in bioscience can be used to block as many other actors as possible. It can also be used to ensure openness and to promote the diffusion of the technology. The licence mechanism offers a number of different ways to enable access to research results. This chapter is built on the assumption that the question of whether intellectual property will promote development is primarily a question of how we use the concepts of intellectual property in research and development (R&D) and in business. This does not mean that we consider it to be unimportant how we handle the balance in the regulation of IPRs. However, we will argue the importance of going beyond an understanding of intellectual property as IPRs in regulation. We will analyse intellectual property as typified objects in transactions of the intangible, that is, of the non-physical. The research question is how the claiming of property and property transactions in early stages can promote innovation as well as openness. We will focus on the role of IPRs in relation to bioscience. Dealing with development and innovation, we are on the one hand interested in understanding the institutional mechanisms that enable the transformation of knowledge processes in bioscience into assets, property and capital and, on the other hand, interested in how different norm structures affect openness and knowledge sharing. Drawing on the experiences from open source and open innovation thinking, we know that intellectual property can be used as a means to build platforms for openness. Thus, in this chapter we shall analyse how R&D and innovations in biotechnology can be governed through the medium of different licence mechanisms, both as transactions of intellectual property and as processes where we promote access, openness in research and openness in utilization. The empirical basis of the research is a number of contractual models used in the bioscience setting.

II. THE CREATION OF ASSETS, PROPERTY AND CAPITAL IN KNOWLEDGE: ELABORATING ON A THEORETICAL FRAMEWORK The starting-point of this chapter is the challenge of transforming research and other knowledge processes into wealth and welfare. In more or less all business sectors based on creativity and new knowledge, we experience a structural transformation where we struggle to find new property-based value propositions and business models (Petrusson 2004).2 In the biopharmaceutical sectors, for example, we elaborate on diagnostic tools and research tools. Many of us wonder whether the claiming of property in early research settings will promote a sustainable wealth and welfare development. At the same time,

156


the development very much illustrates the potential of claiming property in biotechnology, whether and when to construct: (1) products and services, (2) commercial transactions and relations, (3) firms, (4) markets, and also (5) platforms for research and knowledge sharing. Hereby, we are for example moving away from an approach to patents as a means to block others and towards an approach where we use patents to package value propositions to others. Furthermore, for those of us deriving from an industrial setting, it is relatively easy to recognize that capital mechanisms are not adjusted for knowledge-based business. Banks and other actors in the financial machinery are reluctant to recognize the outcome of bioscience as securities. Intellectual assets related to biotechnology are difficult to have recognized as capital in accounting. It does not take much to end up in a bankruptcy situation. In bankruptcies and other insolvency situations, values originating in biotechnology easily disappear. In a market economy setting, where our ambition is to create more and more knowledge-based business, a major challenge is to transform knowledge into assets, property and capital. We are dependent on conceptual frameworks, in particular when to claim: (1) what is valuable for us (assets), (2) what we offer as tradable objects (property) and (3) what should be recognized as objects in the financial machinery (capital). How these frameworks are developed is directly related to how the future knowledge-based market economy will be constructed (Petrusson 2004; Alänge et al. 2006). An overall challenge will be to manage the fact that intellectual assets, properties and capital are intellectual, that is, they do not exist in themselves (Glavå and Petrusson 2002).3 This fact will always mean that the claiming of assets, property and capital in knowledge and other intellectual processes can never be the same as claiming identical processes in the physical. When we claim control of the intellectual, this will generate different restraints than when we claim control of the physical object. In the claiming of property in the intellectual, we will always have to handle the challenge of controlling the level of openness in access and use. It is however not obvious that the claiming of intellectual property will restrict the level of openness; it could possibly also increase it. In theory at least, there will always be a question of how the concepts are constructed and how they are used in the specific context. In the context of business and innovation, it is obvious that intellectual properties are social constructions – constructions that are intertwined with the experienced existence of other social constructions, such as the firm, associations, financial institutions, markets and the like (see Searle 1995). Berger and Luckmann ([1966] 1991, p. 90) explain that ‘the origins of any institutional order lie in the typification of one’s own and others’ performances’. Focusing


157

on patentable inventions and patents in biotechnology, they can be described as typified intellectual objects (intellectual building blocks) created as a result of the communicative interaction of activities within four arenas. Contextually, we know that patent concepts, as communicative means, have different receivers. The patent as a substantive concept means different things to the patent examiner, the judge in an infringement case, the judge in a contract-related case, and the manager in a knowledge-oriented firm (Petrusson 2003). However, this does not necessarily result in confusion. On the contrary, the different communicative processes can be seen as the constructive foundation for the typified patents in the business context. Thus, a fundamental theoretical challenge when analysing the typification of intellectual property in a development setting is to evaluate how it interacts with the typification of other social constructs such as innovations and ventures (see Lundqvist and Petrusson 2002). In this chapter, we are especially interested in how the typification of intellectual property enables the typification of licence structures for the promotion of openness.4 We are interested in the role of intellectual property in the establishment of network-based platforms for knowledge transfer, where the establishment of openness is of fundamental value.

III. CONTRACTUAL PACKAGING OF PROPERTY TRANSACTION: DEFINING THE EMPIRICAL AMBITION Public research organizations (PROs) all over the world are increasingly faced with the challenge of patenting in the field of bioscience (see Eisenberg 1996). For example, according to the programme ‘Responsible Use of Public Research’, European universities are encouraged to develop their activities based on a strategic attitude to intellectual property. One emphasizes inter alia the importance of that collaboration. Partners offer equal rights to research results in order collectively to create research platforms and to favour the long-term public interest on ethical grounds. Gaining the full benefits of Open Science and Open Innovation requires some changes in the way that knowledge generated by PROs is handled and protected. To the question: ‘Why should PROs protect intellectual property’ appropriate answers are: ‘To encourage the economic applications of their discoveries for the benefit of the public’ and ‘To make the research function more attractive and better supported’. To the question: ‘How should this property be protected’, it is appropriate that the PRO takes steps itself, provided that these steps reflect good understanding of where and how the benefits can accrue. (European Commission 2005, p. 3)

158


In the international arena, more and more projects are developed where ‘good understanding’ is a means to developing capabilities to use patent systems and copyright systems to facilitate openness. This action can be seen as a response to a situation where industry and also universities make strategic use of intellectual property ‘carpets’ (in the current discussion, the concept of ‘patent thickets’ is often used; see Shapiro 2000). The opportunity to claim inventions ‘upstream’, that is, in early research phases, has without question generated a pressure to patent. But it is also something more. Copyrights, databases and other IPRs are already used to create platforms for openness. Increasingly, PROs do also recognize the challenge to apply for patents not only to protect inventions but also to promote openness.5 Just by elaborating on the intellectual property mechanism as a building block to establish openness in a bioscience setting, we can distinguish between a number of different possibilities when it comes to establishing openness; for example: • open for evaluation without restrictions; • open for research without restrictions; • open for development without restrictions; • open for research and/or development under payment (with or without fair and reasonable terms); • open for development under payment and with grant back obligation (with or without fair and reasonable terms); • open for commercialization under payment (with or without grant back); • open for development and open distribution/commercialization – ‘an open source model’; • open for development, open distribution and open grant back – ‘another open source model’; and • open without intellectual property claims – ‘traditional open science model’ (Petrusson 2007). The claiming of intellectual property is increasingly used in order to create collective research processes and research platforms. It constitutes an essential part of the development of ‘open arenas’ and ‘open innovation’. Many of the initiatives that so far have been launched can also be characterized as experimental and are in relatively early phases of their development. In this chapter, we will discuss three of these initiatives. We limit the scope of our empirical study so as to focus on licence models developed to promote openness. Particularly we analyse the contractual building blocks specifically designed to facilitate openness. We will analyse one model that is a response to aggressive intellectual property behaviour – Cancer Research UK. We will also


159

analyse an open source model to promote openness in research and utilization in bioscience – the BiOS example. Finally, we will analyse a licence model that differentiates levels of openness in research and utilization – the example of PIPRA.

IV.

CASE STUDIES

A. A Licence Model to Ensure Openness as a Response to Aggressive IP Behaviour: the Cancer Research UK Example Background Myriad Genetics is a company in the field of diagnosing and treating cancer, well known for obtaining a rather strict control over its inventions through patents, databases and other control mechanisms. In the USA, the company has a rather broad patent portfolio controlling research on and testing for the genes BRCA1 and BRCA2, mutations of which are responsible for some hereditary breast cancer. Myriad Genetics’ business model is built on having very few licensees (some parts, such as final sequencing, are not licensed at all), thus ensuring quality control but also excluding both research hospitals and patients. Our subject, Cancer Research UK, was afraid of the consequences of this strategy. Since open science did not offer the same control mechanisms, they decided, and managed, to get a patent on BRCA2 in Europe, which later was held by Cancer Research Technology Limited (controlled by its parent charity, Cancer Research UK; now referred to as CRTL). The purpose was to ensure access to further research through licensing strategies. Therefore, CRTL created a specific licence agreement to be used for not-for-profit research laboratories and publicly funded research laboratories. It is important to note that the purpose, both with the patenting and licensing strategies, was not commercial or profit-driven. On the contrary, the aim was to play by the rules and still reach the objective of ensuring access to further research.6 Object and scope The licence object is limited to patents or, more specifically, one patent. No other intellectual property rights or any other objects are included. This is therefore a rather limited licence agreement, where the object is clearly defined through the packaging in the patent. The patent is referred to by its patent number and since it is a patent granted by the European Patent Office, the European countries in which the patent is valid are listed in an appendix. The scope of the licence is non-exclusive, meaning that CRTL can choose to give out as many other licences as they wish, thus ensuring access to many

160


more licensees. Price is normally an issue as regards exclusivity or nonexclusivity, but since this agreement does not have any profit objective, this becomes a non-issue. Scope can also be limited by the use of technology area, geographical area, type of activity (for example, research, sale and distribution), but the only restriction of the scope is that it shall fall under the patent. The licensee must undertake not to engage in any profit-making activities which fall under the scope of CRTL’s patent. Hence, the purpose of the licence agreement is to ensure access for further research and directly and indirectly benefit patients. However, the right to improvements is not regulated in this agreement. Obligations of the licensor In this agreement, it is notable that the licensor has considerably fewer obligations than in a commercial licence agreement. CRTL has full discretion to decide whether or not to maintain and prosecute the patent rights. This is perfectly reasonable given that there is no licence fee. However, it would have been possible to include an option clause for any licensee. The clause would, if the licensor chose not to maintain or prosecute any patent, get the patent transferred on the condition that the licensee would: maintain the patent and become the new licensor (with the obligation of being bound to keep licensing and also issue new licences to the patent on the same conditions), or prosecute the patent at its own expense (without getting the patent transferred). For this to work, the licensor would need to have an obligation to inform all licensees. Perhaps this could be done through the licensor’s home page, since that would be an efficient communication channel given that the number of licensees could be quite large. This is also the solution chosen in the agreement for the annual activity report. There could also be a stipulation that the licensor should have the possibility to evaluate prospective new licensor(s) and, on each occasion, choose the one most suitable for the task. If these provisions are not included, there is a risk that the access strategy is too strongly connected to one actor. It is very understandable that CRTL does not want to have the duty to assert the patent (or keep it), but this could also be considered a weakness in the business arena. Does this mean that commercial entities will be less likely to obtain a (commercial) licence, given their knowledge of the provisions of this agreement? This weakness could be counteracted by the knowledge that there are others who may step in. Control mechanisms There are no royalty mechanisms in the agreement, but the fact that there is no payment does not mean that it is not possible to regulate obligations and rights of both parties. Open does not equal free and free does not equal open. Thus, although the agreement is open for all actors who are not-for-profit or publicly


161

funded research laboratories, they are not allowed to perform the diagnostic tests for BRCA2, unless they have signed a copy of the licence agreement, sent it to CRTL and received the licence agreement signed by CRTL, too. Furthermore, CRTL also has the right to request information which they consider necessary to confirm the non-profit status of the licensee. This is one control measure. The other important control mechanism is a duty for the licensee to report to CRTL the number of BRCA2 diagnostic tests undertaken annually. The licensee does not have the right to grant sub-licences, which means that it is not possible for not-for-profit actors to further license the patent. By having this control mechanism, CRTL has ensured that they alone will control the not-for-profit status. B. Open Source Model to Promote Openness in Research and Utilization in Bioscience: The BiOS Example Background CAMBIA is a non-profit research institute which invents and shares technologies and also creates new practices of intellectual property management. One of its initiatives is Biological Innovation for Open Society (BiOS), which among other things has created a draft licence agreement for plant enabling technology, where the parties are CAMBIA and the BiOS Licensee.7 This agreement is also open for other actors to use. To a large extent, the agreement is based on open source structures. There are, however, a few notable differences between open source (based on copyright claims) and open patents (based on patent claims). First, there is a cost issue concerning making the claims. Copyright does not cost anything to obtain or maintain, but to assert copyright can be very expensive. Patents, on the other hand, require resources to obtain, maintain and assert. Second, there is a disclosure requirement for patents which does not exist for copyright. Third, in some countries the patent system has an experimental use exemption which allows for experiments on a patented invention. Fourth, the scope of protection differs (copyright is more narrow than patents). Thus, the experiences from open source concerning software are not completely analogous to patents. Object and scope The licence object is rather broad, comprising patents, materials, technology data and any improvements. The patents are explicitly licensed non-exclusively worldwide in the field of agriculture, and there is no royalty. Thus, there can be an indefinite number of licensees to the agreement. Right to improvements There is a rather extensive section on improvements in the agreement. If any

162


licensee comes up with a technology being a plant enabling technology improving the licensed patent but which cannot be exercised without infringing a licensed patent, CAMBIA has a right to a worldwide, non-exclusive and royalty-free licence, including the right to sublicense to other licensees. This means that access is facilitated not just to the original patent(s), but also to further developments thereof. This is a very important contractual mechanism that has traditionally been used for open source. It is exactly this mechanism that is lacking in the Cancer Research UK agreement. However, if an improvement should take place in a country where CAMBIA does not have a patent, it is not possible to exercise any legal control over the improvements. Certainly, this could be done on a voluntary basis, and many non-profit organizations are probably willing to share access anyway, but there would not be any contractual duty to so. Obligations of the licensor Neither party to the agreement has a duty to maintain or prosecute any intellectual property rights against any third party. However, should the BiOS licensee wish CAMBIA to assert its intellectual property rights against a third party, and should CAMBIA agree to this (they have full discretion to decide whether or not to assert any intellectual property rights), the BiOS licensee shall cover all costs. This though, only concerns assertion of IPRs, not maintaining them. Control mechanisms Sub-licensing is allowed to a certain extent, but on the condition that any improvements are considered to be improvements under this agreement. The BiOS licensee has a duty to submit a list of sub-licensees to CAMBIA in writing annually. It is, however, unclear why any sub-licensing should be necessary, as it would be possible to approach CAMBIA directly. Furthermore, the BiOS licensee must submit a detailed description of all improvements annually. This description shall include any filed patent application, submitted publication, technology data, descriptions of compositions, methods, materials, protocols and the like – anything that is necessary to practise the improvement and also, if applicable, in sufficient quantities so that CAMBIA would be able to use the improvement material, and even propagate the material in order to make it available for other licensees. C. A Licence Model to Differentiate Levels of Openness in Research and Utilization: The Example of PIPRA Background Public Intellectual Property Resource for Agriculture (PIPRA) is an organiza-


163

tion with the vision to strategically manage IPRs in the field of agricultural life sciences, and to promote research, access and distribution of subsistence crops for humanitarian purposes in the developing world by means of collaborative efforts of universities and non-profit research institutes. Object and scope PIPRA has created a model clause including three definitions (not an entire agreement) in order to help universities when licensing to keep a right to the invention, regardless of how it is covered by IPRs, for humanitarian purposes.8 When using this clause, any licence agreement the university enters into must be non-exclusive. PIPRA defines humanitarian purposes as: ‘a) the use of Invention/ Germplasm for research and development purposes by any not-for-profit organization anywhere in the World that has the express purpose of developing plant materials and varieties for use in a Developing Country, and b) the use of Invention/Germplasm for Commercial purposes, including the use and production of Germplasm, seed, propagation materials and crops for human or animal consumption, in a Developing Country’ (PIPRA homepage http:/www.pipra.org). There are certain important differences in this clause compared with the other two agreements. First, the distinction is not made purely between profit purposes and not-for-profit purposes. Instead, there are two distinctions. The first one applies worldwide and concerns R&D purposes by any not-for-profit organization, if the express purpose is to develop plant materials and varieties for use in a developing country. Hence, the purpose of the R&D is decisive. The second distinction concerns commercial purposes in developing countries, the right to which is also covered by the clause. Second, the other two agreements use contractual mechanisms to license out a certain invention, thus facilitating access directly. The PIPRA clause works the other way around: instead of licensing out a certain invention, this contractual mechanism is used to ensure that the university has the right to license out the invention for humanitarian purposes as defined by PIPRA at a later stage. This, too, facilitates access indirectly. If it was not for this clause in an agreement, the university could have issued exclusive licences and thus would not have had the freedom to issue licences for humanitarian purposes, which would decrease access.

V.

DISCUSSION

Gradually, as business becomes knowledge-driven, it becomes increasingly clear how one always must consider the degree of openness – concerning both

164


access and use. The more pressure to claim intellectual property, the more important it becomes to develop capabilities to claim intellectual assets in order to ensure openness (Petrusson 2001; Pamp and Petrusson 2004).9 The empirical examples analysed above constitute responses to increasingly aggressive claiming activities, whereas BiOs and PIPRA have taken the lead in the creation of a new logic in which research platforms for openness must be constructed; our discussion below will relate to which different kinds of capabilities are necessary in order to create such platforms. When one is on the brink of developing capabilities to create openness by the use of property transactions, it becomes clear that openness is a multifaceted process in itself. The idea behind ‘open science’ is that publishing makes research results publicly available and thus open. This idea can in practice constitute the basis for academic activities also when one, as for example in the USA, prescribes that universities shall apply for patents on research results if possible.10 Openness is in this regard restricted to freedom of access to the publication in question, but as soon as one has the intention of using the result in any context, a licence agreement must be concluded. Furthermore, in practice only a very small group of academic researchers has the capacity for real access to published research results. The fact that a particular research result is published within the framework of an ‘open science’ model does not necessarily mean anything except that it is hypothetically possible for a handful of persons to get access to the knowledge. Consequently, a lesson from the ‘open science’ era is that we must distinguish between potential openness and actual openness – that is, we must distinguish between situations where the knowledge assuredly is public but in practice nobody has access to it, and where channels are established in practice. Another lesson is that we should distinguish between openness from an accessibility perspective and openness from the perspective of the possibility of use. To get access to a certain result without being able to use the result, perhaps not even in research activities, is definitely something completely different from being able to use available research results without restrictions. In having knowledge of research relations channelled into property transactions, it is further reasonable to distinguish between openness without cost and openness requiring financial contribution. The fact that openness requires fees or royalties does not in itself mean that openness is restricted. It is quite possible that the purpose of taking out a certain royalty for a licence to use a particular research result is to obtain sufficient resources to create and nurture an open knowledge platform, but this is not always the case, and in order to be able to evaluate the differences between these situations, capabilities to evaluate openness are required (see Schollin 2008). Furthermore, it is also possible to distinguish between openness that presupposes access to a particular platform or ‘community’ and openness that does not. With the purpose of


165

establishing an open knowledge exchange between research groups, one can for example create an association where one can be offered access under certain conditions. A.

Creating Openness with as Few Restrictions as Possible

To create total openness for everyone, without any restrictions whatsoever, is consequently not possible to implement. Instead, one must aspire to acquire capabilities to create openness with as few actual restrictions as possible. Such capabilities include for example the capability of creating fee-based licence constructions, the purpose of which is to stimulate actors to commit to a platform where open access and use is offered, as well as the capability to establish ‘communities’ as a feature in the establishment of open knowledge channels. Furthermore, patenting is in this context a capability that can be used in order to ensure that research results shall be available to as large an extent as possible and shall also be used by all who wish to do so (open patents). A complementary capability is to be able to operate agreements in order to ensure that everybody who uses a research result for which a patent has been applied for must offer a licence free of cost to all improvements that are created by them. Using the claiming context in which one exists as a startingpoint, the idea is to see how these capabilities can be used to stimulate a balanced openness with the help of different strategies, organizational solutions, transactions and the like. B.

Creating Openness Within One’s Own Research Group

In a worst-case scenario, a transaction-based attitude can result. Research colleagues within the same research group negotiate the extent to which the knowledge that they have developed and even will develop, shall be made accessible to, and become used by the others. A research colleague might for example have created a database or an information technology research tool, and s/he can be very eager to control this research result. Thus, even within one’s own research group, capabilities are necessary in order to create a balanced openness. The individual researchers’ claiming ambitions can very well paralyse the creativity of an entire group, which means that capabilities for ‘title clearance’ and contractual governance of ownership claims are consequently important to create and maintain openness. C.

Creating Openness Within Research Projects

An important capability is to create a balanced degree of openness in research collaboration and contract research. To what extent should all actors involved

166


get access to research results and to what extent should they be allowed to use the results? How should access be created and what should be required for use? Possibly, a balanced openness could be achieved whereby everyone is allowed to use the result for further research but pays a reasonable royalty for commercialization activities. Capabilities to use project agreements in order to create a balanced openness are thus important. When the collaboration is financed by public funding, for example by the European Union, the capabilities to use the agreement that are created between financier and project partner are of great importance to create a balanced openness. D.

Creating Openness Within a Certain Community

An exciting development, primarily within the biotechnological field, is inspired by the so called ‘open source’ model.11 This model means, when one patents (or claims research in other ways) in an international context, platforms can be created for access and open use by a broader ‘community’.12 In some cases, it can even be interesting to create open source platforms as research projects in themselves. An open source platform is often based on the idea that research, development and commercial utilization shall be open within a specific group. This, however, does not mean that participation or utilization is free; on the contrary, it is often connected to a common governance function and many controls. In this context, capabilities are needed in order to collect claims together, to make research results available without transaction costs becoming too high, and to create structures for open use. The capability to create an IT-based platform for governance, collaboration and documentation of research is decisive. To build a platform, the capabilities of using databases, software, IPRs and contracts are of great importance. In order to ensure access (not only to the original patent), a common construction is to make sure that if additional research is carried out by the licensee on the subject matter and if dependent patents or other intellectual property are created, these are licensed on the same conditions as the original patent. It is possible to regulate through the licence agreement. Hence, not only the original patent is licensed for no fee and licensed to all comers (if those are the terms); all patents dependent on this patent are also licensed on those terms. Thus, access is achieved to a larger extent. However, this requires that the ownership of IPRs that exists prior to the licence agreement is clearly defined, and also that there are rules about research results generated under the scope of the agreement. These are often referred to as ‘background’ and ‘foreground’ intellectual property. The question of which rules should apply for the research results that are created under the term of the agreement but are not related to the scope of the agreement (often referred to as the ‘sideground’) also arises.


167

It is important to regulate who owns background, foreground and sideground, and, furthermore, to regulate who gets access and on what conditions. It could be argued that if the patent is only licensed to not-for-profit actors, the risk that one of those would block others is perhaps not so high. However, such a structure has many advantages. It could allow for anyone who is either a licensee or a licensor to obtain a patent on a patentable invention (if the inventor was not interested in obtaining a patent) and thereafter to license that patent to all comers, but on different conditions, thus ensuring access also to further research results. It would even be possible to differentiate between different actors: for example, all comers would get a non-exclusive licence but commercial actors would have to pay for it. This income could be used to maintain the structure, but also for further research. E. Intellectual Property-Based Structural Arenas for Openness (Open Arenas) Where industry as well as academia increasingly focus on intellectual assets and property, open knowledge channels can still be created. One of the most fundamental challenges is to create arenas where one has the capabilities to offer varieties of openness (open arenas). The challenge is to have such a structural order and yet retain individual capabilities so that one can offer the degree of openness that a balanced ownership responsibility requires. In some cases, one must have the capabilities to create openness among project participants. In other cases, it can be a question of establishing research projects similar to open source. In the international context, the concept of a ‘valley of death’ is often used – a valley between academia and industry where commercial utilization is eliminated. The positions of individual researchers and teachers are too weak to overcome this valley. This also applies to incubators on a regular basis, and only universities have the potentially sufficient structural strength to bridge this valley. This presumes that we shoulder the responsibility for building up portfolios of intellectual assets and that we connect pre-incubators to those portfolios. Academic projects need to be pre-incubated before they enter into a more commercial phase. Often, we need capabilities to manage these projects during a longer time period before the individual project can be capitalized as a company and spun off. We are dependent on capabilities to manage the totality of claiming and managing intellectual assets. Commercial capabilities for managing networks of transactions of intellectual property are important in order to get access to the intellectual assets of others on reasonable terms. With the help of our capabilities to develop crosslicences within strategic alliances, our university hospitals, for example, can get access to research tools and diagnostic tools and thereby make the creation of new innovations possible. In this way we can build platforms.

168


If we have the capabilities to develop platforms for utilization that are based on transactions of intellectual property, then we also have the capabilities to take on development assignments and utilization commissions from industry actors. A comprehensive challenge for universities that intend to create a strong position concerning utilization is, together with industry actors, to establish platforms for development and innovation. Capabilities to ‘preincubate’ projects are something on which industry as well increasingly depends. More and more of those involved within activities close to research emphasize the importance of so called ‘open innovation’, where development is shared between companies and between companies and academic actors (Chesbrough 2003; von Hippel 2005). In business areas where one is dependent on network transactions of intellectual property, it is strategically important to be able to open up both R&D and utilization processes. For smaller companies, it is often strategically decisive to be able to link up to these open platforms.

NOTES 1.

David (1993, p. 21) makes the following remark about economic studies of intellectual property: ‘They should take more pains not to allow familiar, simplifying abstractions to obscure a central fact about the nature of the world for which they would prescribe institutional reforms, namely that the complex body of law, judicial interpretation, and administrative practice that one has to grapple with in this field was not created by some rational, consistent, social welfare-maximizing public agency. What one is faced with, instead, is a mixture of the intended and unintended consequences of an undirected historical process on which the varied interests of many parties, acting at different points (some widely separated in time) ... So, it would be really quite remarkable if the evolution of legal institutions concerning patents, copyrights, and trade secrets had somehow resulted in a set of instruments optimally designed to serve either public policy purposes or the private economic interests of individuals and firms seeking such protections.’ 2. Structural transformation – the concept traces back to Marx’s theory of the economy as a base and a superstructure, and how structures are based upon ideologies. Schumpeter, Hayek and others build upon this fundamental understanding. Schumpeter (1934) developed an understanding of creative destruction in structural transformation. Hayek (1945) developed an understanding of markets as selective information processes that generate spontaneous order. More recently, Nelson and Winter (1982) provided the bridge for a more operative understanding of how cognitive processes – in their words ‘organizational routines’ and ‘genetic material’ rooted in a specific historical and economic reality – are essential in shaping and constraining choices. 3. It is important to acknowledge that, if we are to understand the inherent complexity of intellectual property, we must question our understanding of law as well. Focusing on development, we need a constructionist approach where we recognize that neither law nor intellectual property exists in itself, and that law, as well as intellectual property, is nothing else than the beliefs which are communicated between different actors/governance structures. We need to get rid of the veil of naïve legal formalism and to foster a perspective that not only accepts that intellectual properties are social constructions, but also allows us to handle the construction process with an open mind. Legal realism, with critical analysis of intellectual property and rights as metaphysics, is an important path to increase such awareness. The Danish legal realist Alf Ross explained how law is of a magical nature (Ross 1934,


4.

5.

6. 7 8. 9. 10. 11. 12.

169

p. 19). The Swedish legal realist Karl Olivecrona described legal phenomena as fantasybeliefs (Olivecrona 1942, p. 5). Latour is one of the important contributors to an understanding of the innovation and innovation projects as cognitive construction processes. Latour explains how research results, techniques, innovations, projects, and so on are not autonomous objects and are not part of an existing ontology. Research and innovation projects, even in natural science, are social construction processes. They are cognitive processes that sometimes give themselves a context and sometimes do not. According to Latour, we can have the word ‘context’ replaced by the more friendly word ‘network’ (Latour 1996, p. 135). Latour claims that technological projects are ‘deployed in a variable-ontology world’ (p. 173). This world is a result of the inter-definition of partakers and other actors (p. 173). The project becomes a contextual and network-based translation and negotiation process. Spender, who elaborates on a cognitive approach in management and discusses the implications of Latour’s theory, concludes: ‘it is difficult to imagine how we might operationalize it or even whether Latour would consider this important or useful’ (Spender 1996, p. 57). This is also a common conclusion among recognized scholars in sociology and philosophy, such as Luhmann (1985), who are well aware of socially constructed reality. Luhmann explains how societal and systematic reality on the one hand is the result of cognitive openness and, on the other, is the result of normative closure. The implication of Luhmann’s reasoning is that the experienced existence of institutions, systems and structures – that is, society in itself – is the result of normative closure. If we want systems, structures, institutions and so on to exist, we must have normative closure. This is because they do not exist in themselves and thus cannot be described. The implicit conclusion is that without normative closure, we will have no institutional order resulting in a structural collapse in business as well as in society at large. Spender (1996), who reaches the same conclusion as Luhmann states: ‘Closure mechanisms are those aspects of the firm’s or industry’s internal processes which generate its autonomy and self-regulation facility, define its boundaries and interactions with others and help attenuate consideration of its endless externalities.’ However, cognitive complexity has to be conceptualized and described as objects, persons, relations, autonomous disciplines, roles, and so on, and that is part of our task. In the European Union, there is an increasing discussion of whether intellectual property rights can be used in the promotion of open innovation. ‘To the question: “Why should PROs protect intellectual property”, appropriate answers are: “To encourage the economic applications of their discoveries for the benefit of the public” and “To make the research function more attractive and better supported”. To the question: “How should this property be protected”, it is appropriate that the PRO takes steps itself, provided that these steps reflect good understanding of where and how the benefits can accrue’ (European Commission 2005, p. 3). The licence agreement has been downloaded from http://www.cancertechnology.co.uk/ downloads/European_BRCA2_Licence.pdf. The draft licence agreement has been downloaded from http://www.bios.net/daisy/ PELicense/751/1169.html. These provisions have been downloaded from http://www.pipra.org/docs/ HumResLanguagePIPRA.doc. For example, concerning pharmaceuticals, the prerequisites for developing ‘access-friendly intellectual property policies’ are increasingly discussed. See inter alia Kapczynski (2003). Many are sceptical about the development where more and more research related processes are governed as transactions of intellectual property (see for example David 2000; Nelson 2001; Nelson 2003; Mowery and Sampat 2005). For a more comprehensive description of the development see Hope (2008). To get a legal overview of how the ‘open source’ thought is used within science, see also Boettiger and Burk (2004, pp. 221–31). The Public Patent Foundation project ‘Tropical Diseases Initiative’, the HapMap Project and the Biological Innovation for Open Society inter alia constitute interesting examples.

170


REFERENCES Alänge, S., U. Petrusson, S. Scheinberg and R. Jiménez, (2006), ‘Intellectual Properties: Alternative Strategies to Value Creation in Life Sciences’, http://www.globelicsindia2006.org/. Berger, P. and T. Luckmann ([1966] 1991), The Social Construction of Reality. A Treatise in the Sociology of Knowledge, London: Penguin Books. Boettiger, S. and D.L. Burk (2004), ‘Open Source Patenting’, Journal of International Business Law, 01(I), 221–31. Chesbrough, H. (2003), Open Innovation – The New Imperative for Creating and Profiting from Technology, Boston, Mass: Harvard Business School Press. David, P. (1993), ‘Intellectual Property Institutions and the Panda’s Thumb: Patents, Copyrights, and Trade Secrets in Economic Theory and History’, in M. Wallerstein, M. Mogee and R. Schoen (eds), Global Dimensions of Intellectual Property Rights in Science and Technology, Washington, DC: National Academy Press. David, P.A. (2000), ‘The Digital Technology Boomerang: New Intellectual Property Rights Threaten Global “Open Science”’, http://www-econ.stanford.edu/faculty/ workp/swp00016.pdf. Eisenberg, R. (1996), ‘Patenting Research Tools and the Law’, in National Research Council (1997), Intellectual Property Rights and Research Tools in Molecular Biology: Summary of a Workshop Held at the National Academy of Sciences, February 15–16, 1996 Washington,DC: National Academy Press; go to: http://books.nap.edu/books/0309057485/html/index.html. European Commission (2005), Responsible Partnering: Joining Forces in a World of Open Innovation, A Guide to Better Practices for Collaborative Research and Knowledge Transfer between Science and Industry, Brussels: European Commission. Glavå, M. and U. Petrusson (2002), ‘Illusionen om rätten – Juristprofessionen och ansvaret för rättskonstruktionerna’, in B. Askeland and J. Bernt (eds), Erkjennelse og engasjement: minnesseminar for David Roland Doublet [1954–2000], Bergen: Fagbokforlaget. Hayek, F.A. (1945), ‘The Use of Knowledge in Society’, American Economic Review, Vol. XXXV(4), pp. 519–30. Hope, J. (2008), Biobazaar: The Open Source Revolution and Biotechnology, Cambridge, Mass: Harvard University Press. Kapczynski, A. (2003), ‘Access to Essential Medicines and University Research: Building Best Practices’, Yale University Law School, Yale AIDS Network, Law Policy and Ethics Core, Yale University and Center for Interdisciplinary Research on AIDS (CIRA), http://www.cira.med.yale.edu. Latour, B. (1996), Aramis or the Love of Technology, Cambridge, Mass: Harvard University Press. Luhmann, N. (1985), ‘The Self-Reproduction of Law and its Limits’, in G. Teubner (ed.), Dilemmas of Law in the Welfare State, Berlin: Walter de Gruyter. Lundqvist, M. and U. Petrusson (2002), ‘Designing the Role of the Entrepreneur – Using a Norm Constructionist Approach at the Interface of Research, Learning and Innovation’, paper presented at the Triple-Helix conference, Copenhagen, 6–9 November (available from author). Mowery, D. and B. Sampat (2005), ‘The Bayh–Dole Act of 1980 and University– Industry Technology Transfer: A Model for Other OECD Governments?’, Journal of Technology Transfer, 30, 115–27.


171

Nelson, R. (2001), ‘Observations on the Post-Bayh–Dole Rise of Patenting at American Universities’, Journal of Technology Transfer, 26, 13–19. Nelson, R. (2003), ‘The Market Economy, and the Scientific Commons’, Laboratory of Economics and Management (LEM), Sant’Anna School of Advanced Studies, Pisa, Italy, LEM Papers Series, http://www.lem.sssup.it/WPLem/files/2003-24.pdf. Nelson, R. and S. Winter (1982), An Evolutionary Theory of Economic Change, Cambridge, Mass: Belknap Press of Harvard University. Olivecrona, K. (1942), Lagens imperativ, Lund, Sweden: C.W.K. Gleerup. Pamp, C. and U. Petrusson (2004), ‘Claiming Intellectual Property in Research – Towards an Entrepreneurial University’, working paper presented at CIP Symposium on the Entrepreneurial University, Göteborg, Sweden, 1–3 June. Petrusson, U. (1999), Patent och industriell omvandling, Stockholm: Norstedts Juridik. Petrusson, U. (2001), ‘Structural Transformation Towards License-based R&DStructures’, in European Commission Workshop Report on IPR aspects of Internet Collaborations, Brussels: European Commission, Fifth Framework Programme, pp. 36–9. Petrusson, U. (2003), ‘Patents as Structural Capital – Towards Legal Constructionism’, in O. Granstrand (ed.), Economics, Law and Intellectual Property, Dordrecht: Kluwer Academic Publishers, pp. 363–94. Petrusson, U. (2004), Intellectual Property and Entrepreneurship, Göteborg, Sweden: Centre for Intellectual Property Studies (CIP). Petrusson, U. (2007), Universitetet i kunskapsekonomin, Stockholm: VINNOVA. Ross, A. (1934), Virkelighed og Gyldighed i Retslæren – En kritik af den teoretiske retsvidenskabs grundbegreber, dissertation, København University. Schollin, K. (2008), Digital Rights Management: The New Copyright, Dissertation, Göteborg University. Schumpeter, J. (1934), The Theory of Economic Development, Cambridge, Mass: Harvard University Press. Searle, J. (1995), The Construction of Social Reality, New York: Penguin Books. Shapiro, Carl (2000), ‘Navigating the Patent Thicket: Cross Licenses, Patent Pools and Standard-Setting’, Competition Policy Center, Paper CPC00-011, 1 May, http://repositories.cdlib.org/iber/cpc/CPC00-011. Spender, J.-C. (1996), ‘Making Knowledge the Basis of a Dynamic Theory of the Firm’, Strategic Management Journal, 17 (Winter Special Issue), 45–62. Von Hippel, E. (2005), Democratizing Innovation, Cambridge, Mass: MIT Press.

9. Wikipedia, collective authorship and the politics of knowledge Matthew Rimmer Because the world is radically new, the ideal encyclopedia should be radical, too... It should stop being safe – in politics, in philosophy, in science. Charles van Doren (1962, p. 26)

I.

INTRODUCTION

A ‘wiki’ is a piece of software that allows users to easily create, edit and hyperlink web pages together, and create collaborative and community websites. Ward Cunningham (2002) was the developer of the first wiki in 1994 – the WikiWikiWeb, which he described as ‘the simplest online database that could possibly work.’ He used the Hawaiian phrase, ‘Wiki’, a shorthand for fast, to describe the software. Cunningham (2002) explained, ‘I chose wiki-wiki as an alliterative substitute for “quick” and thereby avoided naming this stuff quickweb.’ He elaborated: Wiki is a piece of server software that allows users to freely create and edit Web page content using any Web browser. Wiki supports hyperlinks and has a simple text syntax for creating new pages and crosslinks between internal pages on the fly. Wiki is unusual among group communication mechanisms in that it allows the organization of contributions to be edited in addition to the content itself. Like many simple concepts, ‘open editing’ has some profound and subtle effects on Wiki usage. Allowing everyday users to create and edit any page in a Web site is exciting in that it encourages democratic use of the Web and promotes content composition by nontechnical users. (Cunningham, 2002)

Since the early 2000s, wikis have been adopted both on private Intranet and the Internet. There have been a number of famous sites that have relied upon wiki software. Most notably, Wikipedia has encouraged amateurs and experts alike to develop a large, multilingual online encyclopedia. The project, the Encyclopedia of Life, has sought to develop an online, global database cataloguing the world’s biodiversity in all its rich complexity. 172

Wikipedia and collective authorship

173

Global economy and culture are also giving rise to powerful new models of production based on community, collaboration and self-organisation. Tapscott and Williams contend that ‘this new economic model extends beyond software, music, publishing, pharmaceuticals, and other bellwethers to virtually every part of the global economy’ (Tapscott and Williams 2006, p. 2). The authors argue that capitalists must ‘embrace a new art and science of collaboration we call wikinomics’ (2006, p. 3). The co-founder of Wikipedia, Larry Sanger, observes that the online encyclopedia has generated a great deal of public debate, noting: ‘Wikipedia inspires unreasonable passions’ (Sanger 2006). He observes: ‘The vigor with which it is attacked, often unreasonably and based on false assumptions, and defended, even against perfectly reasonable criticisms, gives a clue that something interesting is going on’ (Sanger 2006). He notes that Wikipedia has been defended by a range of Internet partisans including ‘high school and college students tired of being talked down to by their teachers, open source devotees, cranks resentful of professionals, privacy and free speech advocates, old-fashioned anarchists, and (indeed) epistemic collectivists who give often undue weight to the so-called wisdom of crowds’. Sanger notes that ‘on the other side, Wikipedia’s most vocal detractors include professional editors, journalists, college professors, librarians, some (not most) bloggers, and victims of Wikipedia’s sometimes defamatory biographies’ (Sanger 2006). He suggests that the underlying debate is about the politics of knowledge: ‘The politics of knowledge, as I will speak of it, concerns this procedural question writ large: how should we, as a society, decide what should pass for knowledge?’ (Sanger 2006). In his book, The Future of the Internet – And How to Stop It, Jonathan Zittrain is a champion of Wikipedia. He notes: ‘A constitutional lawyer might review these tales of Wikipedia and see a mess of process that leads to a mess of substance: anonymous and ever-shifting users; a God-king who may or may not be able to act unilaterally; a set of rules now large enough to be confusing and ambiguous but small enough to fail to reach most challenges’ (Zittrain 2008, p. 141). Zittrain, though, believes that there are virtues in the open evolution of Wikipedia: Wikipedia’s success, such as it is, is attributable to a messy combination of constantly updated technical tools and social conventions that elicit and reflect personal commitments from a critical mass of editors to engage in argument and debate about topics they care about. Together these tools and conventions facilitate a notion of ‘netizenship’: belonging to an Internet project that includes other people, rather than relating to the Internet as a deterministic information location and transmission tool or as a cash-and-carry service offered by a separate vendor responsible for its content. (Zittrain 2008, p. 142)

174


The academic believes that Wikipedia is an exemplar of open, generative technology: ‘Wikipedia shows us that the naïveté of the Internet’s engineers in building generative network technology can be justified not just at the technical layer of the Internet, but at the content layer as well’ (Zittrain 2008, p. 144). In his polemic, The Cult of the Amateur, Andrew Keen inveighs against Wikipedia, disparaging it as ‘an online encyclopedia where anyone with opposable thumbs and a fifth-grade education can publish anything on any topic from AC/DC to Zoroastrianism’ (Keen 2007, p. 4). He submits: ‘By empowering the amateur, we are undermining the authority of the experts who contribute to a traditional resource like the Encyclopaedia Britannica’ (p. 44). Keen contends: ‘In undermining the expert, the ubiquity of free, user generated content threatens the very core of our professional institutions’ (p. 44). He concludes: ‘Wikipedia ... is almost single-handedly killing the traditional information business’ (p. 131). Expressing similar concerns, Oliver Kamm complains of Wikipedia: ‘It combines the free-market dogmatism of the libertarian Right with the anti-intellectualism of the populist Left’ (Kamm 2007). In his piece, ‘Digital Maoism’, Jaron Lanier writes acerbically about Wikipedia and other services, which valorise collectivism (Lanier 2006). He argues that Wikipedia is part of a broader movement on the Internet that aims to promote the collective view above individual judgment (Lanier 2006). The author of The Cathedral and the Bazaar, Eric Raymond, argues that ‘“disaster” is not too strong a word for Wikipedia’ (Schiff 2006). In his view, the site is ‘infested with moonbats’ – a term of abuse referring to dogmatists of any ideological persuasion. Raymond concludes: ‘The more you look at what some of the Wikipedia contributors have done, the better Britannica looks’ (Schiff 2006). This chapter considers the legal ramifications of Wikipedia, and other online media, such as the Encyclopedia of Life. Nathaniel Tkacz (2007) has observed: ‘Wikipedia is an ideal entry-point from which to approach the shifting character of knowledge in contemporary society.’ He observes: ‘Scholarship on Wikipedia from computer science, history, philosophy, pedagogy and media studies has moved beyond speculation regarding its considerable potential, to the task of interpreting – and potentially intervening in – the significance of Wikipedia’s impact’ (Tkacz 2007). After an introduction, Part II considers the evolution and development of Wikipedia, and the legal troubles that have attended it. It also considers the establishment of rival online encyclopedia – such as Citizendium set up by Larry Sanger and Knol, the mysterious new project of Google. Part III explores the use of mass, collaborative authorship in the field of science. In particular, it looks at the development of the Encyclopedia of Life, which seeks to document the world’s biodiversity. This chapter expresses concern that Wiki-based software had to develop in


175

a largely hostile and inimical legal environment. It contends that copyright law and related fields of intellectual property need to be reformed in order better to accommodate users of copyright material (Rimmer 2007). This chapter makes a number of recommendations. First, there is a need to acknowledge and recognise forms of mass, collaborative production and consumption – not just individual authorship. Second, the view of a copyright ‘work’ and other subject matter as a complete and closed piece of cultural production also should be reconceptualised. Third, the defence of fair use should be expanded to accommodate a wide range of amateur, peer-to-peer production activities – not only in the United States but in other jurisdictions as well. Fourth, the safe harbour protections accorded to Internet intermediaries, such as Wikipedia, should be strengthened. Fifth, there should be a defence in respect of the use of ‘orphan works’ – especially in cases of large-scale digitisation. Sixth, the innovations of open source licensing should be expressly incorporated and entrenched within the formal framework of copyright laws. Finally, courts should craft judicial remedies to take into account concerns about political censorship and freedom of speech.

II.

THE REPUBLIC OF LETTERS: WILKIPEDIA Wikipedia is a new paradigm in human discourse. It’s a place where anyone with a browser can go, pick a subject that interests them, and without even logging in, start an argument. In fact, Wikipedia is the largest and most comprehensive collection of arguments in human history, incorporating spats and vendettas on subjects ranging from Suleiman the Magnificent to Dan the Automator. As an unexpected side effect of being the perfect argument space, it’s also a pretty good place to find information about all the characters from Battlestar Galactica. (Sjoberg 2006)

There has long been controversy over copyright law and the creation of encyclopedia. In his erudite book, Encyclopaedic Visions, Richard Yeo considers the place of encyclopedia in the debate over literary property (Yeo 2001, pp. 195–221). He notes that there was a tension between the competing objectives of protecting private rights of authors and respecting the public right to information. Yeo observes that the compilers and owners of encyclopedia were opportunistic. They sought to exploit the arguments of both sides. They sought to assert authorship on the grounds of concise abridgement, clear definition of terms, the explication of theories and the overall organisation of the work. However, in the face of claims of plagiarism, they argued that the collections of such knowledge served the public interest in information.

176


In contemporary times, the debate over the copyright status of encyclopedia has intensified. There has been a great contest between traditional texts such as Encyclopaedia Britannica, new proprietary models such as Microsoft’s Encarta and open access online encyclopedia such as Wikipedia. Jonathan Zittrain has observed of this development: ‘As the Internet exploded, the slow-to-change walled garden content of formal encyclopedia was bypassed by a generative proliferation of topical Web pages, and search engines that could pinpoint them’ (Zittrain 2008, p. 131). There has been much competition between professional authors and amateurs, as well as antagonism between traditional publishers and open access advocates. A.

Wikipedia

In 2000, a Chicago futures trader and Internet entrepreneur, Jimmy Wales, hired Larry Sanger to help him create an online general-interest encyclopedia called Nupedia. The idea was to solicit articles from scholars, subject the articles to a seven-step review process and post them free online. Wales laments: ‘After 18 months and $250 000, we had 12 articles’ (Pink 2005). Wales and Sanger decided to boost the appeal of Nupedia through using wiki-based software. In January 2001, Wales and Sanger founded Wikipedia, a multilingual Web-based free-content encyclopedia (see Goetz 2003; Ciffolilli 2003; Benkler 2006, pp. 70–74; Younge 2006). The encyclopedia exists as a wiki, allowing volunteer Internet users to add, edit and delete entries. Wales called it ‘an effort to create and distribute a multilingual free encyclopedia of the highest possible quality to every single person on the planet in their own language’ (Wales 2005a). On 31 December 2005, Wales observed in his ‘Personal Appeal’: Wikipedia is based on a very radical idea, the realization of the dreams most of us have always had for what the Internet can and should become. Thousands of people, all over the world, from all cultures, working together in harmony to freely share clear, factual, unbiased information … a simple and pure desire to make the world a better place. … This is a radical strike at the heart of an increasingly shallow, proprietary and anti-intellectual culture. It is a radical strike at the assumption that the Internet has to be a place of hostile debate and flame wars. It is an appeal to the best within all of us. (Wales 2005b)

As of March 2008, there are 2 266 463 articles in the English language version of Wikipedia. There are more than 75 000 active contributors working on some 9 million articles in more than 250 languages. Chris Anderson observes that Wikipedia has a ‘long tail’ because it includes entries that no other encyclopedia can because of the constraints of paper or DVD limitations: ‘While many critics focus on the worst entries, the really important thing


177

about Wikipedia’s Tail is that there is nothing else like it anywhere’ (Anderson 2006, pp. 72–3). In his magnum opus, The Wealth of Networks, Yochai Benkler rhapsodises about the dramatic rise and growth of Wikipedia: ‘As the project has grown, it has developed more elaborate spaces for discussing governance and for conflict resolution’ (Benkler 2006, p. 73). He is intrigued by the community norms-based dedication to objective writing: ‘The project relies instead on social norms to secure the dedication of project participants to objective writing’ (p. 72). In 2003, Wales established the Wikimedia Foundation Inc. as the parent organisation to run Wikipedia. The Foundation has sought to build upon the success of the online encyclopedia and develop affiliated projects. Wikiquote is a repository of quotations taken from famous people, books, speeches, films or any intellectually interesting materials. Wikibooks aims to build a collection of free e-book resources, including textbooks, language courses, manuals and annotated public domain books. Wikisource is a multilingual project, started in November 2003, to archive a collection of texts that could be distributed as free and open content. Wikispecies is an open, wiki-based project to provide a central, more extensive species database for scientific taxonomy. The Wikinews project was launched in December 2004 with the mission to report the news on a wide variety of subjects. Wikimedia Commons was launched in September 2004 to provide a central repository for free photographs, diagrams, maps, videos, animations, music, sounds, spoken texts and other free media. All of the projects of the Wikimedia Foundation are collaboratively developed by its users using the MediaWiki software. The Wikimedia Foundation Inc. has registered a trade mark in respect of Wikipedia in order to brand this family of related projects (Wikimedia Foundation 2004). Wales has also been experimenting with commercial ventures. In 2004, he co-founded, with Angela Beesley, the for-profit company, Wikia Inc. Wikia supports the creation and development of wiki communities – it currently supports over 5500 communities in more than 70 languages. Wikia Inc. is also developing an open source web search engine entitled ‘Wikia Search’. Wales hopes that the service will challenge the hegemony of Google, noting: ‘I trust Google reasonably well, but that’s like saying you have a favorite politician’ (Deutschman 2007). According to Wales, ‘It is meant to take on Google by creating a search engine where all the editorial decisions are made by the general public and all the software is open’ (Lewine 2007, p. 30). B.

Authorship and Ownership

Copyright law is founded upon romantic assumptions about creative authorship – its underlying ideals are that authorship is individual, professional and

178


personal. Wikipedia and its relatives pose a number of challenges to the assumptions of copyright law about romantic, individual authorship. First, such open source ventures are based upon a notion of mass collaboration and collective authorship, rather than individual authorship. Rishab Aiyer Ghosh has commented: Humans are social creatures, and our greatest achievements have been collaborative efforts, often vast ones – especially in the realm of knowledge and the mind. That most of us assume creativity as necessarily individual, private and subject to the creative inputs of others only under commercial conditions, is a symptom of the conversion from knowledge and art – whether closely guarded secrets or widely published – to ‘intellectual property’. (Ghosh 2005, p. 1)

As a critic observes, copyright law should acknowledge ‘the new realities of an age where traditional understandings of authorship are inapplicable and where creativity is increasingly an ongoing, collaborative exercise with no true fixed end-point’ (Reuveni 2007, p. 343). Second, Wikipedia and its affiliates rely upon the efforts of amateurs, volunteers and novices – rather than professionals and experts. Dan Hunter and his colleague, Greg Lastowka (Hunter and Lastowka 2004), have used the phrase ‘amateur-to-amateur’ to describe the social phenomenon of popular information creation and free distribution. Third, Wikipedia and its companions rely upon anonymous and pseudonymous contributors – rather than necessarily personally identified authors. Such a model challenges the emphasis of copyright law upon the moral rights of individual authors to attribution and integrity (Adeney 2000). Wikipedia requires all contributions to be licensed under the GNU Free Documentation License, meaning that their content may be freely used, freely edited, freely copied and freely redistributed subject to the restrictions of that licence. Wikinews contributions are licensed under Creative Commons Attribution 2.5. Accordingly, the copyright notice for Wikipedia reinforces this message: The license Wikipedia uses grants free access to our content in the same sense as free software is licensed freely. This principle is known as copyleft. That is to say, Wikipedia content can be copied, modified, and redistributed so long as the new version grants the same freedoms to others and acknowledges the authors of the Wikipedia article used (a direct link back to the article satisfies our author credit requirement). (Wikipedia, Copyrights)

There have been a number of cases in which the courts have recognised that GNU Public Licences could be valid and enforceable.1 There have been a number of disputes about authors wishing to publish both in traditional scientific journals and new Web 2.0 services, such as Wikipedia. Wikipedia poses a particular threat to the norms and hierarchies of


179

learned societies, scientific journals and encyclopedia (Giles 2005; Encyclopaedia Britannica 2005). In 2008, a group of physicists – Jonathan Oppenheim, Robert W. Spekkens and Andreas Winter – have clashed with a scientific publisher, the American Physical Society, which would not allow them to post parts of their work to the online encyclopedia, blogs and other forums. The American Physical Society – http://www.aps.org/about/ – has the mission statement: ‘In the firm belief that an understanding of the nature of the physical universe will be of benefit to all humanity, the Society shall have as its objective the advancement and diffusion of the knowledge of physics.’ According to an article in New Scientist, a group of physicists are protesting the decision of the American Physical Society to withdraw an offer to publish two studies in its journal Physical Review Letters because the authors asked for a rights agreement that was compatible with their uploading to Wikipedia (Editorial 2008). The American Physical Society, however, requires authors to transfer their copyright to the Society before they can publish in an American Physical Society journal. This prevents the scientists from contributing their work to a website such as Wikipedia. In response, Gene Sprouse, editor-in-chief, observed that the American Physical Society would review its copyright policy: ‘A group of excellent scientists has asked us to consider revising our copyright, and we take them seriously’ (Editorial 2008). One of the complainants, Jonathan Oppenheim, a physicist of the University of Cambridge, has explained the nature of the dispute on his website (Oppenheim 2008). He discussed the impulse behind his desire to contribute to Wikipedia, as well as the Physical Review Letters: Technological advances have led to a vast array of tools that scientists can use to communicate their ideas. These tools include open access journals, online archives, paper rating websites, science blogs, quantum blogs, open courses, free universities, and open encyclopedias such as Wikipedia. … Not only does this enable scientists to better share ideas with each other, it creates a commons of scientific information that is freely available to the public. (Oppenheim 2008)

Oppenheim observed that he and his collaborators ‘wanted the option to contribute parts of our paper to the intellectual commons’. He commented that ‘the current transfer of copyright was drafted before the rise of Wikipedia and is simply no longer suited to current realities’. Oppenheim argued that authors should grant the American Physical Society a licence, rather than an assignment of copyright. He noted: ‘I believe that if we want to maximise the rights of both parties to innovate as much as possible, then the most natural thing to do would be for the author to give a highly permissive licence to [the Physical Review Letters] to make use of their work’ (Oppenheim, 2008). Oppenheim suggests that ‘the increased exposure is more

180


likely to increase the revenue of traditional journals’. As a result of the dispute, Oppenheim and his collaborators published their pieces on arXiv, an e-print service run by Cornell University for the fields of physics, mathematics, and computer science (Oppenheim and Winter 2008; Oppenheim, Spekkens and Winter 2008). C.

Plagiarism, Fair Use and Safe Harbours

In addition to grappling with questions about the authorship and ownership of copyright work, Wikipedia has had to contend with larger questions about dealing with the risks associated with copyright infringement. Jonathan Zittrain notes: ‘To be sure, while outside regulation is not courted, Wikipedia’s policy on copyright infringement exhibits a desire to integrate with the law rather than reject it’ (Zittrain 2008, p. 143). The academic notes: ‘Indeed, its copyright policy is much stricter than the laws of major jurisdictions require’ (p. 143). In an effort to ward off potential legal problems, Wikipedia has hired Mike Godwin as its general counsel (see Cohen 2007). The lawyer has gained significant experience working with the Electronic Frontier Foundation and Public Knowledge. Cohen observes: ‘His task is to defend an online encyclopedia created by tens of thousands of (often anonymous) contributors who comment freely on living people and businesses, armed with decades of scholarship, no knowledge at all, or something in between’ (Cohen 2007, p. 3). Godwin has commented: In another 25 years, all of our children will have grown up in a world in which media like these are mutable and changeable and people prank each other, and it will seem less important. Part of my job is to prevent restrictive rules from being put in place that prevent people from participating in massively democratic participatory media. And then let the new norms settle. (Cohen 2007, p. 3)

Godwin hoped to improve the advice that Wikipedia provided to contributors about copyright law. He said: ‘You shouldn’t have to be a lawyer to protect yourself from being sued’ (Cohen 2007, p. 3). Godwin has sought to dissuade Wikipedia users from resorting to litigation. A commentator, Ken Myers, observes: ‘Wikipedia’s best strategy to avoid liability may simply be to avoid plaintiffs’ (Myers 2006, p. 3). Wikipedia has sought to take advantage of the protection provided by the broad and flexible defence of fair use in the United States.2 The developers note: ‘In general, the educational and transformative nature of Wikipedia articles provides an excellent fair use case for anyone reproducing an article’ (Wikipedia website, Copyright FAQ). In its copyright policy, the online encyclopedia observes:


181

Wikipedia articles may also include quotations, images, or other media under the U.S. Copyright law ‘fair use’ doctrine in accordance with our guidelines for nonfree content. It is preferred that these be obtained under the most free content license practical (such as the General Free Documentation License or public domain). In cases where no such images/sounds are currently available, then fair use may be used in certain circumstances as described in the criteria for using non-free media. (Wikipedia, Copyrights)

Wikipedia’s policy for non-free content seeks to ‘support Wikipedia’s mission to produce perpetually free content for unlimited distribution, modification and application by all users in all media’; ‘minimize legal exposure by limiting the amount of non-free content, using more narrowly defined criteria than apply under United States fair use law’; and ‘facilitate the judicious use of non-free content to support the development of a quality encyclopedia’ (Wikipedia, Non-Free Content). Wikipedia also has a fair use project designed ‘to aid the English Wikipedia by improving and monitoring its invocation of the “fair use” clause of U.S. copyright law towards the use of copyrighted media, with the goal of preventing unnecessary and inappropriate copyright infringement, which could potentially cause all sorts of legal trouble for our beloved encyclopedia’ (Wikipedia, Fair Use Project). The operators of the online encyclopedia are no doubt conscious of the ramifications of the Supreme Court of the United States’ decision in the case of Metro-Goldwyn-Mayer Studios Inc. v Grokster Ltd (2005), which established a new inducement doctrine of liability for copyright infringement: ‘We hold that one who distributes a device with the object of promoting its use to infringe copyright, as shown by clear expression or other affirmative steps taken to foster infringement, is liable for the resulting acts of infringements by third parties.’ Accordingly, Wikipedia actively discourages copyright infringement: it instructs participants ‘never [to] use materials that infringe the copyrights of others’ because ‘this could create legal liabilities and seriously hurt the project’ (Wikipedia, Non-Free Content). Wikipedia seeks to take shelter under the safe harbours defence available for intermediaries under the Digital Millennium Copyright Act 1998 (US). In its copyright notice, the Wikimedia Foundation notes that it will respond expeditiously to complaints about copyright infringement of Wikipedia: ‘If you are the owner of content that is being used on Wikipedia without your permission, then you may request the page be immediately removed from Wikipedia’ (Wikipedia, Non-Free Content). First, the Foundation notes to copyright owners: ‘Wikipedia is a wiki, so you can remove the content yourself!’ (Wikipedia, Non-Free Content). Second, it notes that the Foundation will seek to assist the copyright owner, or their representative, after receipt of an informal request (Wikipedia, Non-Free Content). Third, the Foundation observes that its designated agent will respond to the slower process of a formal request

182


under the take-down notice scheme of the Digital Millennium Copyright Act. Ken Myers observes: ‘There is a reason why no one has sued Wikipedia yet: it is amazingly responsive to claims of injury’ (Myers 2006, p. 204). He adds: ‘Wikipedia’s institutional flexibility and lack of dogmatic adherence to free speech might be just enough to keep it on the right side of the cutting edge’ (p. 204). Daniel Brandt, the founder of the website Wikipedia Watch, has suggested that Wikipedia will find it difficult to operate in the current legal environment: ‘The social networking model, from Wikipedia, to Orkut in Brazil, to Napster, Grokster, and YouTube and copyright, is headed for a more restrictive legal environment’ (Brandt 2008). He tried to investigate plagiarism using a sample of about 1 per cent of Wikipedia’s 1.46 million English-language articles (Brandt 2006). He found plagiarism in 1 per cent of those articles. Brandt questioned whether Wikipedia can attract protection under the safe harbour protection as a ‘service provider’ provided by section 512 of the Digital Millennium Copyright Act: This formal structure of power is designed to enforce the many policies, formal and informal, about what constitutes appropriate content on Wikipedia. I believe that this structure in itself means that the Foundation is already much closer to a ‘publisher’ than a ‘service provider’. (Brandt 2006)

Accordingly, he contends that ‘the Foundation requires more due diligence to avoid copyright violations’ (Brandt 2006). He observes: ‘Administrators already make efforts to patrol copyright violations on images posted by users to illustrate articles, but no meaningful efforts have ever been made to detect plagiarism.’ Brandt contends: ‘If the Foundation is not protected by the special status of “service provider” as defined by the Digital Millennium Copyright Act 1998 (US), then a lack of prior due diligence will increase the Foundation’s liability for copyright violations.’ In his view, ‘the Foundation should launch a project to scan for plagiarism on all 1.46 million articles’. D. Imitators, Rivals and Competitors Wikipedia has spawned a host of imitators, emulators and rivals – including its Chinese imitator, Baidu Baike; Conservapedia, a wiki-based web encyclopedia project featuring a Conservative Christian viewpoint; fun projects like Wookieepedia, an online encyclopedia about the universe of Star Wars; and new ventures, such as Larry Sanger’s Citizendium and Google’s mysterious new knowledge project, Knol. In addition to being used to build online encyclopedia, Wiki-based software, such as WikiScanner and Wikileaks, has also had a significant impact upon the conduct and operation of political discourse. The Wikimedia Foundation has expressed concern about a Chinese search


183

engine, Baidu, copying and reproducing its articles on its own Chinese encyclopedia, Baidu Baike, without permission or acknowledgment. Florence Nibart-Devouard, the chair of the Board of Trustees at the Wikimedia Foundation, observed that Baidu failed to respect the terms of the GNU Free Documentation License: ‘They do not respect the license at all. That might be the biggest copyright violation we have. We have others’ (Nystedt 2007). Baidu, in a notice posted on its website, says that ‘it is Baidu’s policy to attach great importance to the protection of copyright and comply with all the applicable [Chinese] laws’ and that it will remove links to copyrighted works ‘in accordance with the applicable laws, regulations, and binding measures’ (Woo 2007). Baidu claims that all content generated on Baike is subject to its own copyright protection. Such an assertion of ownership over content derived from Wikipedia would amount to a breach of the GNU Free Documentation License, which requires new material to be shared alike on the same free terms as the original work. At this point, Wikipedia has no plans to use legal means to resolve its conflict with Baidu. The founder of Wikipedia, Wales, observes: ‘We only appeal to their moral judgment about what is right’ (Woo 2007). NibartDevouard comments that, if need be, the affected authors and editors involved in Wikipedia could organise a class action: ‘The Foundation does not hold a copyright on the articles, the editors or the authors do, so there is very little we can do’ (Nystedt 2007). This conflict has echoes of the litigation being waged in United States courts by the Association of American Publishers and the Authors’ Guild against the search engine, Google, over its appropriation and aggregation of copyright works on the service, Google Book Search (Rimmer 2007, pp. 225–60). As a result of his concerns about the quality and credibility of Wikipedia, Larry Sanger established a new online encyclopedia project ‘Citizendium’ in September 2006. He lamented: ‘While Wikipedia is still quite useful and an amazing phenomenon, I have come to the view that it is also broken beyond repair’ (Thomson 2007). Sanger noted: ‘Wikipedia has gone from a nearly perfect anarchy to an anarchy with gang rule’ (Schiff 2006). The new encyclopedia is designed to harness the participatory capacity of Wikipedia with the strengths of more traditional expert-based resources. Citizendium – ‘a citizen’s compendium of everything’ – aspires to establish a ‘republic of letters’: ‘We welcome experts as well as the general public; we will be built not by topdown orders but as and where contributors wish to work; and we will be organized as a genuine republic of letters governed by a rule of law’ (Citizendium, FAQ). Such a position has certain affinity with recent legal theory. The elegant German legal theorist, Gunther Teubner, has charted a shift in focus from the political constitution of the nation-state to the many civil constitutions of

184


world society (Teubner 2004). As an example, he raises the instance of the autonomous digital law-making of the Internet Corporation for Assigning Names and Numbers (ICANN). Such an analysis would be of equal force if applied to the self-regulation at work in online encyclopedia, such as Citizendium and its rivals. The information technology expert, Roger Clarke, doubts whether the model of Citizendium will be able to compete with the popularity of Wikipedia (Clarke 2006). He observes: ‘Ultimately, the community will vote with its feet, or consumers will determine what the market wants by paying with their clicks and eyeballs (choose your preferred metaphor)’ (Clarke 2006). He suggests: ‘Perhaps the venture’s greatest contribution will be to help us learn about quality assurance of open content.’ In December 2007, the search engine, Google, announced that it, too, would seek to challenge the dominance of Wikipedia with a new project called Knol (BBC News 2007). Udi Manber, the vice president of engineering at Google, commented: ‘A knol on a particular topic is meant to be the first thing someone who searches for this topic for the first time will want to read’ (Manber 2007). He elaborated: ‘The goal is for knols to cover all topics, from scientific concepts, to medical information, from geographical and historical, to entertainment, from product information, to how-to-fix-it instructions.’ It is suspected that Knol represents an attempt by Google to displace Wikipedia entries from high rankings on its search engine.

III. THE BOOK OF ALL SPECIES: THE ENCYCLOPEDIA OF LIFE Imagine the Book of All Species: a single volume made up of one-page descriptions of every species known to science. On one page is the bluefooted booby. On another, the Douglas fir. Another, the oyster mushroom. If you owned the Book of All Species, you would need quite a bookshelf to hold it. Just to cover the 1.8 million known species, the book would have to be more than 300 feet long. And you’d have to be ready to expand the bookshelf strikingly, because scientists estimate there are 10 times more species waiting to be discovered. It sounds surreal, and yet scientists are writing the Book of All Species. Or to be more precise, they are building a Web site called the Encyclopedia of Life. (Zimmer 2008 p. 1)

In Wikinomics: How Mass Collaboration Changes Everything, Don Tapscott and Anthony Williams discuss the adoption of peer-to-peer production strategies in various fields of science (Tapscott and Williams 2006, pp. 151–82). The authors contend: ‘Just as collaborative tools and applications are reshaping enterprises, the new Web will forever change the way scientists publish,


185

manage data, and collaborate across institutional boundaries’ (p. 157). They elaborate: A new age of collaborative science is emerging that will accelerate scientific discovery and learning. The emergence of open-access publishing and new Web services will place infinite reams of knowledge in the hands of individuals and help weave globally distributed communities of peers. (p. 152)

There have been a number of examples of such strategies. In the area of scientific publishing, the Public Library of Science has sought to rely upon Creative Commons licences to facilitate the publication of sharing of scientific articles across a range of fields – including in respect of medicine, genetics, biology, pathology, computational biology and tropical diseases (Rimmer 2007, pp. 283–7). In the field of biology, members of the public human genome project have adopted open source and peer-to-peer strategies in software such as Ensembl to share and annotate genetic information (Rimmer 2003, 2005, 2008; Hope 2008). A good example of such innovation would be the Ensembl Project. The Merck Gene Index and the SNP Consortium sought to put genomics information in the public domain (Moody 2004). In the area of drug discovery, there has been some experimentation with open source strategies. There have been some striking innovations in respect of the use of wiki software in other scientific fields. Of note are specialised encyclopedia such as Quantiki, a Quantum Mechanics and Information Wiki at the University of Cambridge, Qwiki at Stanford University and the statistical mechanics wiki, SklogWiki. Perhaps one of the most ambitious projects has been the Encyclopedia of Life, a monumental initiative to develop a global database to record the world’s biodiversity. A.

The Encyclopedia of Life

The Rio Convention on Biological Diversity 1992 decries ‘the general lack of information and knowledge regarding biological diversity’ and speaks of the ‘urgent need to develop scientific, technical and institutional capacities to provide the basic understanding upon which to plan and implement appropriate measures’. Edward O. Wilson (1998, 2002) is an eminent biologist, researcher and naturalist ardently committed to the protection of biodiversity. In 2003, he envisaged the creation of a global database to document the world’s biodiversity: Comparative biology, crossing the digital divide, has entered a still largely unheralded revolution: the exploration and analysis of biodiversity at a vastly accelerated

186


pace. Imagine an electronic page for each species of organism on Earth, available everywhere by single access on command. The page contains the scientific name of the species, a pictorial or genomic presentation of the primary type specimen on which its name is based, and a summary of its diagnostic traits. The page opens out directly or by linkage with other databases such as ARKive, Ecoport, and GenBank. It comprises a summary of everything known about the species’ genome, proteome, geographic distribution, phylogenetic position, habitat, ecological relationships, and, not least, its perceived practical importance for humanity. The page is indefinitely expansible. Its contents are continuously peer-reviewed and updated with new information. All the pages together form an encyclopedia, whose content is the totality of comparative biology. (Wilson 2003, p. 77)

He suggested that the all-species encyclopedia would serve human welfare in more immediately practical ways: ‘The discovery of wild plant species adaptable for agriculture, new genes for enhancement of crop productivity, and new classes of pharmaceuticals can be speeded’ (Wilson 2003, p. 77). Moreover, he envisaged: ‘The outbreak of pathogens and harmful plant and animal invasives will be better anticipated and halted’ (Wilson 2003, p. 77). It was his hope that ‘creating the base of the all-species encyclopedia, will secure the rightful place in comparative biology within mainstream science’ (Wilson 2003, p. 80). Inspired by this proposal, the Encyclopedia of Life – http://www.eol.org – was established in May 2007 to create an online reference source and database for every one of the 1.8 million species that are named and known on this planet. The Encyclopedia of Life is a collaborative scientific effort led by the Field Museum of Natural History, Harvard University, Marine Biological Laboratory, Missouri Botanical Garden, The Smithsonian Institution, and Biodiversity Heritage Library, a consortium including the core institutions and also the American Museum of Natural History (New York), Natural History Museum (London), New York Botanical Garden and Royal Botanic Gardens (Kew). The Encyclopedia of Life is intended to ‘serve as a global biodiversity tool, providing scientists, policymakers, students, and citizens information they need to discover and protect the planet and encourage learning and conservation’ (Encyclopedia of Life 2007). The Encyclopedia of Life intends to establish a website for every species on the planet. Each site will have two components. There will be an edited, scientific account of the species, managed by a scientific editor who will manage the authoritative content. Ralph Gomory observes: For more than 250 years, scientists have catalogued life, and our traditional catalogues have become unwieldy. The Encyclopedia of Life will provide the citizens of the world a ‘macroscope’ of almost unimaginable power to find and create understanding of biodiversity across the globe. It will enable us to map and discover things so numerous or vast they overwhelm our normal vision. (Encyclopedia of Life 2007)


187

Jonathan F. Fanton, president of the John D. and Catherine T. MacArthur Foundation, commented: ‘Just as a microscope reveals and helps us better understand the small and particular, the Encyclopedia of Life will allow us to discern patterns previously unseen, illuminating relationships, identifying gaps in our knowledge, and suggesting opportunities for new avenues of inquiry’ (Encyclopedia of Life 2007). There will also be an open access, Wikipedia-like component, which will enable members of the public to contribute information. The Encyclopedia of Life acknowledges that it was inspired in part by the online encyclopedia, Wikipedia: Wikipedia accumulated about 1.5 million entries in English in its first four years. That gave us confidence that our tasks are manageable with current technology and social behaviour, although the expert community in a lot of the subjects for pages in Encyclopedia of Life may be only a handful of people. Wikipedia has also created some species pages, as have other groups. Encyclopedia of Life will, we hope, unite all such efforts and increase their value. The Wikimedia Foundation is a member of the Encyclopedia’s Institutional Council. (Encyclopedia of Life, FAQ)

The developers of the Encyclopedia of Life hope to emulate the success of Wikipedia: ‘Encyclopedia of Life has the potential to be a similar phenomenon, serving as a catalogue, database, and learning tool about every organism that has ever lived on the planet’ (Encyclopedia of Life, FAQ). Jonathan Fanton comments: ‘The Encyclopedia of Life will be a vital tool for scientists, researchers, and educators across the globe, providing easy access to the latest and best information on all known species’ (Encyclopedia of Life 2007). In February 2008, the Encyclopedia of Life released its first 30 000 web pages for an alpha test and feedback (Encyclopedia of Life 2008). The executive director, Jim Edwards, commented: The Encyclopedia of Life is a good example of the way the World Wide Web can be used innovatively to assemble diverse kinds of information in an easy-to-use, ever-growing compendium. It can accommodate almost any kind of information about species and, unlike a published book, can be updated instantly. (Encyclopedia of Life 2008)

Later in 2008, the public will be able to contribute text, videos, images and other information about a species. The Encyclopedia of Life will incorporate the best of this information into the authenticated pages. The authenticated pages will also include a diverse range of other materials, including peerreviewed articles and access to DNA barcodes. The Encyclopedia of Life hopes over time to expand the project from its base of English-language text to include a wide variety of other languages.

188

B.


Legal Dangers

The developers of the Encyclopedia of Life note: ‘A possible area of obstacles or dangers is intellectual property’ (Encyclopedia of Life, FAQ). They observe: ‘The Encyclopedia will be very generous with credit and recognition, and we will soon be posting a general statement of principle about open and accessible content, encouraging sharing, and so on’ (Encyclopedia of Life, FAQ). The Encyclopedia of Life will of course be free to publish material that has fallen into the public domain because copyright protection has expired. Graham Higley, the head of Library and Information Services at the National History Museum, comments that the project has sought permission from learned societies and publishers to print material still under copyright protection: The interesting thing is that we thought we would have to spend quite a lot of time and effort persuading people to let us have their journals right up to date. The reverse is happening. A wall of people is coming to us, saying, ‘Will you do our title?’ We’ve signed agreements with more than 20 different institutions [as at December 2007]. And we only had the agreement in document in place a month ago. A lot more are talking to us. (Hyams 2007, p. 33)

The agreements will be attached to the scanned copies on the Biodiversity Heritage Library Project. In many cases, the Encyclopedia of Life is relying upon Creative Commons licences – which provide for flexible uses of copyright material (see Rimmer 2007, pp. 261–95). In its terms of use, the Encyclopedia of Life explains that it relies upon Creative Commons licensing: ‘The Creative Commons has set up a robust, free, easy-to-use set of protocols by which content owners can easily mark their creative work with the freedoms or restrictions that they want the information to carry’ (Encyclopedia of Life, Terms of Use). The project observes that, in most cases, Encyclopedia of Life data partners have made content available for re-use under one of the following Creative Commons licences: • CC-BY (Attribution), which allows users to copy, transmit and reuse the information, and to remix or adapt the information, as long as attribution regarding the source of the information is maintained (see http://creativecommons.org/licenses/by/3.0/). • CC-BY-SA (Attribution-ShareAlike), which has the additional constraint on top of CC-BY that if you alter, transform or build upon this work, you may distribute the resulting work only under the same, similar or a compatible license (see http://creativecommons.org/licenses/bysa/3.0/).


189

• CC-BY-NC (Attribution-NonCommercial), which has the additional constraint on top of CC-BY that you may not use this work for commercial purposes (see http://creativecommons.org/licenses/by-nc/3.0/). • CC-BY-NC-SA (Attribution-NonCommercial-ShareAlike), which has the additional constraint on top of CC-BY that if you alter, transform or build upon this work, you may distribute the resulting work only under the same, similar or a compatible licence and that you may not use this work for commercial purposes (see http://creativecommons.org/ licenses/by-nc-sa/3.0/). The project emphasises: ‘If you wish to re-use any content for purposes other than those allowed by the associated Creative Commons license, you must get permission to do so from the copyright holder (Source)’ (Encyclopedia of Life, Terms of Use). The project notes: ‘Content available on this site has been provided by the Encyclopedia of Life’s data partners, who may have placed restrictions on how you can re-use that content’ (The Encyclopedia of Life, Terms of Use). The project also observes: ‘You are required to check to see which license applies to any portion(s) of the page you wish to re-use and to abide by any restrictions on that content’ (Encyclopedia of Life, Terms of Use). As a large-scale digitisation project of archival and historical materials, the Encyclopedia of Life faces particular problems in gaining access to ‘orphan’ works – works where the copyright owner is untraceable. At present, potential users of ‘orphan’ works are concerned that they may be found liable for statutory damages, amounting to as much as $150 000. Graham Higley noted: ‘We have thought hard about orphan works, and the changes to orphan works legislation, which we hope will be beneficial, but it’s not quite clear exactly how yet’ (Hyams 2007, p. 33). The United States Congress considered, but ultimately did not pass, the Shawn Bentley Orphan Works Act 2008 (US). The legislation proposed to limit liability for copyright infringement where a person ‘performed and documented a qualifying search, in good faith, for the owner of the infringed copyright’ and ‘was unable to locate the owner of the infringed copyright’. Arguably, though, such reforms would provide little solace to the Encyclopedia of Life. The reputation of the large-scale digitisation project would be impugned if it were found to have infringed copyright, even though the remedies available might be limited. It would be preferable for there to be a complete defence for use of ‘orphan’ works – much like there is for the ‘fair use’ of copyright works. C.

Collaborations

One of the participants in the Encyclopedia of Life is the Biodiversity Heritage

190


Library. This group involves an alliance of ten major natural history museum libraries, botanical libraries and research institutions (Biodiversity Heritage Library). The Biodiversity Heritage Library is developing a strategy and operational plan to engage in the digitisation of the published literature of biodiversity held in their respective collections. This literature will be available through a global ‘biodiversity commons’. The Biodiversity Heritage Library is seeking to allow access to this historical material under an open access model: ‘This material will be available for open access and responsible use as a part of a global Biodiversity Commons’ (Biodiversity Heritage Library). The prospectus for the venture notes: ‘In the spirit of open access and responsible use, the libraries will seek to negotiate with relevant publishers, especially those with clear missions to disseminate biodiversity information for the public good – learned societies, museums, botanical gardens and herbaria – for permission to digitize and provide access to publications still protected by copyright’ (Biodiversity Heritage Library). Furthermore, the document notes: ‘Contributing libraries may manage and repurpose as much of the content as they wish with the understanding that all currently public domain material will remain in the public domain and will be made available gratis in an open access mode’ (Biodiversity Heritage Library). Thus the material provided by the Biodiversity Heritage Library to the Encyclopedia of Life will not be encumbered by restrictions in relation to the reuse of that content. The Missouri Botanical Garden has a fine botanical library which was founded in 1859 by Henry Shaw. It is used in conjunction with the herbarium by Garden research staff, botany students and visiting scientists from around the world. The Missouri Botanical Garden has established a freely accessible, web-based digital collection of its 18th and 19th century botanical literature called ‘Botanicus’ (Botanicus). This project, funded by the W.M. Keck Foundation, includes illustrated and non-illustrated works of significant importance to taxonomic botany. The Missouri Botanical Garden has engaged in digitisation of its collection for a number of reasons: ‘Digitizing, indexing, and annotating historical scientific literature is vital to future research in systematic botany, the science of the identification of plants’ (Botanicus). Another participant, the New York Botanical Garden, has established a ‘Virtual Herbarium’. The C.V. Starr Virtual Herbarium is the electronic gateway to the collections of the William and Lynda Steere Herbarium. The Virtual Herbarium aims to make specimen data available electronically for use in biodiversity research projects; to reduce shipping of actual specimens for projects where digital representations will suffice for study; and to reunite data elements derived from a specimen with the catalogue record for that specimen. The New York Botanical Garden allows for use of images of plant specimens for scientific or educational purposes without charge (C.V. Starr Virtual


191

Herbarium, Conditions of Use ). However, it does demand permission for their use, and asks that the source of the image be acknowledged or cited. The ‘Virtual Herbarium’ currently comprises 850 000 herbarium specimens and 120 000 high-resolution specimen images. It is updated on a daily basis as the New York Botanical Garden pursues the goal of digitising all of its 7 million plant and fungi specimens. A United Kingdom participant, the Kew Royal Botanical Gardens, has been creating an electronic catalogue of its herbarium collection for both internal and external uses (Royal Kew Botanical Gardens, Herbarium Digitisation). The Kew Royal Botanical Gardens is also involved in a number of individual digitisation projects – including the African Plants Initiative, the Australian Virtual Herbarium, and Collections from Mount Jaya and West Cameroon (Royal Kew Botanical Gardens, Herbarium Catalogue). The National Museum of Natural History at the Smithsonian Institute is also involved in the Encyclopedia of Life. Interestingly, the Smithsonian Institute has been criticised in the past for its obstructive approach to access to public documents. The advocacy group, Public Resource.org, was highly critical of the Institute placing draconian limits upon access to historical photographs, even though ‘the overwhelming majority of the images in SmithsonianImages.SI.Edu appear to be public domain’ (Public Resource.org 2007). As a result, Public Resource.org downloaded the images and distributed them freely by various means, including on the website Flickr. The Encyclopedia of Life will play an important role in uniting and synthesising the various digitisation projects of individual museums, herbariums and botanical institutions. D.

Challenges

In addition to grappling with the various legal issues, the Encyclopedia of Life faces a number of other challenges. Edward O. Wilson anticipated: ‘Construction of the complete taxonomic base will not, however, be just a smooth compilation of species’ (Wilson 2003, p. 1). It notes: ‘The magnitude of biodiversity and the tangle of evolutionary processes that generated it still present formidable problems’ (p. 1). First, the Encyclopedia of Life will have to overcome the scientific problems involved in classifying and cataloguing such a wide diversity of biodiversity – including micro-organisms. Edward O. Wilson anticipates that the species concept will have to grapple with incongruities. Daniel Brooks, a University of Toronto biologist, reflects pessimistically that previous efforts to catalogue every species on the planet have failed: ‘I have seen 20 years of good ideas go nowhere’ (Zimmer 2008, p. 1). Second, there are a number of technological issues that will need to be

192


addressed in such a large bioinformatics project. Edward O. Wilson acknowledged that there was a need to address questions of inter-operability, quality control, funding and preservation of data: ‘In joining the bioinformatics nation, taxonomists and encyclopedists need to address and overcome the growing problem of information overload already bedevilling those managing DNA microarray analyses, airline schedules and bank accounts’ (Wilson 2003, p. 80). Third, a number of researchers wonder whether the Encyclopedia of Life will have sufficient resources to achieve its goal. The website received startup funds of $US10 million from the John D. and Catherine T. MacArthur Foundation and $US2.5 million from the Alfred P. Sloan Foundation, as well as funds from cornerstone institutes involved in the project. Rod Page, a taxonomist for the University of Glasgow, noted: ‘For electronic material in general, how you sustain it forever is a huge issue’ (Odling-Smee 2007). The project developers acknowledge that ‘the steady-state costs of maintaining the Encyclopedia of Life, including data maintenance and incorporating new information, will cost between $5 million and $10 million per year (that is, less than $6 per species per year)’ (Encyclopedia of Life, FAQ). The Encyclopedia of Life will need to avoid such pitfalls if it hopes to achieve its grand ambitions of enlightenment. Edward O. Wilson observes: ‘These obstacles are daunting, but they are of a technical nature eminently vulnerable to human ingenuity’ (Wilson 2003, p. 80). Hopefully, the Encyclopedia of Life will be able to help what the Rio Convention on Biological Diversity 1992 calls the ‘common concern of humankind’ – the ‘conservation of biological diversity’.

IV.

CONCLUSION

It must be recognised that such Wiki-based services as Wikipedia and the Encyclopedia of Life have been successful, in spite of the strictures of copyright law, confidential information and contract law. Such enterprises remain vulnerable in this legal environment. Intellectual property rights pose certain threats to wiki-based ventures. In an interview with Quarto, Lawrence Lessig discusses some of the challenges in copyright law faced by Wikipedia and its affiliated resources (Wikimedia Foundation 2005). He contemplates some of the long-term risks involved in the reuse of collective content: Enterprises like Wikipedia strain copyright law, certainly. It is hard to express the understandings of the people participating in the construction of this creative work, and nobody understands how the work will develop and change. One of the hardest features of copyright law, for projects like Wikipedia, is the international variability of copyright terms. You could exploit that to create great liability for Wikipedia


193

by finding content which is only free in some jurisdictions ... and then use other jurisdictions to shut it down. (Wikimedia Foundation 2005)

Lessig fears: ‘The greatest long-term risk would be a series of legal decisions that throw into doubt, the ability of the project to rely on the collective assignment [of] rights’ (Wikimedia Foundation 2005). He concludes: ‘The way you deal with that is to figure out the [Free Documentation Licence] structure that best facilitates that collective assignment of rights’ (Wikimedia Foundation 2005). There is a larger question of whether copyright law, and related fields, should be reformed in order to better accommodate wiki-based technologies, with the underlying norms of amateur, anonymous and collective authorship. Neil Weinstock Netanel comments in the conclusion to his book, Copyright’s Paradox, about the need for copyright law to recognise both professional and amateur content: Our system of free expression requires not just a diversity of content but a plurality of types of speech and speakers. It must embrace commercial mass media, cottage industry publishers, professional authors, publicly funded artists and media, nonprofit organizations and political activists, and a host of sundry creators and discussants who exchange their opinions, expression and personal reworkings of bits and pieces of popular culture without any expectation of monetary remuneration. (Netanel 2008, pp. 217–18)

The author laments that ‘today’s increasingly bloated set of Blackstonian proprietary rights too often serves as a tool for private censorship, a burden on non-market and semi-market speech, and a bottleneck for incumbent copyright industries to ward off new media and other potential challengers’ (Netanel 2008, p. 218). A number of doctrinal reforms to copyright law and related rights could help provide a more sympathetic and congenial environment for wiki-based technology in the fields of education, politics and science. First, the presumption of individual authorship under copyright law needs to be reformulated. There should be much greater recognition and acceptance given to forms of mass, collaborative authorship. Second, the definition of a copyright ‘work’ and other subject matter as a complete and closed piece of cultural production also should be reconceptualised. The advent of wiki software shows that a copyright ‘work’ or subject matter can be an open and incomplete form, subject to revision and supplementation. Third, the defence of fair use should be applied in a broad and flexible fashion – not only in the United States but in other jurisdictions as well. There should be scope for the release of unpublished, confidential documents both under copyright law and confidential information. Fourth, the safe harbour protections accorded to Internet intermediaries should be clearly and expressly extended to Web 2.0 services

194


(Myers 2006). Fifth, there should be a defence in respect of the use of ‘orphan works’ – where reasonable efforts have been made to find a copyright owner, without avail. Large-scale digitisation projects of historical and archival material such as the Encyclopedia of Life show the need for such protection. Sixth, the innovations of open source licensing should be expressly incorporated and entrenched into the formal framework of copyright laws. There should be recognition for the validity and legitimacy of the options of free software licences, open source licences and Creative Commons licences at a legislative level. At the same time, the scope of technological protection measures should be curtailed, so that they do not unduly interfere with open source licensing. Seventh, courts should exercise judicial discretion as to remedies to take into account concerns about freedom of speech. Finally, at an international level, a Treaty on Access to Knowledge (A2K) should be established to help assist global information networks – of which Wikipedia and the Encyclopedia of Life are exemplars. The Geneva Declaration on the Future of the World Intellectual Property Organisation paid tribute to the ‘hundreds of innovative collaborative efforts to create public goods, including the Internet, the World Wide Web, Wikipedia, the Creative Commons, GNU Linux and other free and open software projects, as well as distance education tools and medical research tools’ (Consumer Project on Technology 2005). The international intellectual property system should promote such endeavours of enlightenment and learning. As Jack Balkin (2008) has noted: ‘Opening up access to knowledge is a demand of global justice; it is both a human rights issue and a crucial factor in spurring economic development and technological innovation.’

NOTES 1.

Progress Software Corp. v MySQL AB, 195 (2002); see Moglen (2002); In re Harald Welte v SiteCom Germany; and Wallace v Free Software Foundation, Inc. (2005). 2. Section 107 of the Copyright Act 1976 (US) has been interpreted to extend to parody and transformative uses – as per Campbell v Acuff-Rose Music Inc. (1994); thumbnail images as per Kelly v Arriba Soft; and consumer uses such as time-shifting and space-shifting as per Metro-Goldwyn-Mayer Studios Inc. v Grokster Ltd (2005).

REFERENCES Adeney, E. (2000), ‘The Mask of the Author - Anonymity and Pseudonymity in Two Common Law Countries’, International Review of Industrial Property and Copyright Law (IIC), 31, 913-49. Anderson, Chris (2006), The Long Tail: How Endless Choice is Creating Unlimited Demand, London: Random House.


195

Balkin, J. (2008), ‘The Access to Knowledge Conference (A2K3)’, the Information Society Project, Yale Law School, http://www.law.yale.edu/news/6191.htm. BBC News (2007), ‘Google Debuts Knowledge Project’, BBC News, http://news.bbc.co.uk/go/pr/fr/-/2/hi/technology/7144970.stm, 15 December. Benkler, Yochai (2006), The Wealth of Networks: How Social Production Transforms Markets and Freedom, New Haven and London: Yale University Press. Brandt, D. (2006), ‘Plagiarism by Wikipedia Editors’, Wikipedia Watch, http://www.wikipedia-watch.org/psamples.html, 27 October. Brandt, D. (2008), ‘What is the Life Expectancy of Wikipedia?’, Wikipedia Review, http://wikipediareview.com/blog/page/2/, 10 January. Ciffolilli, A. (2003), ‘Phantom Authority, Self-Selective Recruitment and Retention of Members in Virtual Communities: The Case of Wikipedia’, First Monday, 8(12), http://www.firstmonday.org/issues/issue8_12/ciffolilli/index.html. Clarke, R. (2006), ‘Citizendium v Wikipedia: How to QA Open Content?’, http://www.anu.edu.au/people/Roger.Clarke/II/CvW.html, 2 November. Cohen, N. (2007), ‘Defending Wikipedia’s Impolite Site’, New York Times, 20 August, p. 3. Consumer Project on Technology (2005), ‘Geneva Declaration on the Future of the World Intellectual Property Organization’, http://www.cptech.org/ip/wipo/ futureofwipodeclaration.pdf, 4 March. Cunningham, W. (2002), ‘What is Wiki?’, http://www.wiki.org/wiki.cgi?WhatIsWiki, 27 June. Deutschman, A. (2007), ‘Why is this Man Smiling?’, Fast Company Magazine, http://www.fastcompany.com/magazine/114/features-why-is-this-mansmiling.html, March. Editorial (2008), ‘Writer’s Block’, New Scientist, 15 March. Encyclopaedia Britannica (2005), ‘Fatally Flawed: Refuting the Recent Study on Encyclopedic Accuracy by the Journal Nature’, http://corporate.britannica.com/ britannica_nature_response.pdf, March. Encyclopedia of Life (2008), ‘Scientists to Explore Life’s Mysteries Through Encyclopedic “Macroscope”’, Encyclopedia of Life, http://www.eol.org/faq, 26 February. Ghosh, Rishab Aiyer (2005), ‘Why Collaboration Is Important (Again)’, in Code: Collaborative Ownership and the Digital Economy, Cambridge, Mass. and London: The MIT Press (chapter 1, pp. 1–6). Giles, J. (2005), ‘Internet Encyclopaedias Go Head to Head’, Nature, 15 December, 900–901. Goetz, T. (2003), ‘Open Source Everywhere’, Wired Magazine, 11 (11), http://www.wired.com/wired/archive/11.11/opensource_pr.html. Hope, Janet (2008), Biobazaar: The Open Source Revolution and Biotechnology, Cambridge, Mass. and London: Harvard University Press. Hunter, D. and G. Lastowka (2004), ‘Amateur-to-Amateur’, William and Mary Law Review, 46, 951–1030. Hyams, E. (2007), ‘The Encyclopaedia of Life: A Website for Each Species’, Library and Information Update, 6(12), 31–3. Kamm, O. (2007), ‘Wisdom? More Like Dumbness of the Crowds’, The Times, http://www.timesonline.co.uk/tol/comment/columnists/guest_contributors/ article2267665.ece, 16 August. Keen, Andrew (2007), The Cult of the Amateur: How Today’s Internet is Killing Our Culture, New York: Doubleday.

196


Lanier, J. (2006), ‘Digital Maoism: The Hazards of New Online Collectivism’, Edge: The Third Culture, http://www.edge.org/3rd_culture/lanier06/lanier06_index.html, 30 May. Lewine, E. (2007), ‘The Encyclopedist’s Lair’, New York Times Magazine, 18 November, p. 30. Manber, U. (2007), ‘Encourage People to Contribute Knowledge’, http://googleblog.blogspot.com/2007/12/encouraging-people-to-contribute.html, 13 December. Moglen, E. (2002), ‘Declaration in Support of Defendant’s Motion for Preliminary Injunction on its Counterclaims in Progress Software Corp v MySQL AB’, http://www.gnu.org/press/mysql-affidavit.pdf. Moody, Glyn (2004), How Bioinformatics Is Revolutionizing Science, Medicine, and Business, Hoboken, NJ: John Wiley & Sons Inc. Myers, K. (2006), ‘Wikimmunity: Fitting the Communications Decency Act to Wikipedia’, Harvard Journal of Law and Technology, 20, 163–208. Netanel, Neil Weinstock (2008), Copyright’s Paradox, Oxford: Oxford University Press. Nystedt, D. (2007), ‘Baidu May Be Worst Wikipedia Copyright Violator’, IDG News Service, PC World, 6 August. Odling-Smee, L. (2007), ‘Encyclopedia of Life Launched: Online Effort Aims to Bring Together Biodiversity Knowledge’, Nature, 9 May. http://www.nature.com/news/ 2007/070508/full/news070508-7.html. Oppenheim, J. (2008), ‘Traditional Journals and Copyright Transfer’, http://www.damtp.cam.ac.uk/user/jono/item/toc.html. Oppenheim, J. and A. Winter (2008), ‘Uncommon Information (The Cost of Exchanging a Quantum State)’, arXiv, http://arxiv.org/abs/quant-ph/0511082, 5 March. Oppenheim, J., R. Spekkens and A. Winter (2008), ‘A Classical Analogue of Negative Information’, arXiv, http://arxiv.org/abs/quant-ph/0511247, 25 March. Pink, D. (2005), ‘The Book Stops Here’, Wired Magazine, 13(3), http://www.wired. com/wired/archive/13.03/wiki.html. Public Resource.org (2007), ‘SmithsonianImages.SI.Edu’, http://public.resource. org/memo.2007.05.19.html, 19 May. Reuveni, E. (2007), ‘Authorship in the Age of the Conducer’, Journal of the Copyright Society of the USA, 54(218), 1801–59. Rimmer, M. (2003), ‘Beyond Blue Gene: Intellectual Property and Bioinformatics’, International Review of Industrial Property and Copyright Law, 34(1), 31–49. Rimmer, M. (2005), ‘Japonica Rice: Intellectual Property, Scientific Publishing, and Data-Sharing’, Prometheus, 23(3), 325–47. Rimmer, Matthew (2007), Digital Copyright and the Consumer Revolution: Hands Off My iPod, Cheltenham, UK and Northampton, MA, USA: Edward Elgar. Rimmer, Matthew (2008), Intellectual Property and Biotechnology: Biological Inventions, Cheltenham, UK and Northampton, MA, USA: Edward Elgar. Sanger, L. (2006), ‘The New Politics of Knowledge’, Digital Universe Foundation, http://www.dufoundation.org/blog/?p=96, 31 July. Schiff, S. (2006), ‘Know It All: Can Wikipedia Conquer Expertise?’, New Yorker, 31 July. Sjoberg, L. (2006), ‘Wired FAQK’, Wired Commentary, http://www.wired.com/ software/webservices/commentary/alttext/2006/04/70670, 19 April. Tapscott, D. (2007), ‘Wiki Politics’, http://www.wikinomics.com/blog/ index.php/2007/01/26/wiki-politics/, 26 January.


197

Tapscott, Don and Anthony Williams (2006), Wikinomics: How Mass Collaboration Changes Everything, New York: Portfolio, Penguin Books. Teubner, Gunther (2004), ‘Societal Constitutionalism: Alternatives to State-Centred Constitutional Theory?’, in Christian Joerges, Inger-Johanne Sand and Gunter Teubner (eds), Transnational Governance and Constitutionalism, Oxford: Hart Publishing. Thomson, I. (2007), ‘Wikipedia Broken Beyond Repair Says Co Founder’, Information World Review, 13 April. Tkacz, N. (2007), ‘Old Discourse, New Object: Wikipedia’, The Value of Knowledge, University of Melbourne, http://www.inter disciplinary.net/ati/education/ tvk/tvk1/Tkacz%20paper.pdf, 12–14 February. van Doren, C. (1962), ‘The Idea of an Encyclopedia’, American Behavioral Scientist, 6(1), 23–6. Wales, J. (2005a), ‘Wikipedia is an Encyclopedia’, [email protected], 8 March. Wales, J. (2005b), ‘Personal Appeal’, http://wikimediafoundation.org/wiki/ Personal_Appeal, 31 December. Wikimedia Foundation (2004), ‘“Wikipedia” in Relation to “Providing Information in the Field of General Encyclopedic Knowledge via the Internet”’, United States Patent and Trade Mark Office, Trade Mark Registration No. 3040722, Filing Date: 14 September. Wikimedia Foundation (2005), ‘Interview with Lawrence Lessig’, Wikimedia Quarto, 2, http://wikimediafoundation.org/wiki/Interview/Larry_Lessig. Wilson, Edward O. (1998), Consilience: The Unity of Knowledge, New York: Knopf. Wilson, Edward O. (2002), The Future of Life, London: A Little, Brown Book. Wilson, E.O. (2003), ‘The Encyclopedia of Life’, Trends in Ecology and Evolution, 18, 77–80. Woo, E. (2007), ‘Baidu’s Censored Answer to Wikipedia; The Chinese Search Engine’s Baike Online Encyclopedia Blocks Politically Sensitive Entries; Some Say it Condones Plagiarism and Copyright Abuse’, Business Week Online, 14 November. Yeo, Richard (2001), ‘Copyright and Public Knowledge’, Encyclopaedic Visions: Scientific Dictionaries and Enlightenment Culture, Cambridge: Cambridge University Press, Ch. 8, pp. 195–221. Younge, G. (2006), ‘The Addictive Encyclopaedia’, Guardian Weekly, 1–7 September. Zimmer, C. (2008), ‘The Encyclopedia of Life, No Bookshelf Required’, New York Times, 26 February, p. 1. Zittrain, Jonathan (2008), The Future of the Internet – And How to Stop It, New Haven and London: Yale University Press. Zola, Emile (1898), ‘J’Accuse’, L’Aurore, English translation at http://www.marxists.org/archive/zola/1898/jaccuse.htm, 13 January.

CASE LAW Board of Education v Pico, 457 U.S. 853, 867 (1982). Campbell v Acuff-Rose Music Inc., 510 US 569 (1994). Central Hudson Gas & Electric Corp. v Public Service Commission of New York, 447 U.S. 557, 564 (1980). HRH The Prince of Wales v Associated Newspapers Ltd [2006] EWHC 522 (Ch).

198


In re Charlotte Observer, 921 F.2d 47, 50 (4th Cir. 1990). In re Harald Welte v SiteCom Germany, No. 21 0 6123⁄04 District Court of Munich I (19 May 2004), http://www.jbb.de/judgment_dc_munich_gpl.pdf#search= %22netfilter%20sitecom%22. Kelly v Arriba Soft, 280 F 3rd 934 (2002); 336 F.3d 811 (2003). Kleindienst v Mandel, 408 U.S. 753, 762 (1972). McGraw-Hill Companies et al. v Google Inc., Case No. 05 CV 8881-JES (2005). Metro-Goldwyn-Mayer Studios Inc. v Grokster Ltd, 125 S. Ct 2764 (2005). Progress Software Corp. v MySQL AB, 195 F.Supp.2d 328 (D.Mass., 2002). The Authors’ Guild v Google Inc, Case No. 05 CV 8136-JES (2005). Wallace v Free Software Foundation, Inc., Slip Copy, 2005 WL 3239208 (S.D.Ind., 2005).

ONLINE RESOURSES American Physical Society, http://www.aps.org/about/. Biodiversity Heritage Library, http://www.biodiversitylibrary.org/About.aspx. Botanicus, http://www.botanicus.org/. C.V. Starr Virtual Herbarium, http://sciweb.nybg.org/Science2/VirtualHerbarium.asp. Citizendium, http://en.citizendium.org/wiki/Main_Page. Conservapedia, http://www.conservapedia.com/Main_Page. Creative Commons, http://creativecommons.org/. Encyclopedia of Life, http://www.eol.org/faq. Ensembl, http://www.ensembl.org/index.html. E.O. Wilson Biodiversity Foundation, http://www.eowilson.org/. Knol, http://knol.google.com/k. Public Library of Science, http://www.plos.org/. Quantiki, http://www.quantiki.org/. Qwiki, http://qwiki.stanford.edu/wiki/Main_Page. Royal Kew Botanical Gardens, Herbarium Catalogue, http://apps.kew.org/herbcat/ gotoProjects.do. Royal Kew Botanical Gardens, Herbarium Digitisation, http://www.kew.org/data/ herb_digitisation.html. Sklog Wiki, http://www.sklogwiki.org/SklogWiki/index.php/Main_Page. Wikia, http://www.wikia.com/wiki/Wikia. Wikia Search, http://alpha.search.wikia.com/. Wikibooks, http://en.wikibooks.org/wiki/Main_Page. Wikileaks, www.wikileaks.org. Wikimedia Commons, http://commons.wikimedia.org/wiki/Main_Page. Wikinews, http://en.wikinews.org/wiki/Main_Page. Wikipedia, http://en.wikipedia.org/wiki/Main_Page. Wikiquote, http://en.wikiquote.org/wiki/Main_Page. WikiScanner, http://wikiscanner.virgil.gr/. Wikisource, http://wikisource.org/wiki/Main_Page. Wikispecies, http://species.wikimedia.org/wiki/Main_Page. Wikitruth, http://wikitruth.info/index.php?title=Wikitruth:About. Wookieepedia, http://starwars.wikia.com/wiki/Main_Page.

10. Copyright and the new street literature Megan Richardson and Jason Bosland I.

INTRODUCTION

Street literature has ‘always had more influence than polite literature’, according to Leslie Shepard (Shepard 1973, p. 35),1 an influence he attributes to the individuality of its opinions, distinct literary style and its links with the ‘fierce energy of the crowds’ (p. 14). Historically, courts have been remarkably restrained in their use of copyright law to control impolite, street, literature, notwithstanding efforts by established magazines and newspapers including The Times to encourage them to do so in a series of cases in the late 19th century. Indeed, in these cases the seeds of modern ‘fair dealing’ defences to infringement were sown with the result that a rich discourse of diverse opinions was fostered by the law. We suggest that an historical review of the cases is now timely because of a proliferation of a new street literature – the blogs, newsgroups and other ephemeral sites of information and debate about news and current events that can be found on the internet – and a worryingly successful attempt to use copyright law to control its more impolite aspects in the case of Los Angeles Times v Free Republic.2 Our argument is not for statutory law reform but rather for a revival of the historically sympathetic treatment by courts of ‘nontransformative’ uses of copyright material where such uses are for free speech ends.

II.

THE OLD STREET LITERATURE

The old street literature came in many forms. In the centuries following the invention of the printing press, ballads, broadsides, pamphlets and chap-books dispersed news, opinions and stories through the highways and byways of England and its colonies. In England especially, to use Shepard’s words, ‘the streets were the libraries and bookshops of the masses’ (Shepard 1973, p. 35). The uncensored character of the material, coupled with its common language, 199

200


blending of fact, idea and fiction, and the fact that it was cheap, even free, guaranteed an audience, even notwithstanding official disapproval and sporadic attempts to exert legal control. In the mid to late 19th century, cheap books, magazines and penny newspapers took over functions of earlier ephemera. Newspapers, in particular, absorbed material previously communicated in widely varying ways into ‘one cheaply-produced and easily distributed general-purpose sheet’ (Williams 1961, p. 192). Their mix of concentration and cheapness set a course under which, as Raymond Williams says, ‘a wide range of interests was being brought into a literate form’, the catalyst for this ‘the cheapest and most extensive print’ (p. 192). Later concentration under centralised control would become the dominant form of newspaper organisation. But at this stage, the ‘penny’ newspapers were still street literature, being widely accessible, very cheap and highly independent. For these reasons, they make an ideal focus for our street literature study. A.

Copyright’s Marginal Effect

As Table 10.1 shows, the sheer number of newspapers made them difficult to manage. In London, for instance, an estimated 658 newspapers operated in 1885.3 In Melbourne, with a population, still under 500 000, there were 25 newspapers,4 but these were only about a sixth of those in Victoria whose newspaper literature was said to be, in relation to its population, unsurpassed in the world’ (see Morrison 2005, p. 26).5 Nineteenth century newspapers represented a diversity of political views, usually more radical than not. There was also not the will to be controlled, as shown by their independent political positions and opposition to established forces. Even their writing style was of a rather undisciplined character – with headlines in the more extreme cases ranging from ‘Disgraceful Proceedings’ (Morrison 2005, p. 58),6 ‘Separation or No Separation – Aye That is the Question’ (p. 58),7 ‘Daring Conspiracy and Attempted Violation’ (see Williams 1961, p. 191)8 and ‘The Proletarian’s Career from Cradle to Grave’ (the last in the popular Reynold’s Weekly Newspaper).9 If sedition and libel laws exhibited minimal control over style and content, copyright law exhibited even less control over widespread copying practices by newspapers which did not owe their existence to copyright law and certainly did not respect its constraints. In the 1820s and 30s copying was rife (in the same way as, earlier on, stories appeared multiple times in different chapbooks). In particular, the ‘2d papers’ copied, Catherine Seville says (Seville 1999, p. 130). But others did also – except for The Times. And the practices continued throughout the century.

Table 10.1

English and Victorian newspapers by region – 19th century

London

201

English provincial

Melbourne and provincial Victoria

Year

Conservative

Liberal independent

Neutral/

Other

Total

1858 1874 1885 1858 1874 1885 1858 1874 1882

17 25 30 80 294 275 – – –

39 41 48 175 489 374 – – –

64 332 572 139 67 615 – – –

– 12 8 – 66 21 – – –

120 410 658 394 916 1285 351 1112 1633

Source: 1. Drawn from Morrison (2005, p. 111). 2. Morrison (2005, p. 187). 3. Drawn from the Victoria Press Manual and Advertisers’ Handbook (1882).

202

B.


The Times Sues

In the late 19th century, the proprietors of The Times brought a string of actions for copyright infringement in which they challenged copying practices. The fact that these met with mixed success shows something of the problems of controlling the spread of news and opinion under the aegis of copyright. In the first case, Walter v Howe,10 a Howe and Peddle’s pamphlet version of ‘The Life and Works of Benjamin Disraeli’ was admittedly a copy of a version published in The Times. It was also accepted that a newspaper could qualify as a ‘book’ in the broad language of the Literary Copyright Act 1842.11 The infringement claim failed, however, because it was not clear that The Times owned the copyright and the newspaper was not registered. By Walter v Steinkopff,12 the newspaper was duly registered and the ownership question had been resolved. This time infringement was claimed against the St James’ Gazette after extracts from articles originally published in The Times of 13 April 1892 were carried in its evening edition, including two-fifths of a letter on America written by Rudyard Kipling. The claim succeeded, it being accepted that this was not simply a case of trying to claim copyright protection over the news itself – rather the ‘form of the language or modes of expression by which information is conveyed’ warranted copyright protection (Walter v Steinkopff, at 500). However, the limited remedy awarded gave it little real significance. The Gazette’s editor had argued in its defence that the practice of journalistic copying was ‘a well-recognised general custom’ in the newspaper trade, allowing derogation of legal rights (at 491–2). North J. accepted that copying had been ‘notorious for many years’ but rejected the suggestion that The Times had been a party to it (a ‘ludicrous’ suggestion) (at 499–500).13 However, when it came to the remedy, the injunction was maintained only with respect to Kipling’s article. As to the others, North J. said, ‘[t]heir interest has passed away … and it has not been shewn that any damage resulted to The Times from the illegal appropriation of these articles’ (at 500). C.

Rise of Fair Use

If in Walter v Steinkopff a claimed practice of journalistic copying was no answer to copyright infringement, other practices were – at least provided they could fit under the aegis of an exempted ‘fair use’. For instance, North J. observed that had there been ‘reasonable’ copying for purposes of criticism this would be fair use (at 494–5). As to why, North J. referred to Lord Eldon L.C.’s comment in Wilkins v Aikin14 that: ‘the question upon the whole is, whether this is a legitimate use of the Plaintiff’s publication in the fair exercise of a mental operation, deserving the character of an original work’ (at 425–6, emphasis added). Lord Eldon L.C. added there is no breach of the legal

Copyright and the new street literature

203

right even if the effect of the use may be ‘prejudicial’ for the plaintiff (at 425–6). Other judges were more sceptical of the possibility of harm being suffered by a plaintiff as a result of a defendant’s reasonable copying for criticism or review, in particular harm in any market in which the plaintiff was operating. For instance, in Bell v Whitehead,15 where the defendant copied nearly one quarter of an article on atmospheric pressure (a subject of topical interest) from the Monthly Chronicle in the Railway Times for the purpose of a sustained critique, Lord Cottenham said: It is right that, where such remarks and observations as those in question are made, the parties reading them should see the extract of the work which gives rise thereto … [L]arge extracts, moreover, from work are consistently given in the ‘Edinburgh’ and ‘Quarterly’ Reviews, not for the purpose of rendering it unnecessary for the public to purchase the works reviewed by those periodicals, but of explaining their criticisms on the works. The insertion of such extracts, moreover, tends to extend the sale of the works reviewed. The case before me is also very distinguishable from the case of a work being given in a cheaper form, by a stranger, than the original work costs the proprietor. If I were to entertain this application of the plaintiffs, how could I decline to entertain a similar application in the case of those newspapers from the columns of which articles have been extracted and printed in other newspapers, for the purpose of questioning and criticising the opinions expressed therein? (at 142).16

These two approaches continued to feature in British cases where fair use was argued as an exception to copyright infringements by newspapers, magazines, pamphlets and other ephemera. Uses that involved criticism or review were justified as fair either on the basis of mental labour involved in producing something of the character of an original work in its own right17 or on the basis of limited market effect, coupled with a legitimate critical or review purpose in making use of material.18 The separate rationales were also mirrored in separate statutory defences for ‘criticism, review’ and ‘newspaper summary’ (later news reporting) in post-19th century Anglo-Australian Copyright Acts beginning with s. 2(1)(i) of the 1911 Act.19 In the first case a transformative, critical use was required. But in the latter the emphasis was more on the limited effect of the particular use in news reporting on the market for the original work, the eighth edition of Copinger and Skone James on the Law of Copyright (Skone James 1948) observing that ‘in considering whether a dealing with a particular work was fair [for the purpose of news reporting], it would have to be considered whether any competition was likely to exist between the two works’ (p. 137). D.

Fair Use and Free Speech

We might wonder why 19th century courts adopted such a liberal approach to

204


the standard of fair use in copyright infringement. In practical terms the approach is understandable. Given the large number and diversity of newspapers and periodicals devoted in various ways to feeding the public demand for news and opinions about matters of high public interest and concern (as well as more lowly matters), courts may have been reluctant to interfere in practices that were well established and generally popular. But there were arguably intellectual reasons as well. A liberal approach to newspaper-sponsored debates fitted with prevailing ideas about the importance of freedom of speech and discussion in the Victorian age, the ‘Liberty Century’ as A.C. Grayling calls it (Grayling 2007, ch. 6). Free speech and debate had already been seen as implicated by the mid-century copyright reforms, and strong opposition led to some curtailment of the original proposal for extending the term of copyright.20 In the parliamentary session and surrounding debates, extended copyright was characterised as ‘injurious to … the liberty of the press’ (‘Observations’ 1838, p. 2) and more broadly to the spread of literature and knowledge (Seville 1999, p. 47).21 The radical press took up the call and even The Times published editorials, articles and letters attacking the proposal as a severe restraint on ‘the progress of knowledge’.22 It is logical that the same concerns would influence the judicial treatment of ‘fair use’. Importantly, while some philosophical justifications for freedom of speech may have been especially concerned with improvements in knowledge,23 a concern later captured in Holmes J.’s references to ideas competing in ‘the market’ as the ‘best test of truth’,24 others valued the market-place of ideas as a rich and diverse market, in which multiple viewpoints, expressive framings and types of publication flourished alongside each other.25 The emphasis found here on different patterns of understanding and experience and the possibilities these create for communal enrichment and growth – a kind of organic creativity that copyright law does not usually acknowledge – indeed fitted nicely with emerging ideas from Mill and other liberal philosophers about benefits of representative democracy as entailing that ‘not merely a few individuals in succession, but the whole public, are made, to a certain extent, participants in the government and sharers in the instruction and mental exercise derived from it’ (Mill 1865, ch. 6). Such ways of thinking may well have encouraged 19th century Anglo-Australian courts to avoid over-privileging an author’s claim of special contribution and value over prospective other valuable uses of their material. Thus, adoption of a low threshold of ‘authorship’ in the 1900 case of Walter v Lane,26 where Lord Halsbury L.C. went as far as to suggest the most important service provided by a journalist-author may simply be the recording and reporting of something which had not been available to the audience before (at 549), can be seen as culminating a period of serious reflection about the process of making and passing on knowledge, rather than an entirely new beginning.


205

E. End of the Golden Era But the golden era of street literature had almost died by the end of the 19th century. By the early to mid 20th century expansive news agencies had gone a long way towards reducing the cost of gathering and providing news in standard formats for newspapers; and newspapers themselves had become fewer in number but also more organised, substantial and profitable than their 19th century counterparts, so unauthorised copying was less fundamental to their news-reporting operations (Boyce, Curran and Wingate 1978). Courts, for their part, began to apply copyright infringement standards more stringently, restricting the amount of work that could be considered ‘insubstantial’ and construing the statutory exceptions in a progressively narrower fashion (and the exceptions themselves also became slightly curtailed in later enactments by increasingly detailed provisions designed to narrow their compass in specific respects).27 In particular, judges became increasingly ‘aware’ of the market effects on the original work, or at least they were more willing to find such an effect, especially where the newspapers operated in the same locale and the copying was reasonably substantial, as typically was the case with a newspaper summary whose purpose was to give the gist of the original.28 Indeed, by the middle of the 20th century it was commonly accepted that the simple fact that ‘a newspaper summary of an article’ appeared in a ‘rival paper’ was enough to conclude that it ‘would probably be restrained’ (Skone James 1948, p. 137). The defence, not surprisingly, fell largely into disuse as a newspaper defence, except in certain high profile cases where a valuable exclusive was at stake.29 And in the limited other contexts where it was raised, as for instance in the Australian The Panel case, involving a dispute between rival television channels over a fair dealing exception for rebroadcasts of copyright-protected news and current events clips for purposes of humorous commentary by members of The Panel,30 the defence rarely succeeded where the alternative criticism or review defence was not available, and was seen as difficult to distinguish from mere entertainment or otherwise unprivileged use.31

III.

THE NEW STREET LITERATURE

One of the impacts of the internet has been an outburst of new street literature, especially in the United States (although not limited by national borders). As John Quiggin and Dan Hunter (2008, p. 216) say, ‘the expansion of the internet has been one of the most spectacular technological and cultural developments humanity has ever seen’ and this is especially in the development of literary forms based around free sharing of information:

206


Before the rise of the web the most interesting internet content was found on the UseNet systems of newsgroups, covering almost every imaginable topic from atheism to zoology. Vigorous attempts to privatize and commercialize content were made during the ‘dotcom’ era, with only modest success. Newspapers and other mass media built up substantial websites, attracting much attention away from Usenet and personal websites. However, attempts to charge for access to such sites have had very little success, and even reliance on advertising has proved to be problematic. Since the end of the ‘dotcom’ boom, the development of the internet has relied increasingly on collaborative efforts based on free sharing of information. UseNet is no longer a dominant forum for discussion and the provision of content, but not because newspapers or other commercial sources were able to provide a better market-based service. Instead, blogs and other amateur, collaborative content have taken the place of both UseNet and other commercial commentary. Furthermore, the range and scope of amateur content provided for free on the web knows almost no bounds. (Quiggin and Hunter 2008, pp. 216–17)

Moreover, as in earlier times, a common feature of new street literature is copying from ‘respectable’ sources to sources that are less respectable, their activities characterised by impoliteness, opposition to mainstream values and broad public access. Not surprisingly, such practices have led to claims of copyright infringement countered by claims of fair use on the part of copiers. This time the claims have fallen to US courts called on to interpret the ‘fair use’ defence codified in § 107 of the Copyright Act 1976,32 a defence that (notwithstanding the First Amendment) has historically been treated rather stringently when it comes to non-transformative uses of copyright material, stretching back to the classic case of Folsom v Marsh 33 where Story J. distinguished situations of ‘real, substantial condensation of the materials, and intellectual labor and judgment bestowed thereon’ from those involving ‘the facile use of the scissors; or extracts of the essential parts, constituting the chief value of the original work’.34 A.

Free Republic Sued

The keynote case is Los Angeles Times v Free Republic.35 The Free Republic website36 was founded in 1996 by Jim Robinson of California and it is dedicated to ‘roll[ing] back decades of governmental largesse, to root out political fraud and corruption, and to champion[ing] causes which further conservatism in America’. It claims to be ‘the premier online gathering place for independent, grass-roots conservatism on the web’. The following sample (taken in November 2007) gives examples of some typical posts: [On whether the US should bomb Iran (post by blam, 26/10/2007)]: Bomb bomb bomb, bomb bomb Iran


207

The conflict will not end with the taking out of the nuke facilities ... but it will move things along. We should not just bomb Iran – we need to destroy Iran! And soon! [On whether Roe v Wade was progressive, referring to an address b Justice Ruth Bader Ginsburg (post by wagglebee 26/10/2007)]: And we need to start showing this horror for exactly what it is! No progress, but the abortion ovens replaced those used by the Nazis in WWII. Of course, the procedures have been cleansed to some extent but the desired outcome of extermination is still the same. One abortion == one dead + one wounded. I wager this shriveled up old witch (Ginsburp) has never suffered the advances of even the vilest looking of men. Ginsberg is a psychopath – who could have been stopped but was given a pass by our pathetic party leaders (Dole, Lott, Hatch). [On whether Bush is a big spender (post by shrinkermd 26/10/2007)]: Any advanced math goes sailing over my head. If Hillary is elected President, my prediction is that she’ll break all of Bush’s dubious records within 2 years. She’ll spend as if there’s no tomorrow, which will be appropriate since she’s going to get us all nuked anyway. Party hearty, citizens of HillaryHell, the nukes are coming.

Nevertheless, Free Republic provides a vehicle for users to participate in a discussion forum of a quite different, and more radical (here radically conservative), character than offered by mainstream media outlets. As such, it bears a resemblance to the ballads, pamphlets and penny newspapers which served as a vehicle for dissent and debate in streets, squares, sitting rooms and drinking establishments. Similarly, while old street literature may have been low cost, reliant on its ability to keep printing and publishing costs to a minimal level, Free Republic can provide its services at absolutely minimal cost because the basic provision of the internet support system is free. In short, only those who choose to donate to Free Republic pay for the services it provides. Free Republic’s methods were also rather like those of earlier street literature. In short, copying was its modus operandi in its first years of operation. Typically, an article would be extracted or taken in whole from a published newspaper source and presented for comment and criticism by others participating in the newsgroup. A difference from Bell v Whitehead and Wilkins v

208


Aikin is that the first user presented the selected material for others to review rather than engaging in a critique themselves. But the intention was to prompt review and criticism (not so much of the stories but of the underlying ideas). There was no evidence that Free Republic resulted in less purchasing of newspapers or less traffic to newspaper websites. Nevertheless a number of newspapers objected to Free Republic’s republishing of their copyright material. In December 1997 cease and desist letters were sent to Free Republic and Jim Robinson, in September 1998 proceedings were launched by the Los Angeles Times and the Washington Post, and in October 1999 both parties moved for summary judgment. The newspapers argued that Free Republic should limit its activities to summarising and linking from their websites which made articles available at a small fee (in the order of $1.50 to $2.95 an article). Free Republic argued that its use was privileged by the statutory fair use defence (and protected under the free speech guarantee in the Constitution’s First Amendment). In March 2000 judgment was given for the newspapers. Applying a rather literal treatment of earlier US authorities,37 District Court Judge Margaret Morrow held that defendants’ verbatim copying failed to meet the standards of fair use prescribed by the section which states that: the fair use of a copyrighted work … for purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research, is not an infringement of copyright. In determining whether the use made of a work in any particular case is a fair use the factors to be considered shall include: (1) the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes; (2) the nature of the copyrighted work; (3) the amount and substantiality of the portion used in relation to the copyrighted work as a whole; and (4) the effect of the use upon the potential market for or value of the copyrighted work.

The decision, elaborated in Table 10.2, was (in essence) that ‘[s]ince the first posting of an article to the Free Republic site often contains little or no commentary, it does not significantly transform plaintiffs’ work’ (at 1461); the fact that defendants did not charge for the use (whereas the newspapers did) did not change the fact that ‘posting to the Free Republic site allows defendants and other visitors to avoid paying the “customary price”’ (at 1466), and given the number of registered visitors to the Free Republic website and the page views Free Republic attracts, ‘the court cannot accept defendants’ assertion that the site has only a de minimis effect on plaintiffs’ ability to control the market for the copyrighted works’ (at 1469).


Table 10.2

209

Free Republic case and the Statutory Fair Use Factors

Fair use factor: 17 USC § 107

Assessment by the Judge Morrow in Free Republic1

(1) – purpose and character of use:

The primary purpose of the posting to the Free Republic site was ‘to facilitate discussion, criticism and comment by registered visitors’ (at 1463). However, Free Republic did not show that the wholesale copying of articles was necessary to achieve the purpose of criticism.2 The first posting to the site often did not contain (much by way of) commentary. Visitors to the site also did not comment on the presentation of news in the reproduced articles, but instead commented on the underlying news itself, which requires a recital of the underlying facts, not verbatim copying the entire articles. Finally, the fact that Free Republic did not derive direct financial benefit from the use was not enough to tip the scales in its favour – rather ‘it is more appropriate to conclude that, while defendants do not necessarily “exploit” the articles for commercial gain, their posting to the Free Republic site allows defendants and other visitors to avoid paying the “customary price”’ (at 1466). (2) – nature of work: Because the works were predominantly factual, this weighed in favour of Free Republic – more so than if the copyright works had been fiction, for instance. (3) – amount and Wholesale verbatim copying of the plaintiffs’ articles weighed substantiality of heavily against Free Republic. Free Republic’s argument that it portion used: was necessary to copy the article to achieve its critical purpose was not accepted for the reasons in relation to the first factor – the amount copied was not proportional to the critical purpose. (4) – effect of use The articles posted on the Free Republic website were a upon potential substitute for the plaintiffs’ articles, made available in their market for or value newspapers and on their websites.3 The fact that internet users of work: would be able to view the plaintiffs’ articles on the Free Republic site was enough to show that there would be some effect on the market for the works, even though the plaintiffs did not provide evidence of lost revenue or that people viewing the Free Republic website would have instead visited the plaintiffs’ websites. The argument that the plaintiffs actually benefited from referrals from the Free Republic site was rejected for lack of evidence,4 noting courts have routinely rejected the argument that a use is fair because it increases demand for a copyrighted work. Notes: 1. Material from this table is drawn from the discussion in Los Angeles Times v Free Republic 54 USPQ 2d (BNA) 1453 (2000) at 1459–72.

210


Table 10.2

continued

2.

Especially when another option was hyperlinking: see Los Angeles Times v Free Republic 54 USPQ 2d (BNA) 1453 (2000) at 1463, the judge adding that ‘[d]efendants’ assertion that links expire after a period of time is presumably a reference to the fact that articles are available on plaintiffs’ websites free of charge only for a certain number of days. Thereafter, there is a charge for viewing and/or printing them. That this is so does not make linking plaintiffs’ websites to the Free Republic site “impractical”. It merely requires that Free Republic visitors pay a fee for viewing plaintiffs’ articles just as other members of the public do’. 3. Albeit a ‘poor substitute’: (at 1469). Here the judge appeared to be influenced by the fact that ‘the undisputed evidence is that Free Republic has approximately 20 000 registered users, receives as many as 100 000 hits per day, and attracts between 25 and 50 million page views per month’ (at 1469). The judge also appeared to assume that untapped, or not fully tapped, markets (of the plaintiffs) should be included in an assessment of market effects, on the authority of the Campbell and American Geophysical cases, to the effect that ‘[m]arkets for derivative works, ie, those markets “that creators of original works would in general develop or license others to develop,” must be considered in addition to the market for the original’ (at 1469). 4. Indeed there was evidence given for the defendants that the Los Angeles Times website ‘receives approximately 20 000 hits per month from users who visit the Free Republic site before accessing the Times’ site’, but this was regarded as inconclusive by the judge on the basis that it ‘does not address how many hits are diverted from plaintiffs’ websites as a consequence of the posting of articles to the Free Republic site, and this is the pertinent inquiry in terms of potential market harm’ (at 1471).

B.

Implications for Free Speech

Not surprisingly, there have been criticisms of the Free Republic decision from a range of commentators. As Frank Pasquale (2004, pp. 808–9) points out, the implicit assumption is that an order requiring linking to a newspaper site that charged for access to its archived articles can provide adequate access: But ‘book reviews would mean little if they could only quote from a book after obtaining permission from the book’s copyright owner’ and ‘I would rarely send articles from the New York Times website to friends if I had to pay a fee each time … [and] it is likely that bloggers who now generate traffic for such sites would not do so [either]’. The point is put even more strongly by Quiggin and Hunter: Free Republic highlights a problem with the cultural practice of some types of amateur content: the blogosphere could not exist in its current, vibrant form if copyright owners actually enforced copyright in relation to all the millions of infringements that take place on it each day. It is not an answer to say that copyright owners do not usually bother to sue, since this just leads to selective enforcement and an uncertainty in the scope of appropriate use. It would be better to establish a principle that, for example, non-commercial use of copyright material (as on a blog or in other amateur content forms) is not copyright infringement. (Quiggin and Hunter 2008, p. 247)


211

We agree that Free Republic is problematic. There is a certain illogicality to its stringent approach to the ‘purpose and character of the use’ factor with respect to copying of material that is little more than a mere chronicle of facts or news. In particular, why should transformative use be such a crucial aspect in assessing the character of Free Republic posts? Indeed, we might go further and ask why transformative use should not encompass the re-contextualisation of material for criticism and commentary? The re-contextualisation of works – placing them in new contexts – can have just as much an influence on their meaning as uses which change the attributes of the works themselves, drawing on insights from anthropology and cultural studies (for instance Hebdige 1979; Harrington and Bielby 2001) where the impact of consumption on cultural meaning has been closely examined. Arguably, such re-contextualisation should be considered a kind of transformation.38 But, at very least, recontextualising (and in their own way creative) uses should be considered valuable uses in offering new opportunities for speech and debate by new audiences for the material, adding to the spread of knowledge in much the same way as the radical 19th century street literature, irrespective of ‘transformation’. In the light of this conclusion, it is also a matter of particular concern that the court adopted such a restrictive view of the market harm element of fair use.39 The question, according to the judge, seems to be one of entirely hypothetical market harm rather than actual or even potential market harm based on a realistic assessment of the plaintiffs’ and defendants’ respective ideologies and target audiences (which in this case were arguably quite different – the plaintiffs being mainstream, moderately liberal newspapers, the defendants an extreme right-wing discussion group). Whether it would be preferable or even desirable to have a blanket exception for non-commercial use, as Quiggin and Hunter suggest, is debatable given that the history of street literature has been one that has included highly commercial albeit very cheap forms of publication. But we agree with these authors’ assessment that the Free Republic decision offers a distinct vision of copyright law being used to limit free speech in the new street literature environment. In the end, we would hope the District Court’s approach to fair use in Free Republic will not find favour in other courts. There are some indications that it might not be, with the US District Court taking a rather different approach to the question of assessing ‘market harm’ in the music file sharing case of A & M Records, Inc. v Napster, Inc.,40 decided soon after Free Republic.41 There the emphasis was on assessing actual and potential market effects with reference to survey and expert evidence. Unfortunately, the Court of Appeals chose to blur the question of which of the two approaches should prevail, suggesting rather bizarrely that both were correct.42 Thus, until the conundrum of which approach will prevail is resolved in a future case, it seems Free Republic can

212


be taken as good authority – including in Australian courts (in line with a trend, that of influence, that can generally be observed in copyright cases). If so, that would be unfortunate, in light of the potential chilling effect for a vibrant street literature in the internet domain.

IV.

CONCLUDING COMMENTS

In this chapter we have argued for a measured judicial development of fair use and fair dealing doctrines to deal in a nuanced fashion with cases of potential copyright infringements involving posting of copyright-protected material on blogs, newsgroups and other internet-based sites of discourse and debate about news and political events. In particular, we have argued that the rigidly freespeech restrictive approach of the US District Court in Los Angeles Times v Free Republic should not be followed by other courts, which should rather look to the more liberal history of fair use in the 19th century English newspaper cases. Of course, we should add, publishing practices are not entirely controlled by decisions on copyright liability. For one thing, remedies can also be important.43 So can bad publicity – and it is noteworthy that the terms of the postjudgment settlement between the parties in Free Republic were substantially lower than was awarded by way of damages.44 Third, and perhaps the most positive development after Free Republic, posts that now appear on the Free Republic website have adopted a practice of excerpting and linking to articles from newspapers that treat their archival news stories as open for free public access – suggesting that, notwithstanding Quiggin and Hunter’s scepticism (2008, p. 57) as to the merits of relying on selective non-enforcement, this may work well in at least some situations – although, of course, it does leave open the prospect that practices that are thought acceptable in the current climate may yet be challenged at a later date.45

NOTES *

We are grateful to Dan Hunter and Julian Thomas for their inspiration and comments on ideas presented in this chapter. The preparation of this chapter was funded in part by a MRGS Grant on ‘A Legal History of Opinionated Journalism’ and an ARC Discovery Grant on ‘Amateur Hour: The Socio-legal Construction of Amateur Media’. 1. Talking especially of ballads here but the point is generally made with respect to street literature in the rest of his introduction. When Shepard talks of street literature he means ‘the cheap ballad-sheets, pamphlets and other ephemera of the masses, which circulated from the dawn of printing right up to the end of the nineteenth century’ (1973, p. 13). Later he adds that ‘the broadside ballad, then, was a kind of musical journalism, the forerunner of the popular prose newspaper’, suggesting that he includes such newspapers in a broader understanding of ‘street literature’. As will be seen, we have an even broader definition in mind, embracing the internet era.


213

2. Los Angeles Times v Free Republic 54 USPQ 2d (BNA) 1453 (2000). 3. See Lee (1976, p. 291). Also see Harris (1990) (reasons attributed to population growth and dispersal, political interests and, to some extent, growing respectability and security of the joint-stock company in the wake of mid-century reforms to the companies legislation). 4. According to the State Library of Victoria’s Guide to Newspapers on Microfilm (Victoria, Melbourne) on the Library’s website at http://www.slv.vic.gov.au/about/using/guides/ resources/newspapers/vic_m2.html. 5. Morrison (2005, p. 167), quoting a spokesman for the press in Victoria in the Argus, 13 April 1872, although Morrison is sceptical, noting that Queensland also had a flourishing country press community at the time. 6. See Harris (1990, p. 114) (article in normally ‘respectable’ Walthamstow Guardian, which, galvanised by the radical Walthamstow Reporter, ‘became apoplectic in its coverage of local events’ (in the 1890s). 7. Editorial on question of Port Philip’s future, Portland Mercury, 19 April 1843. 8. Quoting from an 1842 edition of Bell’s Penny Dispatch. 9. See Humphries (1990, p. 40) (referring to p. 1 of 7 July 1850). Humphries points out that the article’s signatory, G.M.W. Reynolds, was a writer of lurid popular fiction, of similar style to his non-fiction writing. 10. Walter v Howe (1881) 17 Ch D 708. 11. 5 & 6 Vict, c 45, s 2 (the term defined to mean and include, … every volume, part or division of a volume, pamphlet, sheet of letter-press, sheet of music, map, chart or plan separately published’). It was also accepted that the form of expression could be protected under copyright law, even if the content of the expression (the news) could not – and it seems there was no statutory support for deciding otherwise, an attempt to gain protection of news in the 1842 Act having failed; see Seville (1999, p. 131, n. 7). 12. Walter v Steinkopff [1892] 3 Ch 489. 13. It was also held that, even if the mutual practice had been established, it could not override the law but the conclusion is questionable. Of two cases cited, Wyatt v Barnard 3 V & B 77 and Maxwell v Somerton (1874) 30 LT 11, one involved a practice that did not extend to the plaintiff and the other involved a one-way practice under which the publisher of Belgravia Magazine and Belgravia Annual allowed a 2d paper, the Bristol Mercury, to reprint extracts (and the minimal remedy with respect to past uses suggests the court had a nascent appreciation of the binding force of a waiver). In Walter v Steinkopff it was argued the practice was mutual. Had the evidence shown this, why should it not function as a contractual licence? 14. Wilkins v Aikin (1810) 17 Ves 422. 15. Bell v Whitehead (1839) 8 LJ Ch 141. 16. Lord Cottenham L.C. also suggested that this would not be a case for a remedy: ‘Here, the value of the extract is very minute and trifling, and if there is nothing else in the case, the extreme minuteness of value in the extract, and of injury sustained by the plaintiff, would be sufficient to induce the Court not to interfere.’ 17. See Wilkins v Aikin (1810) 34 Eng. Rep. 163; Walter v Steinkopff [1892] 3 Ch 489. 18. See Bell v Whitehead (1839) 8 LJ Ch 141; Campbell v Scott (1842) 11 LJ Ch 166; Whittingham v Wooler (1817) 36 Eng. Rep. 679; Scott v Standford (1867) LR 3 Eq 718 (involving the copying of a factual compilation of statistical returns of all coal imported into London). 19. The Copyright Act 1911 1 & 2 Geo 5, c 46, provided that: ‘the following acts shall not constitute an infringement of copyright: – (i) Any fair dealing with any work for the purposes of private study, research, criticism, review, or newspaper summary.’ 20. See generally, Seville (1999) for a detailed survey of the law reform process. 21. citing Hansard Parliamentary Debates, 25 April 1838, 27 February 1839. 22. See Seville (1999, pp. 132–4 and references). See also Editorial in The Times, 20 June 1838; p. 5, col. B (proposal for extended term ‘depriving the public to that access to sound and economical reading which (subject to the reasonable restrictions now existing) we maintain they have a clear right to’). 23. Most prominently, John Stuart Mill in his classic essay, On Liberty (1859 [1974]): (‘[w]e can

214

24. 25.

26. 27.

28.

29.

30. 31.

32. 33. 34.

Intellectual property policy reform never be sure that the opinion we are endeavouring to stifle is a false opinion; and if we are sure, stifling it would be an evil still’, pp. 76–7). For instance, in Abrams v United States 250 US 616 (1919) at 639 (‘the best test of truth is the power of thought to get itself accepted in the competition of the market’). Shepard (1973, pp. 14 and 35), for instance, doubts whether truth is important at all (‘crowds love lies and half-truth better than truth itself’) and sees the value of street literature as lying in its link with the ‘fierce energy of the crowds that created history, rather than to the culture of its rules or manipulators’. Although Mill himself did not use this argument for freedom of speech (being strongly of the view that the truth would prevail), it is noteworthy that in his arguments for freedom of conduct he does acknowledge that there may be merits in the freedom irrespective of the rightness of the conduct: On Liberty, pp. 132–3 (‘[i]f a person possesses any tolerable amount of common sense and experience, his own mode of laying out his existence is the best, not because it is best in itself, but because it is his own mode’ – there is no suggestion here that ‘his own mode’ is the right mode for anyone other than the individual). Walter v Lane [1900] AC 539, confirming the judgment of North J. at first instance. See Burrell (2001, p. 366). An example is Express Newspapers plc v News (UK) Ltd [1990] 3 ALL ER 376, where the defence of fair dealing was not made out because the defendant newspaper did not comply with s. 30(2) of the Copyright, Designs and Patents Act 1988, which required acknowledgement to be made to the plaintiff newspaper as the source. This was despite the fact that it was not, as observed by Browne-Wilkinson VC, ‘ordinary newspaper conduct’ to acknowledge the particular journalist who wrote the original story. See Burrell (2001), and further, for example, Associated Newspapers Group plc v News Group Newspapers Ltd and Ors [1986] RPC 515, where the court held that the owners of The Sun newspaper were not acting fairly for the purposes of reporting the news when they reproduced an exchange of letters between the late Duke and Duchess of Windsor, the rights to which were held exclusively by the plaintiff newspaper. Walton J. said that it was not necessary to reproduce the letters to deal with current events, citing Hubbard v Vosper [1972] 2 QB 84 (at 93) that it was not ‘fair dealing for a rival in trade to take copyright material and use it to his [sic] own benefit’. As for instance in the Express Newspapers and Associated Newspapers cases above (ns 27 and 28). Such cases may also raise breach of confidence issues: see, for instance (although here the plaintiffs were the actual subject of the reporting), Commonwealth v John Fairfax & Sons Ltd (1980) 147 CLR 39 and HRH Prince of Wales v Associated Newspapers Ltd [2007] 3 WLR 222 (in both cases fair dealing defences failed). TCN Channel Nine Pty Ltd v Network Ten Pty Ltd (2002) 55 IPR 112. For a useful discussion of this case, see Handler and Rolph (2003). In The Panel case, for instance, the news reporting defence was successfully relied on in respect of only 2 of the 20 clips complained about by Channel Nine (The Sunday Program – Drugs at Olympics and The Today Show – Opera House, the latter not appealed to the full Federal Court). In fact, there was some disagreement between the judges in the case as to where the defence should be available: see, for instance, Sundberg J. at 55 IPR 112, 113 (by a narrow majority the defence did not succeed in respect of The Today Show – Boris Yeltsin clip). Part of the difficulty appears to be the judges’ perception that news needs to be distinguished from mere entertainment and cannot easily be combined (even if theoretically it may be possible to combine news and humour): see Sundberg J. at 112–13 and further Hely J. at 131 (both apparently agreeing with Conti J. at first instance that there can be considerable difficulty in distinguishing news from entertainment). 17 USC § 107. Folsom v Marsh 9 F Cas 342, 345 (1841). In this classic case, concerning the copying of President Washington’s letters, Story J. appeared to have a rigorous standard of fair use in mind when he said (emphasis added): a reviewer may fairly cite largely from the original work, if his design be really and truly to use the passages for the purposes of fair and reasonable criticism. On the other


215

hand, it is as clear, that if he thus cites the most important parts of the work, with a view, not to criticize, but to supersede the use of the original work, and substitute the review for it, such a use will be deemed in law a piracy … There must be real, substantial condensation of the materials, and intellectual labor and judgment bestowed thereon; and not merely the facile use of the scissors; or extracts of the essential parts, constituting the chief value of the original work.

35. 36. 37.

38.

39.

40. 41.

42.

Similarly, the leading 20th century ‘newspaper’ case of Harper & Row Publishers, Inc. v Nation Enterprises 471 US 539 (1985) concerned publication of excerpts of President Ford’s memoir (that it was unpublished at the time it was copied also was significant to the Supreme Court’s conclusion against fair use). Per O’Connor J: ‘The Nation article is structured around the quoted excerpts which serve as its dramatic focal points’ and the unauthorised publication of the excerpts (approximately 13% of President Ford’s memoir) ‘directly competed for a share of the market for prepublication excerpts’ (at p. 565). Los Angeles Times v Free Republic 54 USPQ 2d (BNA) 1453 (2000). Extensive details of Free Republic’s purposes, aims and methods, as well as a running record of post and discussions, are given on its website at http://www.freerepublic.com/home.htm (27 November 2007). Including Folsom v Marsh 9 F Cas 342 (1841) and Harper & Row Publishers, Inc. v Nation Enterprises above n. 34, as well as the Supreme Court’s more recent fair use decision in Campbell v Acuff-Rose Music, Inc., 510 US 569 (1995), which emphasised the importance of transformative use in the context of a music parody, and the ‘derivative works’ case of American Geophysical Union v Texaco, Inc., 60 F3d 913, 918 (2d Cir. 1995), which Lemley (2007, p. 189) has described as ‘the beginning of a new approach to the analysis of market effect’. In the same way as Jeff Koons’ appropriation art was considered fair, and especially transformative, use in the recent Koons case: Blanch v Koons, 467 F3d 244 (2006) – and see especially the interesting comments of Sack Cir J at 253: ‘Koons is, by his own undisputed description, using Blanch’s image as fodder for his commentary on the social and aesthetic consequences of mass media. His stated objective is thus not to repackage Blanch’s ‘Silk Sandals’, but to employ it ‘“in the creation of new information, new aesthetics, new insights and understandings” …When, as here, the copyrighted work is used as “raw material” in the furtherance of distinct creative or communicative objectives, the use is transformative.’ See Pasquale (2004, p. 778), arguing that it seemed to be enough that the defendant had ‘impaired the copyright holder’s ability to demand compensation’ from any individual or group which has the ability and means to pay (the hypothetical lost opportunity to charge for use deemed to belong to the copyright owner by virtue of the copyright, without any assessment of the real-life prospects of the opportunity being realised), citing Blackmun J. (dissenting) in Sony Corp of America v Universal City Studios Inc. 464 US 417, 485 (1984). In the Free Republic case, for instance, it might well be argued that if the point of the website was to offer an independent voice in political debates, seeking permission from a mainstream newspaper (especially of a more liberal persuasion) was a very unrealistic prospect. A & M Records, Inc. v Napster, Inc., 114 F Supp 2d 896 (2000). In this case users’ files-sharing activities, suing Napster’s services, were held to have harmed the plaintiff record companies’ markets on evidential grounds, viz, that a survey introduced into evidence by the plaintiff found that ‘file sharing had resulted in a loss of album sales within college markets’ and, second, that, again agreeing with expert evidence adduced by the plaintiffs, ‘plaintiffs entered the digital download market very recently, or plan to enter it in the next few months, they are especially vulnerable to direct competition’ (at 908–11). While we might question the readiness of the judge here to believe that record companies did at the time of the Napster case have a genuine interest in entering the internet market for downloading single songs being exploited by Napster, we find this to be a more realistic and therefore superior approach to the assumption of market harm in Free Republic. See A & M Records v Napster, Inc., 239 F.3d 1004 (2001) at 1017 (‘We, therefore, conclude that the district court made sound findings related to Napster’s deleterious effect on the

216


present and future digital download market. Moreover, lack of harm to an established market cannot deprive the copyright holder of the right to develop alternative markets for the works.’) See also Los Angeles Times v Free Republic 54 USPQ 2d (BNA) 1453 (2000) at 1469–71. 43. As, for instance, in Walter v Steinkopff [1892] 3 Ch. 489. However, for an injunction and $1 million statutory damages awarded without consideration of possible flexibilities in the Free Republic case, see Los Angeles Times v Free Republic 56 USPQ 2d (BNA) 1862 (2000). Unfortunately there seems to be limited scope for flexibility in the award of statutory damages under the terms of § 504 of the US Copyright Act of 1976, and Lemley (2007) has argued that the provisions are in urgent need of reform. 44. The Los Angeles Times and the Washington Post, according to Pasquale (2004, pp. 8–9, n. 131), agreed to ‘payment of only nominal damages of roughly $5,000 to each plaintiff’. 45. See Hansell (2008) and Cadenhead (2008) for the Associated Press’s efforts to impose more stringent guidelines on newsgroups and blogs, and so on, which use its stories, including issuing take-down notices to the Drudge Retort asking it to remove items that contained quotations from AP articles ranging from 39 to 79 words, although the news service later backed down a little after criticisms from prominent bloggers.

REFERENCES Boyce, G., J. Curran and P. Wingate (eds) (1978), Newspaper History from the Seventeenth Century to the Present Day, London: Constable. Burrell, R. (2001), ‘Reining in Copyright: Is Fair Use the Answer?’, Intellectual Property Quarterly, 4, 361–88. Cadenhead, R. (2008), ‘AP Settles Dispute with Drudge Retort’, Workbench, 20 June, http://www.cadenhead.org/workbench/news/3372/ap-settles-dispute-drudge-retort. Grayling, A.C. (2007), Towards the Light: The Story of the Struggles for Liberty and Rights, London: Bloomsbury. Handler, M. and D. Rolph (2003), ‘“The Real Pea Souper”: The Panel Case and the Development of the Fair Dealing Defences to Copyright Infringement in Australia’, Melbourne University Law Review, 27, 381–422. Hansell, S. (2008), ‘The Associated Press to Set Guidelines for Using its Articles in Blogs’, New York Times, 16 June, http://www.nytimes.com/2008/06/16/ business/media/16ap.html. Harrington, C. Lee and D. Bielby (2001), ‘Constructing the Popular: Cultural Production and Consumption’, in C. Lee Harrington and D. Bieldby (eds), Popular Culture – Production and Consumption, Malden, Massachusetts: Blackwell, pp.1–16. Harris, M. (1990), ‘London’s Local Newspapers: Patterns of Change in the Victorian Period’, in L. Brake, A. Jones and L. Madden (eds), Investigating Victorian Journalism, London: Macmillan, pp.104–19. Hebdige, D. (1979), Subculture: The Meaning of Style, London: Methuen. Humphries, A. (1990), ‘Popular Narrative and Political Discourse in Reynold’s Weekly Newspaper’, in L. Brake, M. Harris and L. Madden (eds), Investigating Victorian Journalism, London: Macmillan, pp. 33–47. Lee, A. (1976), The Origins of the Popular Press 1855–1914, London: Croom Helm. Lemley, M. (2007), ‘Cultural Environmentalism @ 10: Should a Licensing Market Require Licensing?’, Law and Contemporary Problems, 70, 185–203. Mill, J.S. (1859 [1974]), On Liberty, Middlesex, UK: Penguin Classics. Mill, J.S. (1865), Considerations on Representative Government, London: Longman, chapter 6.


217

Morrison, E. (2005), Engines of Influence: Newspapers of Country Victoria, 1840–1890, Carlton, Victoria: Melbourne University Press. Observations on the Law of Copyright: In Reference to the Bill Introduced into the House of Commons by Mr Sergeant Talfourd (1838), London: Scott, Webster and Geary. Pasquale, F. (2004), ‘Breaking the Vicious Circularity: Sony’s Contribution to the Fair Use Doctrine’, Case Western University Law Review, 55, 777–811. Quiggin, J. and D. Hunter (2008), ‘Money Ruins Everything’, Hastings Communications and Entertainment Law Journal, 30, 203–56. Seville, C. (1999), Literary Copyright Reform in Early Victorian England, Cambridge: Cambridge University Press. Shepard, L. (1973), The History of Street Literature, Newton Abbot, UK: David and Charles. Skone James (1948), Copinger and Skone James on the Law of Copyright, London: Sweet and Maxwell. Williams, R. (1961), The Long Revolution, London: Chatto and Windus.

PART III

Intellectual property transfers

11. Commercialization of university research and free diffusion – what does experience show works best in and for Australia? Ann L. Monotti I.

INTRODUCTION

Universities have evolved in a haphazard and organic way. As Eric Ashby wrote in 1967, ‘A university is a mechanism for the inheritance of the Western style of civilization. It preserves, transmits, and enriches learning, and it undergoes evolution as animals and plants do. Like animals and plants, universities are products of heredity and environment’ (Ashby 1967, p. 417). Heredity accounts for the core themes in the business of Australian universities, namely teaching, research and involvement with the community. However, as Ashby notes, the environment modifies the ways in which we teach and research and engage with the community. While the fundamental principles remain, they evolve to meet the challenges of the environment in which universities choose to operate or find themselves compelled to operate in order to survive and grow. For instance, solitude and freedom in research were the essential features of the new idea that Wilhelm von Humbolt contributed to the foundation of Berlin University in 1810. While Australian universities have inherited these features, they translate in a very different manner in 2008 from the understanding of solitude and freedom in 1810 (Ashby 1967, p. 419). By way of example, in many areas of multidisciplinary research that require collaboration, it is not possible nor desirable for solitude in research, whereas solitary inquiry remains common in areas of the humanities and law. The same is true for other essential features of a university. The concept of ‘academic freedom’, which extols the virtues of the maintenance of truth, independence, objectivity, freedom of intellectual inquiry, scientific rigour and open intellectual inquiry, has been modified to some extent and even curtailed as the environment changes to meet practical challenges in both teaching and research. For example, a law school might be required to teach certain subject matter in core units that a law student must 221

222


pass to be admitted to legal practice. Industry requirements for accreditation of this kind will affect the ability of academics to maintain complete autonomy in how they teach specific units. This environmental impact upon inherited values is particularly evident in the means by which academics conduct research and diffuse its outcomes to the community, an aspect of the core theme of involvement with the community. Traditionally, university researchers within one or more institutions conducted research and published the outcomes openly through such vehicles as the press, journal articles and conference presentations. This remains the position for the bulk of university research which is made ‘publicly available to promote industry-wide economic growth, or provide environmental and social benefits for the community rather than sold commercially to generate direct economic benefits’ (HRSCSI 2006, p. 17). However, over time we observe an increasing focus upon protecting intellectual property rights in a relatively small area of university research to enable its ‘commercialization’: its diffusion in the eventual form of commercial products. This chapter is concerned with the relatively small amount of research that results in intellectual property (IP) (usually inventions) that is suitable for commercialization in conjunction, commonly with an industry partner. It will usually fall within one of two categories, namely: 1.

2.

the university identifies what research outcomes might be of interest to industry and seeks its financial involvement – a ‘technology push’ mode of innovation. This approach relies upon the supply of commercial ideas from the universities into business and is founded upon the assumption that basic research is the origin of the majority of innovation (HRSCSI 2006, 3.39); or the researchers work with industry partners to tailor research to meet market demand – ‘market pull’.

A common mode for commercialization uses the ‘knowledge production model’ which involves an ‘idea – patent – licence’ and possibly a ‘spin-off company’ in which the inventor or inventors remain involved. This is the concept of ‘commercialization’. It is the focus of this chapter because this is likely to be the more controversial aspect of university commercial activity. Formerly, the opportunities for university research to result in patentable inventions that could be commercialized were limited because most university research was too ‘upstream’ to be of interest to industry. Two factors in particular have changed this situation, namely: 1. developments in basic science, especially information technology, molecular biology (Wofsy 1986) and, most recently, nanotechnology

Commercialization of university research

223

(Lemley 2006) that have an immediately identifiable commercial application; and 2. judicial and patent office decisions that have clarified and broadened the scope of patentable subject matter, with the result that more fundamental university research may qualify for patentability, particularly in the abovenamed areas. Governments are increasingly aware of the connections between the research they fund, or that secures private funding, and the opportunities for universities to aid ‘innovation’ by enhancing the commercial outcomes from that research through protecting, developing and commercializing IP. They have identified ways in which universities should contribute to the economy and have developed policies to achieve these goals. In particular, these policies actively encourage the expansion of collaborative research between universities, government and industries and the identification, protection and commercialization of inventions with commercial potential. In the UK, this activity is articulated as a ‘third stream activity’ which follows on from the two priorities of teaching and research (Lambert 2003, p. 3).1 This chapter pursues the line that 21st century Australian universities must be commercially minded to some extent, although they need to be realistic about the money that might be earned through commercializing patentable inventions. The bulk of university research will be unsuitable for patenting and commercialization and will continue to be disseminated in the traditional ways that generate little if anything in the way of direct commercial benefits. However, patenting of inventions and the pursuit of their commercialization in a realistic and discerning way that assists the community’s drive for innovation is a sensible strategy provided that it is pursued in ways that protect timely diffusion of the research and the other traditional university values and goals as far as possible. This line of inquiry is pursued by first explaining the concept of commercialization of patentable inventions with reference to the criteria that bodies in various jurisdictions adopt to measure public research commercialization activities. Second, it identifies the arguments that support the proposition that some degree of commercialization of IP is a sensible strategy for universities to follow. Third, it explores the broader context of worldwide policy in this area to demonstrate the pressures that impact on university autonomy and influence their approach to commercialization of research. Fourth, it reviews a small selection of Australian university statutes and mission statements to show how institutional practice varies and does not necessarily remain constant within an individual institution. Finally, it draws some conclusions on how current Australian policy might assist the pursuit of commercialization in ways that protect timely diffusion of

224


the research and the other traditional values and goals. It recommends that we can learn from the US experience to develop more detailed guidelines upon (a) the way in which the commercialization occurs and (b) the publication practices of the institution. Provided that we have minimum standards applied across institutions that protect the traditional university goals and values as far as possible, registering patents selectively and effecting their commercialization with the financial support of external industrial partners is an important method for disseminating knowledge that is created in universities with both public and private funding.

II.

DEFINING COMMERCIALIZATION ACTIVITY

Until 2005, the then Department of Education, Science and Training (DEST) defined research commercialization in terms that limited it to commercial returns (DEST 2005, p. 28). One definition was ‘the process of converting science and technology, new research or an invention into a marketable product or industrial processes’ (Yencken and Ralston 2005, p. 6). A more limited concept again was the ‘knowledge production model (i.e. idea – patent – licence – spin-off)’ (HRSCSI 2006, 2.32), and the data recorded in surveys which commenced in 2000, followed by joint surveys for 2001–02 and 2003–04, were limited accordingly to gross income derived from licences and start-up company formation and levels of patenting (NSRC 2007, ch. 2). While these latter metrics enabled international comparisons, few universities were successful in generating substantial returns from a patent portfolio. Hence, this narrow definition made it difficult for universities to receive credit for the many ways in which they contribute to the nation’s innovation performance (HRSCSI 2006, ch. 2). In 2005, the Group of Eight2 universities questioned whether commercialization should have a wider scope (Group of Eight 2005). After detailed consideration in two further reports, namely the Howard Report (Howard Partners 2005) and the CCST Working Group on Metrics of Research Commercialisation Report (CCST 2005), DEST broadened the scope and adopted the following CCST recommended definition: For Australia’s publicly funded research, research commercialisation is defined as the means by which universities and publicly funded research agencies generates commercial benefits, thereby contributing to Australia’s economic, social and environmental well-being. This is achieved through developing intellectual property, ideas, know-how and research-based skills resulting in new and improved products, services and business processes transferable to the private sector. (DEST 2005, p. 44)


225

It is this definition that was subsequently endorsed by the House of Representatives Standing Committee on Science and Innovation (HRSCSI 2006, ch. 2) and which has been applied in the latest National Survey of Research Commercialisation Years 2003 and 2004 (NSRC 2007, p. 22). The results of this latter survey, which included 18 case studies of survey respondents, were published in August 2007 (NSRC 2007). The data collected3 spanned the following areas: 1. 2. 3. 4. 5.

IP activity (invention disclosures, patent and plant breeders’ rights (PBR) issued, patent and PBR holdings, and patent and PBR culled or lapsed); licensing activity; start-up company activity; research contracts and consultancy activity; and skills development and transfer.

This range is broader than that which is collected by the licensing surveys conducted in the US by the Association of University Technology Managers (AUTM) and in the UK by the University Companies Association (UNICO). Those surveys have a more limited purpose focus upon commercialization of IP alone. In this chapter, we also limit the concept of commercialization to refer only to IP activity, licensing activity and start-up company activity.

III. SOME DEGREE OF COMMERCIALIZATION OF UNIVERSITY IP IS A SENSIBLE STRATEGY That it be an objective of Australian universities to facilitate and encourage commercial exploitation of IP, and in particular patents for inventions, remains controversial to some extent (May 2006). It connotes images of secrecy and obligations of confidentiality that are imposed upon researchers to delay unreasonably the public release of the research. While it is acknowledged that ‘commercialization’ and ‘diffusion’ are not synonymous terms (the private sector frequently relies upon secrecy to maintain its competitive commercial edge), they are not necessarily incompatible. Public disclosure of the patented invention in return for a limited monopoly is the essence of the patent system. Proponents of commercialization point out the vital role of universities in the innovation process. On the other hand, its detractors raise a number of concerns. The criticism largely radiates from the US, principally within the universities and research institutions, as a result of the unintended consequences of the Patent and Trademark Law Amendments Act of 1980, known as the Bayh–Dole Act4 (‘Bayh–Dole Act’) and the later amendments passed in 19845 which impose statutory obligations on universities to license inventions

226


created with federal funding. The main areas of concern that appear in the literature are: 1.

2.

3.

4. 5. 6. 7.

The conflict between wide and timely diffusion of research and the ability to exploit commercially the IP in research that has commercial potential. There is a potential clash here between the demands of commercial partners for confidentiality agreements that may delay or prevent publication of results and the impact this may have upon free and open discussion among peers with a consequential reduction in the free dissemination of knowledge (Cohen et al. 1998; Fine 2000). Allied with this is the need for research generally to be accessible to the public on reasonable terms. However, open publication does not necessarily achieve wide access because copyright owners can use their rights to levy high charges for access to the material (Mitchell 2008; IFLA 2004; Monotti with Ricketson 2003, pp. 337–42; Eisenberg 1996; Rai and Eisenberg 2003; Hope 2007; Texas Digital Library).6 Personal and institutional conflicts of interest. Personal conflicts of interest can arise in consequence of financial interests of the researcher, but also of a spouse and dependent children, or where any foundation or entity controlled by the individual or his or her spouse is involved. They include such things as a direct or indirect financial interest in a private sponsor of research (Monotti with Ricketson 2003, pp. 457–85). The risk for personal conflicts of interest is particularly evident in projects involving clinical research and the use of start-up companies to exploit university IP. Institutional conflicts of interest may arise in consequence of the university’s financial investments, ‘especially investments that are not publicly traded, such as direct equity holdings or rights to equity in start-up companies which might in any way benefit from the results of the research’ (Stanford University 2005). Liaisons with industry and involvement in the patents regime may divert attention from pursuit of knowledge to pursuit of industry-driven goals, commercial outcomes and eventual development of saleable products. Critics allege that some universities have become greedy in prosecuting their patents to the detriment of open science and the intellectual commons. Failure to give the public its value for the federal contribution to research and development (Schacht 2007; Monotti with Ricketson 2003). The effect on universities of inconsistency in their policies for research commercialization (In the Public Interest 2007). Costly litigation may be necessary to clarify ownership of IP and to protect and defend patents. Universities have turned into ‘companies’ vying with industry and individual scientists for IP rights and patents.


8.

9.

227

Risk of harm to third parties caused by IP developed by the university – for example, product liability or environmental harm (Dueker 1997, p. 470). Lack of attention to teaching and university matters.

These are significant issues that will arise when universities commercially exploit their IP. However, the presence of these risks is not a reason in itself for rejecting commercialization of IP in universities as an appropriate activity in limited circumstances. It is probably fair to say that the opposition to patenting of university research and commercialization tends to be directed to particular fields of inquiry, such as biomedical, nanotechnology and IT research, and to the mode of commercialization and publication practices, rather than to commercialization per se. Even some critics accept that universities will maintain ‘robust intellectual property rights’ in certain areas while promoting open knowledge and creation in other areas (May 2006, p. 302). Moreover, the above risks are not necessarily exclusively related to commercialization of IP. For example, delays in dissemination of research may occur to protect the property of a commercial sponsor, but equally common motives are likely to include the desire to protect the researchers’ lead in the field and to delay dissemination of undesired results (Harman 2002, p. 157). There are at least six practical problems with any proposal that rejects commercialization of university IP that results from research as a valid activity for universities: 1. The loss of licensing income and other revenue that can arise from wellmanaged commercialization of IP. The key findings of the National Survey of Research Commercialization (NSRC 2007) showed an increase in research commercialization activity and outcomes flowing from Australia’s universities which, according to the report, ‘create significant income for the institutions’. In 2004, ‘institutions reported income from licenses, options and assignments (LOAs) of over $59 million, contracts and consultancies with end-users worth almost $900 million, and equity holdings valued at almost $185 million … Although determining the level of sales resulting from licensed technologies is difficult, institutions estimated these returns were in the order of $3 billion in 2003 and $4 billion in 2004 … In 2004 the cost to institutions of conducting research commercialization activities, including staffing and associated administration costs, was almost $70 million’ (NSRC 2007, p. 12; Jaffe and Lerner 2004). 2. The impact this may have on entrepreneurial researchers and their continuing employment within our universities. There are many reasons why individuals dedicate their careers to research, but the underlying common factor is passion. It may be passion to explore basic questions, passion to

228

3.

4.

5. 6.


contribute to the search for a solution to a major problem for society, passion to do something that is relevant to society. Whatever the motivation may be, universities must provide researchers with the right environment to help them pursue their passion. This requires the correct equipment, resources and intellectual stimulation. In a limited number of areas of research, the right environment also requires some collaboration with industry and the prospect for commercialization of IP. If a university rejects commercialization of research as an appropriate activity on the basis that it is incompatible with academic principles, it will limit the academic freedom of its researchers to choose their line of inquiry and may limit the ability of universities to attract the best minds to research. The belief of government that universities have an obligation to play a constructive role in economic and social renewal of their communities. As Marginson and Considine stated in their 2000 study of Australian universities: ‘There can be little dispute about the need for universities to play an active and constructive role in the economic and social renewal of their communities. That this involves closer ties with business and greater attention to the needs of consumers should not be in doubt’ (Marginson and Considine 2000, p. 5). The fact that effective transfer of research for community use may sometimes require patent protection for the basic science, particularly in the medical and pharmaceutical fields where significant risk and investment funds are involved in successful development of commercial products. Lack of alternative funding for developing biomedical and other capitalintensive advances. The increasing financial pressures on Australian institutions.

In addition, to reject commercialization of intellectual property as an appropriate activity for universities to pursue could result in the loss of the many other benefits that flow from well-managed, realistic and selective commercialization. These benefits include increased research contracts and consultancies, research grants, training and professional development activity; improvement in teaching and research arising from a better understanding of the needs of industry; enhanced university and staff profiles and promotional prospects; increased industry–university interaction; improving business and negotiation skills; access to increased resource capability (for example, personnel and research infrastructure); return of benefits to the community; increased flow of knowledge from and into the university; and educational opportunities for staff and students. In appropriate cases, patenting an original invention can protect universities against the impact that ‘lock-up by third parties’ can have on university research and teaching programmes (Monotti with Ricketson 2003, p. 47).


229

As long as governments, research authorities and universities are able to agree on parameters in managing these issues when academic research and commercial interests mesh, there seems to be no convincing reason to deny Australian universities and researchers the ability to exercise the rights of exploitation that IP laws give to them as owners. Nor is there any obvious reason to place an absolute prohibition on researchers pursuing their research passions within the academic environment in ways that may result in commercialization. Patenting of inventions and the pursuit of their commercialization in a realistic and discerning way that assists the community’s drive for innovation is an appropriate and sensible strategy for Australian universities, although they need to be realistic about the money they may earn from this activity. The relevant question, and the one that continues to trouble commentators, government policy makers and funding agencies as well as universities and researchers, is not whether they should commercialize their IP and collaborate with industry, but how they achieve a balance in which they reaffirm their highest values of protecting the integrity of their research, the well being of the human subjects who participate in it, and the trust of the public in maintaining integrity and open dissemination of research (Monotti with Ricketson 2003; van Overwalle 2006; Rai, and Eisenberg 2003; Capart 2006). Consideration of international policy and development in this area can provide considerable assistance in the formulation of a policy that will achieve that balance.

IV. THE CONTEXT: INTERNATIONAL POLICY AND DEVELOPMENT Before examining the position in Australia, it is instructive to consider briefly the international context in which Australian universities operate. This context is relevant for at least two reasons. First, it is clear that the international context influences the national policy development on innovation, which in turn directs the approach that Australian universities should take on the commercialization of research. Second, Australian universities do not operate in isolation of the developments that occur in other countries. Innovation networks are becoming increasingly global as universities and firms collaborate across national boundaries (OECD 2006, pp. 10–11). Australian universities look particularly to the US, but also the UK, to observe how their national funding bodies and universities manage IP ownership and exploitation, conflicts of interest and other associated issues. Governments in a number of our major trading partners have encouraged universities to protect and commercialize suitable IP which results from their

230


research. While they encourage publication, this is no longer seen to be necessarily a sufficient method of dissemination on its own to give the public its full value for public expenditure on the research. Moreover, they do not assume that productivity is an automatic function of their research grants (Guston 2000, p. 137) and they expect universities to protect and actively exploit the results of university research where appropriate. Policies to foster innovation have grown in importance and more countries are developing ‘formal plans and strategies for science, technology and innovation – and are backing them up with funding increases and changing institutional structures’ (OECD 2006, p. 5). A.

United Kingdom

Since 1948, when the British government established the National Research Development Corporation (NRDC) under the Development of Inventions Act 1948 (11 & 12 Geo 6 c 60, now repealed), successive governments have encouraged the development and exploitation of inventions resulting from public research (Monotti with Ricketson 2003, pp. 227–31). The encouragement of commercialization of university research is now an entrenched government policy but is not a legislated requirement. Instrumental in its implementation are the grant conditions imposed upon institutions that are successful in winning public research funding. A review of the core terms and conditions which the Research Councils UK7 imposes upon grant recipients discloses a policy that requires universities to balance the responsibility to commercialize in appropriate cases8 with the imperative to actively communicate research to the public.9 Reviews and reform recommendations In 2003, Richard Lambert completed his review of business–university collaborations for the UK government (Lambert 2003). The review strongly supported a policy that involves such collaborations, but pointed out that the ‘technology push’ model does not work with much research that is performed within universities. Typically, much of this research will be too upstream to be of interest to industry. An exception lies in areas of biomedical research, information technology and nanotechnology, where the fundamental science may also have commercial applications. The report recommended that universities and industry should be encouraged to engage also with an alternative model for innovation that looks to the research and development being tailored to meet market demand – ‘market pull’. The Lambert Review has been influential on UK policy and the government and policy makers now adopt both innovation models for university involvement, anxious to increase levels of university–industry collaboration in research and development. A similar


231

policy that endorses increasing university involvement in the innovation process appears in the 2007 review by Lord Sainsbury of Turville of the government’s science and innovation policies (Lord Sainsbury 2007). Some universities have grasped the bit of commercialization in their teeth. For example, Imperial College London established Imperial Innovations in 1986 with the aim of realizing the commercial potential of research conducted at Imperial College. In 2006, Imperial Innovations Group plc, majority owned by Imperial College London, floated on the London stock exchange, the first university-owned technology transfer company in the UK to take this step. At the time, the Rector of Imperial College, Sir Richard Sykes, stated that: ‘Our universities are powerhouses of innovation and it is vital for the UK’s economy that we get our best ideas out of the laboratory and into the marketplace quickly. Imperial College is world renowned for carrying out leading edge research, and Imperial Innovations has a track record of making this research commercially viable’.10 Imperial College London is not alone in pursuing commercialization with zeal. In its fourth Survey of University Commercialisation in 2005, covering data for the financial year 2003–04, UNICO reported that total licensing income had ‘increased from £31.3m in 2003 to over £40m in 2004, with the number of licence agreements more than doubling during this period.’ It concluded that ‘[t]he recent news that in the past two years alone, 20 spin-outs from UK universities have floated with a combined value of over £1billion, demonstrates the importance of the university sector to the national economy.’11 Nevertheless, it is important to place the significance of commercialization of university IP in the right perspective. The UNICO press release noted that while some universities are ‘international benchmarks’, commercialization activity in many UK universities is still in its infancy, 30% of participants in the survey having commenced their activities as recently as 2000. B.

United States of America

The practice of patenting in US universities goes back many decades. The issue was on the table as long ago as 1934. In that year the American Association for the Advancement of Science expressed the view that patenting could threaten open science and ‘undermine the public’s trust in universities as institutions committed to advancing knowledge for its own sake’ (Sampat and Nelson 2002, p. 140). Nevertheless, over the following years various universities experimented with patenting and by the 1970s there were many examples of patenting activity within US universities (Monotti with Ricketson 2003, pp. 231–8; Sampat and Nelson 2002), although it seemed that there was little evidence of patenting activity that was picked up and developed by industry (USGAO 1998; contra Eisenberg 1996).

232


The trend towards commercialization of university inventions accelerated with the passing of the Bayh–Dole Act (Bok 1982; Dickson 1984; Walterscheid 1990; Lee 1997), which was intended to promote wide use of technologies that might otherwise not be commercially developed. The current provisions, which are now expressed in 35 USC § 200, focus the policy upon dissemination of the research in the public interest to business rather than making profit for the universities. A university must give priority to US small business firms12 when licensing the invention if they show they have the resources and capability to bring the invention to practical application.13 Among the many other conditions imposed upon universities by the legislation is an obligation to protect the government’s interest in the invention14 and a prohibition on assignment of rights without permission.15 Hence, the US approach differs from that of the UK in that the obligation to commercialize inventions that arise from publicly funded research has detailed requirements that have statutory force. The US government and its funding organizations strongly endorse policies that require grantees to freely disseminate and share knowledge. Under the National Institutes of Health (NIH) Grants Policy,16 grantees are subject to specific obligations in relation to rights in data (publication and copyright), sharing of research data and sharing of unique research resources. In relation to data, NIH grantees are encouraged (but not required) to publish original research in primary scientific journals as a means for sharing knowledge. They are also expected to engage in timely release and sharing of final research data from NIH-supported studies for use by other researchers. ‘Timely release and sharing’ is defined as no later than the acceptance for publication of the main findings from the final data set. Subject to certain restrictions, investigators conducting biomedical research are required to make the research tools they develop with federal funding readily available for research purposes to qualified individuals within the scientific community (NIH Principles and Guidelines 1999). In addition, the NIH must be given a royalty-free, nonexclusive and irrevocable licence for the Federal government to reproduce, publish or otherwise use publications and data arising from the research and to authorize others to do so for Federal purposes. Reviews and reform recommendations The statutory obligations imposed upon universities under the Bayh–Dole Act mean that direct comparisons are not possible between the US, UK and Australia. The differences are evident when one considers the content of the reviews and reform recommendations in these countries. As discussed above and below, reviews in the UK and Australia tend to focus upon ways to increase industry–university collaborations and to enhance commercialization of IP in universities. In contrast, US reviews and reform recommendations


233

address problems that have arisen through the increasing involvement that universities have with industry. Those problems were identified above and a number which have led to more discriminating policy are discussed in more detail below. Licensing The first area of note concerns the conditions on which universities license their inventions. AUTM produces an impressive annual Licensing Survey to measure the levels of activity. However, without direction those who create and license university inventions risk losing sight of the government objectives of this activity, and in particular of the government policy underlying Bayh–Dole that universities are expected ultimately to license technologies in the public interest and for society’s benefit. While the emphasis of the Bayh–Dole Act is upon transferring grant-supported technology to local industry for development for the public good, generating income has become very important for the universities which are recipients of these grants. A conflict may arise when the commercialization of IP in ways that maximize the public interest limits the revenue and the available options for commercialization. Not all institutions and academics would necessarily place the public interest ahead of profit. Licensing practices vary considerably from institution to institution and from case to case. Furthermore, the reality is that more generous packages at one institution could lure ‘superstar’ researchers from one institution to another (Glanz 1999; Press and Washburn 2000), so there may be powerful reasons to induce an institution to minimize the importance of the public interest in any commercial negotiations. In 2006, Arthur Bienstock, then Dean of Research at Stanford University, convened a small meeting of research officers, licensing directors and a representative from the Association of American Medical Colleges (AAMC) to brainstorm about important ‘societal, policy, legislative and other issues in university technology transfer’ (In the Public Interest 2007). This meeting was partly in response to a number of factors, including negative press (Press and Washburn 2000; Silverman 2007) and criticism of the Bayh–Dole Act. The participating institutions produced In the Public Interest: Nine Points to Consider in Licensing University Technology17 to reflect the consensus that emerged from that meeting. This consensus recognizes, inter alia, the special role universities have in commercializing technology in the public interest and for society’s benefit. The document is lengthy, but the key points are summarized as follows: 1. Universities should reserve the right to practise licensed inventions and to allow other non-profit and governmental organizations to do so. 2. Exclusive licences should be structured in a manner that encourages technology development and use.

234

3. 4. 5. 6. 7. 8. 9.


Strive to minimize the licensing of ‘future improvements’. Universities should anticipate and help to manage technology transferrelated conflicts of interest. Ensure broad access to research tools. Enforcement action should be carefully considered. Be mindful of export regulations. Be mindful of the implications of working with patent aggregators. Consider including provisions that address unmet needs, such as those of neglected patient populations or geographical areas, giving particular attention to improved therapeutics, diagnostics and agricultural technologies for the developing world.

Since their release by Stanford University on 6 March 2007, AUTM has encouraged its members to endorse these nine points. However, the rate of endorsement has been slow.18 Patenting Demarcation between inventions that are appropriate to protect and commercialize and those that should be diffused freely is a second area in which policy guidelines have appeared to mitigate undesirable consequences of the Bayh–Dole provisions. The Bayh–Dole amendments were not intended to allow patents and other restrictions to stifle the broad dissemination of new discoveries and limit further avenues for research. However, as patent law allows almost anything to be patented, the fear is that proprietary claiming can create significant barriers – ‘anticommons’ or ‘proprietary rights thicket’ – to subsequent research and product development (Rai and Eisenberg 2003; Eisenberg 1996; Heller and Eisenberg 1998). The NIH has been proactive in addressing these issues. The Patent Policy of the Public Health Service (PHS) of the NIH provides a clear demarcation between inventions that are appropriate to protect and commercialize and those that should be diffused freely. The general principles, which provide a useful checklist for universities generally, can be précised as follows: NIH will not patent if: • no further research and development is necessary (exception is research tools); • the invention is in a low public health priority or there is no commercial interest; or • patenting will hinder technology transfer and access to the technology. NIH will patent if: • it is necessary to attract investment in research and development;


235

• the commercial or public health value of the technology warrants the expenditure of funds for patenting; or • patenting will facilitate access to the technology (Driscoll 2007, NIH Patent Policy). Conflicts of interest The third area which has been the subject of extensive review and proposals for reform is that of personal and institutional financial conflicts of interest (Knight 2003, p. 741; Monotti with Ricketson 2003, pp. 457–85). These must be managed effectively to maintain public confidence in research. In the US, the Federal government has regulated financial conflicts of interest in federally sponsored research since 1995 when the Public Health Service and the Department of Health and Human Services promulgated its regulation titled ‘Responsibility of Applicants for Promoting Objectivity in Research’19 (Monotti with Ricketson 2003, pp. 467–9). Under the regulation, responsibility for managing conflicts is given to the institutions. Since the promulgation of the regulation, a number of initiatives have proffered guidance for consistent implementation of these responsibilities among institutions. One such initiative was that of the AAMC, which published a report in 2001 of its Taskforce on Financial Conflicts of Interest in Clinical Research (AAMC 2001). A spate of conflict of interest scandals in the NIH in 2005 resulted in the AAMC joining forces with the Association of American Universities (AAU) to review the 2001 policy. In Protecting Patients, Preserving Integrity, Advancing Health: Accelerating the Implementation of COI Policies in Human Subjects Research, these bodies strengthened the earlier call for the effective management of serious conflicts of interest that university–industry collaborations can create in this area of research. The executive summary includes the following: A principled partnership between the academic community and industry is essential if we are to realize the promise of biomedical research, but such collaboration can also create serious conflicts of interest. These pressures compel academic institutions to reaffirm their highest values of protecting the integrity of their research, the well being of the human subjects who participate in it, and the trust of the public. (AAMC–AAU 2008)

The critical feature underlying the more detailed and nuanced recommendations in this revised report is the call to institutions to ensure that any variations among their conflict of interest policies and procedures do not lead to standards that are ‘less rigorous’ than those which the report contains. Effectiveness of diffusion of research The fourth area in which problems have arisen concerns the timely dissemination of research. In response to these

236


concerns, the NIH actively promotes the effectiveness of diffusion of research by acting to assist with the public access to federally funded published research. Since 2 May 2005, recipients of grants from the NIH have been subject to a voluntary policy on enhancing public access to archived publications resulting from NIH-funded research (NIH 2005). More specifically, the policy is intended to: 1.

2.

3.

create a stable archive of peer-reviewed research publications resulting from NIH-funded research to ensure the permanent preservation of these vital published research findings; secure a searchable compendium of these peer-reviewed research publications that NIH and its awardees can use to manage more efficiently and to understand better their research portfolios, monitor scientific productivity and, ultimately, help set research priorities; and make published results of NIH-funded research more readily accessible to the public, health care providers, educators and scientists.

However, it appears that there has been inadequate voluntary compliance with this policy and libraries20 have lobbied to secure a revised policy (issued on 11 January 2008 with an effective date of 7 April 2008) that makes the Public Access Policy mandatory (NIH 2008). All investigators funded by the NIH are now required by statute to submit or have submitted for them to the National Library of Medicine’s PubMed Central21 an electronic version of their final, peer-reviewed manuscripts upon acceptance for publication, to be made publicly available no later than 12 months after the official date of publication. Investigators must ensure that ‘any copyright transfer or other publication agreements allow the article to be submitted to NIH in accordance with the Policy’.22 The burden of compliance with this obligation falls on the investigators. However, as the Council on Governmental Relations (COGR) noted in a comment presented to the NIH on 20 March 2008, the administrative burden of this policy is likely to be challenging also for institutions which do not join in the relationship between authors and publishers, claim no ownership to copyright in scholarly publications but are required to meet the terms and conditions of all their agreements. It pointed out that, although a large number of journals submit journal articles directly to PubMed Central, the burden for compliance would be eased: if more journals would collaborate with NIH and the research community in meeting these statutory obligations. Recognizing the requirement, we hope that publishers will modify standard copyright agreements to include a provision that acknowledges that the author retains the right to provide a copy of the final peerreviewed manuscript to the NIH and to make the article available in PubMed Central within 12 months after publication by the journal. (COGR 2008)


V.

237

POLICY AND DEVELOPMENT IN AUSTRALIA

While there has been a degree of interaction between tertiary institutions and industry for some time (ASTEC 1980), it was not until the late 1970s that there was explicit encouragement for universities to develop research and other links with industry (ASTEC 1978–79; ASTEC 1980; AATS 1983; DST 1984; OECD 1985; ASTEC 1989; Buchbinder 1993, p. 335). Increased interaction was seen as one way in which to encourage research in areas that were relevant to the nation’s needs. Government innovation and commercialization programmes have mushroomed since then (HRSCSI 2006, Appendix E), and the pressure intensifies on universities to be an active part of the innovation system (NSRC 2007, p. 3) and to actively transfer ‘knowledge, skills and innovation, specifically via increased commercialisation’ (HRSCSI 2006, 2.78). As in the UK, Australia imposes no legislative obligation on universities to patent and license inventions but implements its policies through the conditions of research grants. The underlying expectation is that universities should file patent applications in ‘appropriate circumstances’ and have the autonomy to choose when to do so and how to protect and manage research commercialization generally within their institutions (Singhe et al. 2005; Allen Consulting Group 2004). Reviews and Reform Recommendations In contrast to the nature of reform proposals that are developed in the US, Australia is still focused upon finding ways to increase university involvement in innovation. The most relevant key policy development in recent times was the five year innovation strategy that was introduced in 2001, with a funding commitment through to 2006: Backing Australia’s Ability – An Innovation Action Plan for the Future (‘BAA’) Following the 2003 Mapping Australian Science and Innovation Report (Commonwealth of Australia 2003) which identified strengths and weaknesses in Australia’s innovation performance, the BAA was extended for a further five years to 2011. Three key themes underpin BAA, namely: 1. 2. 3.

the generation of new ideas (research and development); the commercial application of ideas; and developing and retaining skills.

Commercialization is one of the many ways in which universities transfer the knowledge that they acquire and generate for the benefit of the broader community (Phillips KPA 2006) and a number of recent reviews have evaluated ways in which to succeed with this venture (Allen Consulting Group 2004).

238


As in the US and UK, reviews have shaped policy that the government imposes upon universities principally through terms and conditions of public research funding agreements (Monotti with Ricketson 2003, pp. 221–6). In this regard, a critical development in 2001 that arose from the BAA strategy was the development by the Australian Research Council (ARC); the Australian Tertiary Institutions Commercial Companies Association; the Australian Vice-Chancellors’ Committee (AV-CC); the Department of Education, Training and Youth Affairs; the Department of Industry, Science and Resources; IP Australia, and the National Health and Medical Research Council (NHMRC), of a set of National Principles of Intellectual Property (‘NP’; ARC et al. 2001). While the emphasis of public research funding bodies is to support the best research in the national interest, the purpose underlying the development of the NP was ‘to assist researchers, research managers and their research institutions, in ensuring that they have access to best practices for the identification, protection and management of IP, and therefore, to maximise the national benefits and returns from public investment in research’ (NP, Preface). Compliance with these National Principles became a condition of national competitive grants.23 It is important to mention three significant features of accepting federal research grants. The first is that observance of these National Principles requires institutions to have an IP policy that covers specific matters, including provisions as to when institutional claims will be made to ownership, staff responsibilities and procedures to identify IP. The implementation of an IP policy was no arduous task because universities had flurried to revise and redraft new IP policies (Monotti with Ricketson 2003, pp. 221–6) following a 1993 AV-CC Discussion Paper on this subject that was later revised in 1995 (AV-CC 1993; 1995). By 2001, all had IP policies that dealt with such matters as who would own any IP; whether there would be any restrictions on the freedom of the academic to disseminate the research in his or her chosen mode and time; who would be responsible for deciding to patent any inventions; and how the researchers and their department would share any commercial return from their exploitation. The second is that institutions must consider the most appropriate way of exploiting the IP generated from the publicly funded research. The discretion for deciding to commercialize lies with the institution and no funding body provides guidelines for when patenting might not be appropriate. Compliance with this obligation requires considerable administrative infrastructure but, as with IP policies, by 2001 most universities had already responded to government pressures and/or blandishments by establishing companies and separate commercial offices to aid the identification and management of IP with commercial potential. The government retains no rights in any data or publications that arise from the research.


239

The third concerns the ‘obligation’ to disseminate the research, which arises from the Australian Code for the Responsible Conduct of Research (‘the Code’), not from the grant conditions. Developed in 2007 by Universities Australia, the ARC and NHMRC to replace the Joint NHMRC/AVCC Statement and Guidelines on Research Practice (1997), the Code includes the following: Dissemination of research findings is an important part of the research process, passing on the benefits to other researchers, professional practitioners and the wider community. Research activities supported by public funding are rarely complete until the results have been made widely available. (NHMRC et al. 2007, Section 4)

Unlike NIH funding conditions, there are no provisions that encourage the sharing of unique research data or research materials (Monotti 2006b). In response to widespread concerns about access to published scholarly work, the ARC introduced a condition in its 2008 grant conditions commencing in the 2009 funding year to provide for voluntary deposit of research data and publications in institutional and other repositories. Whereas under the 2008 conditions, the ARC encouraged researchers to consider the benefits of such deposits, the 2009 conditions impose this obligation upon the Administering Organization. Clause 20.3 provides as follows: The Administering Organisation shall consider the benefits of depositing the data and any publications arising from each Project in an appropriate subject and/or institutional repository wherever such a repository is available. If the Administering Organisation is not intending to deposit the data from a Project in a repository either before, or within six months after, the completion of a Project the reasons for not doing so must be detailed in the Project’s Final Report. Any research outputs that have been or will be deposited in appropriate repositories should be identified in the Final Report. (ARC Grant Conditions, cl. 20.3)

A voluntary obligation imposed upon institutions is likely to involve an expensive administrative burden and it remains to be seen how many universities are able to comply with the above provisions. It is to be recalled that the voluntary NIH policy on enhancing public access to archived publications resulting from NIH-funded research became compulsory in 2008, presumably because a voluntary obligation proved to be ineffective. In this regard, Australian universities will gain practical assistance from the methods that US universities adopt to comply with the compulsory access requirements imposed on PHS research grants. Some universities retain a non exclusive and irrevocable worldwide licence to make each faculty member’s scholarly articles available in this way (Mitchell 2008; Monash University 1994, s. 2.6), but this represents a minority. An added complexity for Australian universities will be their lack of authority to deal with the copyright scholarly works of their

240


employees if they have assigned or waived their copyright in those works under their institutional IP policy. It may not be a simple matter just to reverse the policy if their university statutes do not give them the power to expropriate the intellectual property rights of their employees24 (Monotti 1997). Objectives and Policies As pressures intensify from national and international sources for universities to supplement their funds with other sources of income, and engage directly in ‘innovation’, we see a shift in emphasis among some of the research-intensive universities. A report commissioned by DEST in 2005, Evaluation of Incentives for Commercialisation of Research in Australian Universities, (Yencken and Ralston 2005) found a commitment to an effective research commercialization strategy by senior management in most of the top performing (in terms of research commercialization) universities. Hence, as in the UK, universities participate in the innovation process without the need for a statutory obligation of the kind that the Bayh–Dole Act contains. A brief review of the objectives of three research-intensive Group of Eight universities, the University of Melbourne, Monash University and the University of Queensland, and a member of Innovative Research University Australia, Macquarie University, shows that they all adopt a role for commercialization of research, although as testamur to their individuality and autonomy, they place different degrees of emphasis upon that role. Both the Universities of Melbourne and Monash make subtle and indirect reference to this objective of commercialization. In section 4A of the Melbourne University Act 1958 (Vic), the objectives include teaching, research, equipping graduates to excel, serving the public interest and conferring degrees. There is no express mention of commercialization of research, although it could be implied as a possible object from the specific research objective ‘(b) to undertake scholarship, research and research training of international standing and to apply that scholarship and research to the advancement of knowledge and to the benefit of the well-being of the Victorian, Australian and international communities’. When one looks at the 2006 University Plan Growing Esteem (The University of Melbourne 2006), commercialization is well and truly on the radar. This publication ‘reaffirms Melbourne’s intention to be one of the finest universities in the world, highly regarded for making distinctive contributions to society’ in research, learning and teaching and knowledge transfer. The latter term expressly includes commercial development of research. Consistently with the policy in Growing Esteem, the university revised its Statute 14.1 Intellectual Property on 11 December 2006 (approved by the Minister 12 March 2007) to again claim for the university the principal role in


241

commercialization. In the repealed Statute 14.1 Intellectual Property (1999), it had expressly abandoned any role in the application for and prosecution of patents (The University of Melbourne 1999), and adopted instead a policy that passed both responsibility and opportunity for patenting to the originator of the invention. The guiding principles under the 2006 Statute are clear in their emphasis upon university commercialization of IP. Guiding principle 2 notes that assertion of ownership ‘provides clarity and security of ownership rights, thereby promoting the commercialisation of research and the transfer of knowledge to the wider community’. Guiding principle 3 provides: ‘In pursuit of its objects, the University strives to deal expeditiously, expertly and thoroughly with opportunities to commercialise IP in which it asserts rights, and in so doing, seeks to realise its full value for creators.’ Monash University’s commitment to commercialization is perhaps more pronounced. The university amended its objectives in Monash University Act 1958 (Vic) in 2003 to include in s. 5(b) the additional italicized words below, so that the objective now reads: ‘to aid by research and other means the advancement of knowledge and the pursuit of the benefits of its practical application to primary and secondary industry and commerce.’ Hence, commercialization of research was raised to the level of a university objective alongside research and teaching, but in indirect terms.25 At the most extreme is the University of Queensland, which now includes commercialization of research as an express function of the university. Originally established under the University of Queensland Act 1965 (Qld), this Act was repealed and replaced with the University of Queensland Act 1998 No. 5 (Qld) to, inter alia, extend ‘the capacity of the University to make policy on matters necessary to its good governance’. Section 5(h) of this new Act provides as follows: ‘to exploit commercially, for the University’s benefit, a facility or resource of the university, including, for example, study, research or knowledge, or the practical application of study, research or knowledge, belonging to the University, whether alone or with someone else.’ In contrast to these three universities, Macquarie University has one objective in s. 6(1) of Macquarie University Act 1989 (NSW), namely ‘the promotion, within the limits of the University’s resources, of scholarship, research, free inquiry, the interaction of research and teaching, and academic excellence’. It then distinguishes ‘principal functions’ and ‘other functions’. Commercialization of research falls within the ‘other functions’ category. Hence, while commercialization is a function of the university, it is expressly subordinated to the primary function of disseminating knowledge freely. These examples demonstrate that commercialization is clearly entrenched in the statutory functions that the modern Australian university expects to perform, alongside research and teaching. The manner in which commercialization of research can be performed is addressed in institutional policies,

242


including the IP policies which universities must have in place if they are to be recipients of national research funding from the ARC and NHMRC. A variety of models exist, but the general approach is for the university to claim ownership of IP created by its staff in the course of their employment, but to relinquish ownership of copyright in scholarly works (Monotti with Ricketson 2003, ch. 7). Often, the claims extend to IP that is not necessarily created in pursuance of employment duties but utilizes university facilities and resources. These IP policies provide each university with the opportunities to control the commercialization of research, provided that care is taken to ensure their validity26 (Monotti 1997; Monotti with Ricketson 2003). Many factors will influence the extent to which universities pursue commercialization and the levels of their success in these ventures. As an indication of the extent to which Australian universities have embraced commercialization activities, the annual survey measuring commercialization activity in Australian universities found a 74% increase in commercialization staff over the period from 2000–2004. Perhaps not unexpectedly, a large proportion of IP and licences, options and assignments activity was carried out by a relatively small number of institutions – the Commonwealth Scientific and Industrial Research Organisation (CSIRO) and a selection of universities, principally in the Group of Eight. For example, in 2003 and 2004, 10 universities filed no patent applications. In contrast, the majority of institutions were found to be active in research contracts and consultancies, skills development and transfer.

VI.

CONCLUDING COMMENTS

The international trend towards knowledge protection and appropriation through patenting may be subject to criticism and calls for greater controls, but is not one that shows any sign of being abandoned. Nor is it necessarily a trend that should be reversed. It is unrealistic to suggest that there should be no level of secrecy within university research, that no applied research is appropriate and that all commercialization in universities should cease. If one reads the numerous reviews conducted in this area, it is difficult to reject the proposition that some degree of commercialization offers positive benefits for universities, researchers and society. Prohibition of commercial links with universities to commercialize IP ignores the impediments that such a policy could pose to research, particularly in the biomedical and pharmaceutical areas that demand extensive financial investment before products are available for the marketplace. Assuming that universities ensure that they have valid claims to own and deal with IP that their employees create, the critical issues in this area for the


243

Australian government, research bodies and universities are selective and appropriate protection and management of IP that has commercial potential. The US experience suggests that if universities adopt profit as the overriding goal for commercial exploitation, either through greed or necessity, there can be undesirable consequences that can be challenging to manage. However, it also shows that difficulties can be minimized when there is a common set of standards that subordinates the profit motive to their traditional essential core values: freedom of inquiry; pursuit of truth; academic freedom; open and timely publications; advancing fundamental research and new knowledge; integrity and objectivity; and safety of human participants (SQW 2002; ARC et al. 2001). The relative lack of abuse in Australia may be the reason why our government and research authorities maintain a relatively laissez-faire policy that provides minimal guidelines in these areas and hands responsibility to the institutions themselves for their implementation. While direct comparisons are not possible with the US, it may nevertheless be wise to use this period of relative calm to review the existing grant conditions, policies and guidelines for their ability to provide universities with clear and sufficiently detailed central policies and guidelines on those areas that have caused trouble in the US. In the process, it would also be worth considering whether any of the various restrictions on commercialization that are contained in the Bayh–Dole Act merit inclusion in such policies and guidelines. Ultimately, what works in and for Australia is a commercialization environment where universities impose limitations and controls on this activity that satisfy a minimum national standard so that commercial interests do not threaten to displace principles of academic freedom and integrity of research.

NOTES 1.

2.

3. 4. 5. 6.

Two years after publication, the Lambert Working Group on Intellectual Property developed the Model Agreements toolkit (2005) which comprised a number of model collaborative research agreements. The objectives of the toolkit are to: facilitate negotiations between potential collaborators, reduce the time and effort required to secure agreement, and provide examples of best practice; http://www.innovation.gov.uk/lambertagreements/, 18 February, 2008. This term refers to the following universities: Australian National University, Monash University, Sydney University, University of Adelaide, University of Melbourne, University of New South Wales, University of Queensland and University of Western Australia. Seventy-nine institutions were approached to take part, with the following involvement: 4 publicly funded research institutions (all responded); 39 universities (38 responded for 2003, 39 for 2004); 36 medical research institutes (27 responded). Pub L No 96-517 (12 December 1980), implemented in 37 CFR Part 401. Pub L No 98-620 (1984). The chapter does not deal with the various consumer-led actions (particularly emanating

244

7. 8.

9.

10. 11.

12. 13. 14. 15. 16. 17.

18. 19. 20. 21. 22. 23.

24. 25. 26.

Intellectual property policy reform from the US, where commercial exploitation of university research is well entrenched) to reduce the adverse effects that assertion of IP rights may have on dissemination of research results. These include: free software, open source software, creative commons (which provides free tools that allow authors, scientists, artists and educators to easily mark their creative work with the freedoms they want it to carry), science commons (which aims to remove unnecessary legal and technical barriers to scientific collaboration and innovation), and the development of institutional and other repositories of published research and research papers. Core terms and conditions: www.rcuk.ac.uk/cmsweb/downloads/rcuk/documents/ tcfec.pdf, 16 February, 2008. RG 21: ‘It is the responsibility of the Research Organisation, and all engaged in the research, to make every effort to ensure that any potentially valuable results obtained in the course of the research are exploited, and that there is a suitable return to the Research Organisation and the researchers from any such exploitation.’ RG 20 ‘Public Engagement’: ‘It is the responsibility of the Research Organisation and the Principal and Co-Investigators to actively communicate the research to the public at both local and national level, and to raise awareness of the role of science and research in any related issues of public interest.’ See press release, 20 July 2006 at: http://www.imperial.ac.uk/P7996.htm, 28 March, 2008. The data in this survey were collected from 106 institutions, including 47 of the top 50 UK universities (by research income). But note the criticism in the Lambert Review, Executive Summary, p. 5: ‘there has been too much emphasis on developing university spinouts, a good number of which may not prove to be sustainable’. Defined in 35 USC § 201(h) to mean ‘a small business concern as defined at section 2 of Pub L No 85-536 (15 USC§ 632) and implementing regulations of the Administrator of the Small Business Administration’; 37 CFR 401.14(a)(a)(5). 35 USC §§ 200, 202(c)(7)(D) and 204; 37 CFR 401.14(a)(k)(4). 35 USC § 202(c)(4). 35 USC § 202(c)(7)(A). Part II: Terms and Conditions of NIH Grant Awards. Subpart A: Availability of Research Results: Publications, Intellectual Property Rights, and Sharing Research Resources, http://grants.nih.gov/grants/policy/nihgps_2003/index.htm, 21 March, 2008. The participating institutions that crafted this document were: California Institute of Technology; Cornell University; Harvard University; Massachusetts Institute of Technology; Stanford University; University of California; University of Illinois, Chicago; University of Illinois, Urbana-Champaign; University of Washington; Wisconsin Alumni Research Foundation; Yale University, and the AAMC. www.autm.net/ninepoints_endorsement.cfm, 19 February. The Australian National University is the only Australian university to endorse these points; www.anu.edu.au/ ro/info/newsletters/29november07.htm, 21 March, 2008. 42 CFR § 50.602 et seq. See: http://www.arl.org/sparc/, 24 March 2008. http://www.pubmedcentral.nih.gov/, 18 March 2008. http://publicaccess.nih.gov/, 21 March 2008. For example, the ARC Discovery Grant Funding Rules for Funding commencing in 2009 contain in clause 14.4.4: ‘Except with written approval from the ARC, all Proposals and ARC-funded research projects must comply with the National Principles of Intellectual Property Management for Publicly Funded Research … and accord with any intellectual property policies of the researcher’s organisation.’ University of Western Australia v Gray (No 20) [2008] FCA 498. Commercialization also features in the Excellence and Diversity Strategic Framework – 2004–2008, www.monash.edu.au/about/monash-directions/excellence.html, 24 March, 2008. University of Western Australia v Gray (No 20) [2008] FCA 498.


245

REFERENCES AAMC (2001), Association of American Medical Colleges, Protecting Subjects, Preserving Trust, Promoting Progress – Policy and Guidelines for the Oversight of Individual Financial Interests in Human Subjects Research, Introduction, www.aamc.org/research/coi/firstreport.pdf, 24 March, 2008. AAMC–AAU (2008), Association of American Medical Colleges and Association of American Universities, Protecting Patients, Preserving Integrity, Advancing Health: Accelerating the Implementation of COI Policies in Human Subjects Research: A Report of the AAMC–AAU Advisory Committee on Financial Conflicts of Interest in Human Subjects Research, https://services.aamc.org/ Publications/showfile.cfm?file=version107.pdf&prd_id=220&prv_id=268&pdf_id =107, 24 March, 2008. AATS (1983), Australian Academy of Technological Sciences, Developing High Technology Enterprises for Australia, Parkville, Vic.: AATS. Allen Consulting Group (2004), Building Effective Systems for the Commercialisation of University Research, www.universitiesaustralia.edu.au/documents/publications/policy/submissions/BCA-AVCC%20Report_Final.pdf, 24 March, 2008. ARC et al. (2001), Australian Research Council; Australian Tertiary Institutions Commercial Companies Association; Australian Vice-Chancellors’ Committee; Department of Education; Training and Youth Affairs; Department of Industry; Science and Resources; IP Australia; National Health and Medical Research Council (2001), National Principles of Intellectual Property: Management for Publicly Funded Research, www.arc.gov.au/pdf/01_01.pdf, 24 March, 2008. Ashby, E. (1967), ‘Ivory Towers in Tomorrow’s World’, Journal of Higher Education, 38(8), 417–27. ASTEC (1978–79), Australian Science and Technology Council, Science and Technology in Australia 1977–78, Canberra: Australian Government Publishing Service. ASTEC (1980), Australian Science and Technology Council, Interaction between Industry, Higher Education and Government Laboratories, Canberra: Australian Government Publishing Service. ASTEC (1989), Australian Science and Technology Council, The Core Capacity of Australian Science and Technology, Canberra: Australian Government Publishing Service. AV-CC (1993), Australian Vice-Chancellors’ Committee, Ownership of Intellectual Property in Universities: A Discussion Paper, Canberra: AVCC. AV-CC (1995), Australian Vice-Chancellors’ Committee, Ownership of Intellectual Property in Universities: A Discussion Paper, Canberra: AVCC. Bok, D.C. (1982), Beyond the Ivory Tower: Social Responsibilities of the Modern University, Cambridge, Mass: Harvard University Press. Buchbinder, H. (1993), ‘The Market Oriented University and the Changing Role of Knowledge’, Higher Education, 26, 331–47. Capart, G. (2006), ‘Should Universities File Patent Applications?’, Ethical Perspectives: Journal of the European Ethics Network, 13(2), 221–30. CCST (2005), Coordinating Committee on Science and Technology, Metrics of Research Commercialisation: A Report to the Coordination Committee on Science and Technology, www.dest.gov.au/sectors/science_innovation/science_agencies_ committees/coordination_committee_on_science_and_technology.htm, 15 February, 2008.

246


COGR (2008), Council on Governmental Relations, Comment Presented 20 March 2008, National Institutes of Health Concerning NIH Revised Policy on Enhancing Public Access to Archived Publications Resulting from NIH-Funded Research, http://206.151.87.67/docs/PublicAccessComment.doc, 30 March, 2008. Cohen, W., R. Florida, L. Randazzese and J. Walsh (1998), ‘Industry and the Academy: Uneasy Partners in the Cause of Technological Advance’, in Roger G. Noll (ed.), Challenges to Research Universities, Washington, DC: Brookings Institution Press. Commonwealth of Australia (2003), Mapping Australian Science and Innovation – Main Report, Canberra: Commonwealth of Australia, available at: http://www.dest. g o v. a u / N R / r d o n l y r e s / B O A 1 9 2 9 - F D 9 E - 9 9 D 9 - D 4 E E C 7 0 4 3 6 4 C / 2 6 8 8 / backing_Aust_ability.pdf Creative Commons, http://creativecommons.org/about/, 16 February, 2008. DEST (2005), Department of Education, Science and Training, Definitions and Methodological Notes: Statistics on Science and Innovation, www.dest.gov.au/ sectors/science_innovation/publications_resources/profiles/definition_methodological_notes_2005.htm, 15 February, 2008. Dickson, D. (1984), The New Politics of Science, New York: Pantheon Books. Driscoll, C.T. (2007), Technology Transfer Policies and Practice at the U.S. National Institutes of Health, E-Research Conference, 11–12 July 2007, Surfers Paradise. DST (1984), Department of Science and Technology, Proceedings of the National Technology Conference, Canberra: Australian Government Publishing Service. Dueker, K.S. (1997), ‘Biobusiness on Campus: Commercialization of UniversityDeveloped Biomedical Technologies’, Food and Drug Law Journal, 52(4), 453–510. Eisenberg, R.S. (1996), ‘Public Research and Private Development: Patents and Technology Transfer in Government-Sponsored Research’, Virginia Law Review, 82(8), 1663–727. Excellence and Diversity Strategic Framework (2004–08), www.monash.edu.au/about/ monash-directions/excellence.html, 24 March, 2008. Fine, P. (2000), ‘Academic Freedom Stifled by Contracts’, The Times Higher Education, www.timeshighereducation.co.uk/story.asp?storyCode=153737 §ioncode=26, 24 March. Glanz, J. (1999), ‘Headhunters Stalk the Halls of Physics’, Science, 284(5415), 733–4. Group of Eight (2005), Report on Outcomes of the Technology Transfer and Research Commercialisation Workshop, www.go8.edu.au/, 15 February, 2008. Guston, D.H. (2000), Between Politics and Science: Assuring the Integrity and Productivity of Research, Cambridge, UK: Cambridge University Press. Harman, G. (2002), ‘Australian University-Industry Research Links: Researcher Involvement, Outputs, Personal Benefits and “Withholding Behaviour” ’, Prometheus, 20(2), 143–58. Heller, M.A. and R.S. Eisenberg (1998), ‘Can Patents Deter Innovation? The Anticommons in Biomedical Research’, Science, 280(5364), 698–701. Hope, J. (2007), ‘Open Source Licensing’, in Krattiger et al. (eds), Intellectual Property Management in Health and Agricultural Innovation: A Handbook of Best Practices, New York: MIHR–PIPRA. Howard Partners (2005), The Emerging Business of Knowledge Transfer: Creating Value from Intellectual Property and Services, www.dest.gov.au/NR/rdonlyres/75374026-6458-4D0E-9CF7-6D589645D093/4079/7253HERC05A1.pdf, 15 February, 2008. HRSCSI (2006), House of Representatives Standing Committee on Science and


247

Innovation, Pathways to Technological Innovation, Canberra: Commonwealth of Australia, www.aph.gov.au/house/committee/scin/pathways/report.htm, 15 February, 2008. IFLA (2004), IFLA Statement on Open Access to Scholarly Literature and Research Documentation, www.ifla.org/V/cdoc/open-access04.html, 24 March, 2008. In the Public Interest (2007), In the Public Interest: Nine Points to Consider in Licensing University Technology, www.autm.net/ninepoints_endorsement.cfm, 19 February, 2008. Jaffe, A.B. and J. Lerner (2004), Innovation and Its Discontents: How Our Broken Patent System is Endangering Innovation and Progress, and What to Do About It, Princeton, NJ: Princeton University Press. JISC (2003), Digital Repositories Programme, www.jisc.ac.uk/whatwedo/ programmes/digitalrepositories2007.aspx, 24 March, 2008. Knight, J. (2003), ‘Accusations of Bias Prompt NIH Review of Ethical Guidelines’, Nature, 426(6968), 741. Lambert, R. (2003), Lambert Review on Business–University Collaborations: Final Report, UK: HM Treasury, p. 3, www.hm treasury.gov.uk/consultations_and_ legislation/lambert/consult_lambert_index.cfm, 24 March, 2008. Lee, Y.S. (ed.) (1997), Technology Transfer and Public Policy, Westport, Conn: Quorum. Lemley, M. (2006), ‘Patenting Nanotechnology’, Stanford Law Review, 58, 601–30. Lord Sainsbury of Turville (2007), The Race to the Top: A Review of Government’s Science and Innovation Policies, http://www.hm-treasury.gov.uk/ independent_reviews/sainsbury_review/sainsbury_index.cfm, 29 April, 2008. Marginson, S. and M. Considine (2000), The Enterprise University: Power, Governance and Reinvention in Australia, New York: Cambridge University Press. May, C. (2006), ‘Patents, Universities and the Provision of Social Goods in the Information Society’, Ethical Perspectives: Journal of the European Ethics Network, 13(2), 289–304. Mitchell, R. (2008), ‘Harvard to Collect, Disseminate Scholarly Articles for Faculty’, Harvard University Gazette Online, www.news.harvard.edu/gazette/2008/ 02.14/99-fasvote.html, 24 March, 2008. Monash University (1994), Statute 11.2. – Intellectual Property 1994 (as amended). Monotti, A.L. (1997), ‘Who Owns my Research and Teaching Materials – My University or Me?’, Sydney Law Review, 19(4), 425–71. Monotti, A.L. (2006a), ‘Research Use of Patented Technologies’, Agenda, 13(4), 351–64. Monotti, A.L. (2006b), ‘Access to Tangible Research Materials in Biomedical Research: Conditions of Access and their Effect on Research’, Journal of Law and Medicine, 14, 86–101. Monotti, A.L. with S. Ricketson (2003), Universities and Intellectual Property: Ownership and Exploitation, New York: Oxford University Press, Inc. Mowery, D.C., R.C. Nelson, N.S. Bhaven and A.A. Ziedonis (2004), Ivory Tower and Industrial Innovation, Stanford, California: Stanford Business Books. NHMRC et al. (2007), National Health and Medical Research Council, Australian Research Council and Universities Australia (2007), Australian Code for the Responsible Conduct of Research, www.nhmrc.gov.au/publications/synopses/ r39syn.htm, 24 March, 2008. NIH (2005), National Institutes of Health, Policy on Enhancing Public Access to Archived Publications Resulting from NIH-Funded Research (NOT-OD-05-022,

248


2005), http://grants.nih.gov/grants/guide/notice-files/NOT-OD-05-022.html, 18 March, 2008. NIH (2008), National Institutes of Health, Revised Policy on Enhancing Public Access to Archived Publications Resulting from NIH-Funded Research (NOT-OD-08-033, 2008), http://grants.nih.gov/grants/guide/notice-files/NOT-OD-08-033.html, 1 May, 2008. NIH Principles and Guidelines (1999), Principles and Guidelines for Recipients of NIH Research Grants and Contracts on Obtaining and Disseminating Biomedical Research Resources, National Institutes of Health (64 FR 72090, 1999), www.ott.nih.gov/policy/rt_guide_final.html, 21 March, 2008. NSRC (2007), National Survey of Research Commercialisation Years 2003 and 2004: Selected Measures of Commercialisation Activity in Universities, Publicly Funded Research Agencies and Medical Research Institutes, http://www.dest.gov.au/ sectors/research_sector/publications_resources/profiles/National_Survey_of_Resea rch_Commercialisation.htm, 1 May, 2008. OECD (1985), Organisation for Economic Co-operation and Development, Science and Technology Policy Outlook, Paris: OECD. OECD (2006), Organisation for Economic Co-operation and Development, Science, Technology and Industry Outlook, Highlights, www.oecd.org/document/62/0,3343,en_2649_34273_37675902_1_1_1_1,00.html, 1 May, 2008. Patent Policy of the Public Health Service (PHS) of the NIH, http://ott.od.nih.gov/policy/phspat_policy.html, 21 March, 2008. Phillips K.P.A. Pty. Ltd. (2006), Knowledge Transfer and Australian Universities and Publicly Funded Research Agencies, Department of Education, Science and Training, www.dest.gov.au/sectors/research_sector/policies_issues_reviews/ key_issues/commercialisation/knowledge_transfer.htm#Full_Report, 15 February, 2008. Press, E. and J. Washburn (2000), ‘The Kept University’, The Atlantic Online, www.theatlantic.com/issues/2000/03/press.htm, 21 March, 2008. Rai, A.K. and R.S. Eisenberg (2003), ‘Bayh–Dole Reform and the Progress of Biomedicine’, Law and Contemporary Problems, 21(66), 289–314. Sampat, B.N. and R.R. Nelson (2002), ‘The Evolution of University Patenting and Licensing Procedures: An Empirical Study of Institutional Change’, The New Institutionalism in Strategic Management, 19, 135–64. Schacht, W.H. (2007), The Bayh–Dole Act: Selected Issues in Patent Policy and the Commercialization of Technology, CSR Report for Congress, http://italy.usembassy.gov/pdf/other/RL32076.pdf, 21 March, 2008. Science Commons, http://sciencecommons.org/, 28 March, 2008. Silverman, E. (2007), ‘The Trouble with Tech Transfer’, The Scientist, 21(1), 40–45. Singhe, K., S. Playford, R. Percy and A. Quader (2005), Survey of Patenting and Commercialisation Activities of Australian Universities, Canberra: Innovation Division, DITR. SQW (2002), SQW Economic & Management Consultants, Managing Intellectual Property: A Guide to Strategic Decision-Making in Universities (Commissioned by UUK, AURIL and the Patent Office), www.ipo.gov.uk/managingipoverview.pdf, 24 March, 2008. Stanford University (2005), Institutional Conflict of Interest Policy (RPH 4.7, 2005), www.stanford.edu/dept/DoR/rph/4-7.html, 1 May, 2008. Taskforce on Finanical Conflicts of Interest in Clinical Research (2001), Association of American Medical Colleges, http://www.aamc.org/research/coi/firstreport.pdf.


249

Texas Digital Library, www.tdl.org/, 24 March, 2008. The University of Melbourne (1999), Statute 14.1 – Intellectual Property. The University of Melbourne (2006), Growing Esteem, University of Melbourne Strategic Plan, http://growingesteem.unimelb.edu.au/__data/assets/pdf_file/ 0010/86653/strategic_plan2006.pdf. UNICO (2005), The University Companies Association, Survey of UniversityCommercialisation, London: UNICO. USGAO (1998), United States General Accounting Office, Report to Congressional Committees (May), Technology Transfer: Administration of the Bayh–Dole Act by Research Universities, RCED 98-126, Washington, DC: USGAO. van Overwalle, G. (2006), ‘Reconciling Patent Policies with the University Mission’, Ethical Perspectives: Journal of the European Ethics Network, 13(2), 231–48. Walterscheid, E.C. (1990), ‘The Need for a Uniform Government Patent Policy: The D.O.E. Example’, Harvard Journal of Law and Technology, 3, 103–66. Wofsy, L. (1986), ‘Biotechnology and the University’, Journal of Higher Education, 57(5), 477–92. Yencken, J. and L. Ralston (2005), Evaluation of Incentives for Commercialisation of Research in Australian Universities: A Survey of Selected Australian Universities, report commissioned for the Department of Education, Science and Training, www.dest.gov.au/sectors/research_sector/policies_issues_reviews/key_issues/com mercialisation/incentives.htm, 15 February, 2008.

12. Pervasive incentives, disparate innovation and intellectual property law William van Caenegem I.

INTRODUCTION

Current interest in patent policy reaches beyond the academic community, as two recent newspaper articles demonstrate: one concerned how ‘a new technique for creating embryo-free human stem cells sidesteps a controversial US patent that has slowed the pace of scientific discovery worldwide’.1 James Thomson of the University of Wisconsin generated the patent but also the breakthrough which circumvents it. Since another group simultaneously reported the technique no ‘one team can control it’, according to the author. The other article related to the Australian National University (ANU), which was told by the Australian Universities Quality Audit (AUQA) to, ‘better promote the intellectual property attached to its research to raise its profile’.2 AUQA stressed that even if the ANU was not set to make much money from exploiting intellectual property (IP), at least exploitation ‘could raise ANU’s standing in the global community’.3 The first story illustrates that diverse and unpredictable events affect the value of patents. Patenting has created a dense web of rights within which interactions and exchanges occur in a highly complex and contingent manner. Single invention, single patent, single product linear innovation is not the norm. The second story has a now familiar ring to it: patents can play a significant (but unintentional) role in enhancing the value of goodwill or reputation attached to an innovator’s brand. Below the surface, however, lies a meta-tale about patent policy’s constant adaptation and mutation in response to theoretical critiques or uncomfortable empirical data: that patents do not actually induce innovation does not matter so much, since they have the benefit of strengthening your reputation. The ‘mutation’ literature and the rising tide of condemnatory critiques of the patents system4 invite us to revisit the complex relationship between intellectual property law and creativity within a contemporary industrial and commercial context. Creativity flourishes not simply because of incentives 250

Pervasive incentives

251

provided by intellectual property rights (IPRs), or specifically by patents in the industrial context; rather, all IPRs interact dynamically with multiple other factors to influence the nature, locality, direction and volume of creative output – whether in the arts or industry. The other factors that engender creativity or, focusing on industry, that act as incentives to invent are so multifarious and complex in their interaction that they are difficult to map.5 Nonetheless, we should not shrink from contextual accounts of the role of IPRs; in that light I revisit the account of patents as incentives within a contemporary socio-economic context below. Attention is first drawn to the perceived core role of patents as an incentive for firms to invent. Then the focus shifts to incentives for individuals to invent, or more particularly to generate, divulge and realise inventive ideas. In relation to the former I suggest that it is arguable that in the light of the pervasiveness of incentives to invent in contemporary industry and society the need for patents as incentives must be exposed to serious debate. At best patents could amount to little more than an additional and superfluous incentive. It may be that the account of patents as incentives to invent is so uncertain that a perception of patents as measures to enhance competition in particular circumstances should be preferred. In relation to individuals I argue that because incentives pervade society, innovation is disparate, that is, it happens simultaneously and heterogeneously in multiple organisations. This generates the need for legal instruments that underpin coordination of innovation, and also for legal settings that enhance the transfer of knowledge and information between organisations. Perhaps the most significant mechanism by which the latter occurs is through the mobility of knowledgeable individuals between organisations. In terms of the law, this means that employees should not be subject to unwarranted restraints, whether contractual or equitable, on their ability to draw on all they have learned from previous employment (in the same way that they can draw freely on anything learned during formal courses of instruction). Furthermore, in making their choices as to where their know-how and creative abilities will be most productively used they must be as free as possible from legal restraints on their use of knowledge. This freedom will have beneficial effects in terms of innovation performance.

II. PERVASIVE INCENTIVES FOR INNOVATION IN FIRMS AND THE NEED FOR PATENTS A.

Patents as Incentives to Invent

Patents are said to constitute a critical incentive to invent. However, clearly a

252


patent per se cannot operate as an incentive – a patent imposes costs. In fact the incentive is the potential return from monopolistic exploitation. But this return is very uncertain at the time of patenting, and many patents are never turned to account, so that the cost is never offset by monopoly gains. If one is turned to account, it is not clear to what extent profits are increased by the patent – in other words, would equivalent returns on an innovation have been generated in the absence of the patent? Further it is not clear at what level the incentive is supposed to operate – at the individual or the firm level? At the very least, considerable uncertainty surrounds the incentive account. Below I make an attempt to address it by rephrasing the incentive question: rather than ask ‘Does a patent work as an incentive to invent?’, I ask ‘Is there a shortage of incentives to invent which requires an additional incentive in the form of a technology monopoly?’ This is not a question to which I attempt to give a definitive answer; rather I point out that it is arguable both at the socio-cultural level, and at the industrial or economic level, that incentives to invent already pervade society in the absence of patents. This affects the account that posits that firms require an additional regulatory incentive to invent. Although there is no underlying consensus about the inducement effects of patents or about the need for patents as an incentive across industry, they are certainly perceived by both participants and commentators to be crucial in particular sectors, pharma being the outstanding example. But we have to be cautious in interpreting that fact as supporting the incentive theory of patents. The pharmaceutical products sector has certain very particular characteristics: the marketed product and the patented invention tend to be coextensive;6 the risk and sunk cost of investment in adopted research pathways is high, with great potential for total losses;7 alternative strategies to appropriate the advantages from innovation are largely inoperable; the potential lifespan of the product is relatively long; the cost of imitation and reverse-engineering is low; the required skills in manufacturing are generic and readily available; there is little room for product improvements and enhancements; and potential imitators tend to be highly informed and highly skilled. Substitution by new technologies (new drugs) is potentially the most significant source of competition in the sector, yet it is stymied by these characteristics, which may be further aggravated by particular conditions in any given geographical market. Products in other sectors may be different in any number of crucial respects: the patent and the product not coextensive; the time to market shorter; risks and expense of research and development smaller; potential for continual improvements greater; natural barriers to imitation higher (imitation lag is greater); secrecy more useable; other (legal) appropriation strategies available, and the like. Think software for instance, where the need for patents is greatly contested. So, rather than demonstrating a lack of incentive to invent new drugs, the


253

perceived sectoral importance of patents points to the peculiar characteristics of that product sector: they militate against dynamic competition by introduction of new technologies. There is good ground for arguing that there is a competition problem, which could be efficiently addressed by a technology monopoly, but it is not at all clear that there is an incentive problem. To expand this beyond the pharma sector, the problem a technology monopoly might usefully address is a failure of competition in particular product sectors, rather than a lack of incentives to invent in industry. B.

Pervasive Incentives to Invent

In reality incentives to invent pervade post-industrial society. The healthcare sector is a paradigmatic example of this: nowhere is the incentive to generate new solutions stronger. This pervasiveness is apparent both at the individual and socio-cultural level, and at the level of industry and the economy. Furthermore, the significance of the former cannot be ignored at the level of the latter: social and cultural factors that compel people to invent also impact at the level of the economy. They cannot be ignored in favour of an exclusively ‘financial’ analysis at the firm level. In industrialised nations such as those in North America, Europe and Australasia, a high premium is attached to individual creativity and originality. Conversely, copying, imitation and ‘following’ are ascribed less value.8 The culture of individualism engendering personal originality and creativity is very strong. In other words, in a technological setting, there is a strong cultural incentive to search for new and creative practical answers to problems, rather than to meet them with fatalistic acceptance or an exclusively spiritual response. This applies to even the most intractable issues facing society, whether they be more social (for example, sexism in society) or environmental (for instance, global warming or disease). There is also a strong socio-cultural innovation pull resulting from public perceptions about science: it is perceived as a potential source of new solutions to most problems. In other words, people inherently expect that solutions to problems will emerge from new and creative applications of scientific knowledge; science is not exclusively valued for its own sake. The public automatically assumes that creative individuals are somewhere engaged in plumbing the depths of science to find new solutions to problems. At the more institutional level of education and research, creativity, inventiveness and originality are also core values. Solid peer-review based incentive mechanisms underpin and institutionalise the search for new and original ideas and insights. Prize and other evaluation systems focus on creativity, novelty and originality, as does assessment for publication. The whole tendency of science and humanities research and teaching is towards the new, original and

254


different. Scientific research is dynamic, with an exponential growth in new knowledge in decentralised and multifarious institutions. None of this new knowledge would come about without deep and pervasive personal and institutional incentives. These incentives also compel individuals to seek out and immerse themselves in new environments that will enhance their own creative output and inventiveness. Labour market policy tends to prioritise mobility, so that the expectation that knowledgeable individuals will seek out the most fertile environment for creative work can be fulfilled. Another factor encouraging innovation is both socio-cultural and market based: consumer focus on novelty.9 The latest is the best, and much advertising, product development and market strategising is centred on and accelerates the cultural desire for the new and different. There is a constant demand for new versions and enhancements, improvements, changes in design and so on. At the same time society is very techno-focused: technology pervades lives and technology always changes and, either in reality or in perception, improves our lives over time. This view may not be universally accepted, but it is sufficiently pervasive to be a significant force in society. Economically or industrially also incentives for firms to innovate are pervasive. The conditions that now sustain the drive to innovate in industry are completely different from what prevailed during the industrial age. Then information channels were few, they operated slowly and were not universally accessible. Education levels were poor. The industrial infrastructure was heavy, solid and difficult to replace, lead times for innovation in plant and machinery were long, and specialisation in machine manufacture was not as advanced as it is now. The risks inherent in innovation in industry, with potential downtime and so forth, were very high. Anti-competitive conduct was not actively policed. Innovation was not a central competitive strategy because neither supply nor demand factors were as solid and pervasive as they are now. Investment levels in basic and applied research in the public and private sectors in education were far lower than they are today. Innovation was not disparate and widely distributed, nor so comprehensively studied or serviced in terms of management, the law and finance. In contrast, in this day and age innovation is a central competitive strategy for firms. Innovation-thinking permeates industry in most sectors of economic activity. No business can afford not to innovate or at least adopt innovation as a significant competitive strategy, whether at the level of business practices or technology. This competitive imperative is absolutely critical to thinking about patents law – if innovating is overwhelmingly mandated by competitive market conditions then the justification for patents becomes less obvious. At the same time a solid educational and broader infrastructure further encourages and enables innovation in industry: apart from general education in science and technology, specialists study innovation and teach it in business


255

and management schools around the industrialised world.10 Modern technology is readily available, including R&D-focused instruments and machinery – in other words, there is a dynamic specialist tooling sector that underpins innovation in industry. At the same time, flexible and innovative financial and management tools are constantly being refined to generate the required capital and management expertise to bring innovations to market. Various kinds of risk-financing are possible and available, with sectorial specialisation. Underpinning that activity are the legal skills and tools required to bring innovation to fruition in a complex and multifarious market with diverse and dispersed actors.11 This infrastructure invites the pursuit of constant improvements, adaptations and developments, while the university and research sectors provide a matching supply of new information, knowledge and data that can be turned to account. The focus on ensuring diffusion by commercialisation of public sector innovations is now very strong. Today, innovation is not sporadic or occasional but continuous – not innovating automatically equals losing the competitive race. Innovating is not a choice but an inevitable necessity, an essential survival mechanism. The above account seeks to give prominence to the centrality of the innovation dynamic in post-industrial society, both at the socio-cultural and economic level. This suggests that strong incentives to invent may already operate pervasively. Ascribing some merit to this account detracts from the potency of the incentive theory in patents law, that is, that patent monopolies, which come at a well-documented cost,12 are required because incentives to invent are lacking. That may have been the case in the past, but it is at the very least more contestable in a modern society and economy with genuinely competitive markets. C. Problems with Competition Rather Than Incentives? If we assume no shortage of incentives to innovate in a post-industrial economic climate, what does this mean for patents? Part of the answer may lie in the early origins of the system: patents offered protection for the introduction and establishment of industries relying on technologies new to the realm (either by invention or from abroad). In a market whose competitiveness had been corroded by the irresponsible grant of monopolies in existing trades and industries, patents encouraged competition by new entry into local manufacturing.13 A lack of competition was addressed by the grant of narrow monopolies (technology based, rather than product or industry based), encouraging new entrants to compete with incumbents. The grant in itself guaranteed no return to the grantee. Another pointer is the generally accepted fact, canvassed above, that certain

256


product sectors such as pharma consider patents to be crucial. Yet as also contended above, at the same time medicine is an area of science and research where the incentive to innovate is strongest: the social impetus is compelling and public R&D investment correspondingly high; peer based incentive structures operate acutely; and demand for new health solutions is close to insatiable. So arguably it is not a lack of incentives that necessitates patents here; it must be something else. The most obvious source is the presence in the pharma sector of the kind of conditions, discussed above,14 which strongly deter new technological entrants, and for which a state granted technological monopoly may be an effective remedy. In contrast, competition failures in other product sectors may neither be amenable to ‘the gale of creative destruction’ nor responsive to the grant of a technological monopoly. Other competition-enhancing regulatory measures (competition policy provisions) may be more appropriate. Yet as we know, patents apply without distinction to all technological products and processes, without express regard to the competition context – therein perhaps lies the rub. D. Patents Regulating Competition All this points towards fashioning patents law into an instrument for addressing competition failures in industries with certain product and market characteristics, rather than regarding it as a mechanism to redress a shortfall in incentives to innovate.15 The preconditions of grant and scope of the exclusive right should be adapted to modify stagnant markets by the introduction of temporary technological monopolies. If viewed in this manner the whole patents system is integrated into the broader complex of competition law. The apparent contradiction between patent monopolies and competition policy/regulation dissolves and the two can operate seamlessly to address competition deficiencies across all markets by different remedial interventions (some of which will be technological monopolies but others not).16 On a less ambitious scale, at the very least a competition focus suggests a conservative approach to patent extension. In the absence of the clearest evidence of competition failure in some identifiable product/process market, which a technological monopoly is actually adapted to redress, the boundaries of patentability should not be extended.17 Debates should go beyond the question of whether some new area involves technology, and centre on whether competition failure exists and is of a kind which can be efficiently addressed by grants of technological monopolies. Critiques of the patent system are now commonplace, and the approach above simply reinforces conclusions suggested by different angles of approach, which are largely based on concerns about proliferation and poor administration.


E.

257

Pervasive Innovation, Private Ordering and Mobility

Because incentives to innovate pervade society their effect is dispersed: they operate on many and diverse actors, resulting in polycentric and disparate innovation. Multiple independent firms and organisations in private and public sectors initiate innovation. As a result information and knowledge are generated and held in disparate locations by different actors.18 Innovation is not centrally planned nor solely Government controlled. Information and knowledge are everywhere and not limited to specific reservoirs and channels. An acute need for efficient coordination and diffusion results from this polycentricity. Innovation while disparate remains highly interdependent. Dispersed actors need to learn and acquire previously or concurrently accumulated knowledge and data from each other. Knowledge and information have multiple possible uses which may be recognised and exploited by others rather than by the actual holder or creator. There must be a system of revealing and exchanging dispersed knowledge and information. Our system is complex and organically grown: it incorporates the public domain, embedded in a number of structures, and the proprietary domain in which communication and diffusion on the basis of the rules of intellectual property play a significant role.19 But non-proprietary legal mechanisms also impact significantly on how knowledge is revealed, dispersed and exchanged between diverse actors. Equity and contract regulate an individual’s access to and use of tacit, nonproprietary knowledge and know-how.20 The contractual obligations entered into by autonomous parties and equitable default principles are ‘relational’ rather than proprietary, and they impose standards on personal conduct. In terms of innovation, their impact is critical: the most effective information exchanges are personal. They come about by person to person contact, and most importantly by the mobility of knowledgeable individuals between firms and organisations. People take their knowledge with them to work in different environments, new firms and organisations with distinct complementarities, and direct and daily interaction with new colleagues. These interactions and exchanges are productive of ideas, insights and innovation. The critical legal instruments that underlie them, that is, employment contracts, and associated principles of equity21 are further examined next.

III. A.

INCENTIVES FOR INDIVIDUALS TO INNOVATE Optimal Conditions for Individual Creativity

At the outset, this piece focused on incentives for firms. I surmised that in

258


today’s socio-cultural and economic environment such incentives are so pervasive that there is no demonstrable need for an additional statutory incentive to create or invent. That contention then pointed to a reconceptualising of patents as instruments to address competition failures in certain markets or industries, such as pharma, which might effectively be addressed by tailored technology-monopolies. But incentives to create must also be considered at the level of the individual rather than the firm. Not only do individuals in a mobile society act as critical diffusers of knowledge, as argued in Part I, but they are also the source of all invention. The creative genius ‘myth’ may rightly have been debunked in IP law, but there is no escaping the fact that new knowledge cannot be generated without the creative faculties, the mental processes of individuals. Nothing new in the practical arts comes about without the human mind. It is this insight that leads us to focus on the incentives that apply to individuals to create, to reveal, to communicate. However, in the modern environment individuals can rarely operate independently: they are dependent on firms and organisations which can accumulate and organise the complex and multifarious complementary resources that permit creative activity. The firm is the common environment for critical inter-personal exchanges. Most often, individuals are legally embedded in firms or other organisations by contractual employment relationships. Therefore the firm based employment context is the primary focus of attention here. Two interesting questions arise in this context: first, how to optimise incentives for individuals to invent, create, innovate in the firm environment; and second, how to optimise the conditions for the realisation of individuals’ creative ideas and inventions? Innovation policy and management literature tends to focus on the first question: how to structure material incentives for employees most efficiently. In terms of patents this becomes a debate concerning ownership; rights to exploit a patented invention as between employer and employee; royalty rights; patenting premiums and the like, and also about the primacy of contract or the need for statutory intervention in this area. But even though the incentive to innovate is universal, the micro-conditions for realisation of innovations are variable, which demands attention for the second issue: what are the optimal conditions for the realisation of the creative insights of an individual employee? Individual incentives should operate to encourage the revelation of creative ideas where the chances of realisation are the greatest.22 But what are the optimal conditions for revelation and realisation of new ideas? How do we generate or organise such conditions? These are exceedingly difficult and complex questions to attempt to answer in theory and a priori. Instead, I suggest a simpler starting point – that creative individuals themselves are best able to identify and evaluate conditions that stimulate their own creativity.23 And therefore knowledgeable individuals should be free to


259

exercise employment choices between firms and organisations that offer different conditions for creative activity. Presuming that alternatives do exist, a number of conditions must be fulfilled for individuals to be able to make and realise such choices. First, they must have access to information concerning the resources of, and conditions within, relevant firms;24 and second, labour market conditions and the law must leave them at liberty to put their choices into effect by migrating between firms and organisations, whether on a concurrent or consecutive basis. If there is sufficient freedom to migrate between organisations individuals will be more likely to find the conditions that optimise opportunities for realisation. B. Relevant Legal Mechanisms In terms of the relevant legal mechanisms, because most creative individuals are employees we are substantially within the province of labour law.25 The latter takes us too far away from the present IPR context, but there is a close connection between labour law and aspects of intellectual property law relevant to mobility. These aspects are employee entitlement to statutory IPRs; and trade secrets and contracts law. The predominant focus below is on the latter: legal restraints on individuals’ ability to draw on previously acquired know-how will act as a disincentive to move, and also restrain an individual’s creativity in a new environment. Mobility cannot have positive effects on creativity if individuals are not free to draw on knowledge they have previously gained. Some comments concerning employee entitlements to statutory IPRs are nonetheless appropriate here. If we favour the mobility of employees in full possession of their knowledge, experience, ideas and know-how – their creative tools – then there are two broad preferences: one favouring employee IPR ownership, or at least participation; and another favouring dealing with inventions retrospectively and individually rather than prospectively and universally. In other words, the rights of ownership and exploitation should be dealt with at the time the employee makes the invention.26 This policy approach may appear counterintuitive to firm managers, who might incline towards the broadest possible a priori claims over employee IP, but experience shows that commercial outcomes are not necessarily less favourable for firms that are more prepared to negotiate with creative employees.27 However, if the crucial question is to what extent individuals are able to migrate with all their knowledge, and conversely what rights an ex-employer can exercise over such knowledge, then property rights or ownership of inventions only provide part of the answer. With a view to mobility between firms (employers), equity and contract are as significant as patent ownership, if not more so. The freedom to use all knowledge held in the mind, all learning,

260


insight, know-how and skills absorbed with a previous employer, without restraint, in new and different conditions which an individual considers more conducive to invention and creativity, is critical. Legally this freedom depends on how rules and principles deal with knowledge that is either not subject to property rights, or over which no property claims have been staked. Often this is compendiously and loosely referred to as tacit knowledge, or legally as know-how. The contract of employment may seek to resolve contested claims to such knowledge, but equity and the law concerning restraints of trade impose marked limitations on contractual terms in this regard. In the absence of express contractual clauses, the equitable obligation of confidence imposes restrictions on the use of particular identifiable trade secrets post-termination. Other than in relation to those, the employee is generally absolutely free to use and deploy everything she has previously learned, whether or not she is now in direct competition with the ex-employer. However, the scope of the rights that employers enjoy over trade secrets under the mantle of the equitable obligation of confidence is both narrow and unpredictable. Narrow, because it extends no further than what can be properly particularised, demonstrated to be outside the public domain and communicated in circumstances of confidence. Unpredictable, because the dividing line between what is a trade secret and what is general know-how and experience lying outside the confines of trade secret protection is often very difficult to draw.28 What is more, there are various policy imperatives (freedom of competition, tools of trade theory and the like), and consequent principles underlying the law in this area, which do not generally favour the employer in a particular case. An employer’s ordinary reaction to this is to shore up its position by seeking to insert favourable terms into the contract of employment. These tend to take two forms, although there is considerable legal ingenuity in this area. The first is expansive confidentiality clauses; the second is covenants in restraint of trade. The latter are a well-recognised weapon in the employer’s armoury but are subject to very severe restraints in scope and enforceability. The general policy of the law is to hold such covenants unenforceable.29 However, in exception to that rule, reasonable restraints will be enforced by the courts if they are demonstrably justified to protect particularised employer trade secrets. This means that the employer must be in a position to establish, if the covenant is challenged in court, that it has such trade secrets in its possession, exactly what they concern, and that the ex-employee had access to them (or generated them in the pursuit of her duties). Furthermore, it must establish that the prohibition on working in a certain area and industry for a certain time is reasonably adapted to protecting the employer’s legitimate interests in its own trade secrets. The difficulty lies in predicting what a court will hold reasonable in the particular circumstances of a given case.30


261

In terms of express confidentiality clauses, the employer might be tempted to insert terms that include as much information as possible under the umbrella of contractual trade secrets protection. However, the courts have held that contracts cannot make trade secrets out of information that is in the public domain; and they equally demand particularisation. Abstract ‘umbrella’ terms that purport to cover otherwise unidentified items of information will often not be particular enough to enforce. But the alternative of identifying particular trade secrets in substance is of course difficult to do in an employment contract. It operates prospectively, intending to apply to as yet unidentified information that will emerge at some later date, either by way of the employee herself or in some other manner within the organisation. It may be possible to identify existing trade secrets at the time of entering into the employment contract, but usually it is too onerous to do so comprehensively. But it will be far more difficult, if not impossible, to do in relation to future or anticipated trade secrets. So the employer may have little alternative but to refer to abstract categories (the ‘trade secrets’, the ‘confidential information’ and so on) in the contract. However, when seeking to enforce such terms, it will be in no better position than when, in the absence of express provisions in the contract, relying on equity alone: having to establish what is a genuine trade secret, as opposed to general know-how the employee is entitled to use. And then the law has a further difficulty where an employee’s ideas are inchoate: courts tend to limit the grant of equitable remedies where an idea is not fully worked out, its practical use not fully described and its commercial value uncertain. Thus the law may superficially appear to favour the employer but in its particular application in fact tend to support an employee’s right to leave and escape the ex-employer’s control over all her acquired knowledge, other than in relation to a narrow and difficult to prove category of genuine trade secrets.31 This favours employee mobility, and means that knowledgeable and creative individuals have a large and unrestrained measure of freedom to determine where to work – in other words, to decide where their chances of creative output, realisation and/or dissemination are optimal. If we accept, as suggested above, that individuals are the most capable of assessing optimal conditions for creative work and realisation, then this is positive. Rather than attempt to identify ideal conditions for creative work theoretically and a priori, we can presume that those firms with the most conducive conditions will be most effective in attracting creative individuals, and will thus be most likely to survive and prosper. However, two worrying trends invite some comment. The first is a tendency in some quarters to elevate notions of contractual autonomy, in particular in relation to knowledgeable (that is, well-educated) employees, above notions of public policy that require active policing of restraints of

262


trade.32 The second is the fact that there is a dichotomy between the perception and the reality of legal rights in relation to restraints and broad confidentiality clauses. Although there may be severe doubts about enforceability of certain clauses and restraints, it may be that most employees are unwilling to challenge them in court, given the risks involved in doing so. Restraints that are unreasonable in terms of the law may in fact be observed in practice. Arguably this would be detrimental in terms of the policy settings discussed above.

IV

CONCLUSION

I have highlighted the importance of thinking differently about incentives and innovation in two ways. First, at the level of the firm I have argued that the focus should not be on incentives to innovate, because such incentives arguably pervade post-industrial societies.33 Rather the focus should be on patents as a legal device to enhance competition in markets that are not dynamic due to various factors, where there is reason to suppose that the problem can be effectively addressed by technological monopolies. Additionally, I have argued that thinking about incentives must also focus on individuals. This requires thinking about which incentives will induce individuals to be inventive, but also about the ideal conditions for generating and realising creative output from individuals. Identifying those conditions theoretically or empirically is arguably too complex; a reasonable alternative is to accept individuals as rational maximisers of their own inventive or creative capacity. They are the best placed to make the ex post facto choice between existing firms and organisations as to which will maximise their chances of creative output. From that perspective it is beneficial to ensure, in terms of legal policy, that individuals have sufficient access to information to identify the firms with the most potential, and that they enjoy conditions of mobility with all their knowledge and know-how, so they can act on their preferences and transfer between organisations with a view to maximising their inventive output. This will encourage the realisation and diffusion, whether by commercialisation or otherwise, of their creative ideas.

NOTES 1.

Leigh Dayton, ‘Cell Find Sidesteps Patent’, Higher Education Section, The Australian, Wednesday 28 November 2007, p. 25. 2. The Australian Universities Quality Audit (AUQA) is the Australian Government Agency for regularly reviewing the universities’ quality control processes; its reports are available at: http://www.auqa.edu.du.

Pervasive incentives 3. 4.

5. 6. 7. 8. 9. 10.

11. 12.

13.

14. 15. 16.

17.

18. 19.

263

See Brendan O’Keefe, ‘ANU Told of Ways to Raise Profile’, Higher Education Section, The Australian, Wednesday 28 November 2007, p. 26. See for example, Kingston (2004); Lichtman and Lemley (2007); Schankerman and Pakes (1986); Schankerman (1998); Moore (2005); Abramowitz (2007); Janis (2002); Bessen (2008); and Jaffe and Lerner (2004). In relation to weeding out valueless patents, see Ayres and Parchomovsky (2007). Patent renewals also indicate value attached to patents: see for example, Scotchmer (1999). As to alternatives to patents, see for example, van Ypersele and Shavell (2001). Ranging from the personal and psychological to structural and economic factors, education and the like. In today’s complex technological environment this is, generally speaking, a rare occurrence. In other words, the investment in research may be totally lost, in that there are no alternative or tangential uses of the results of research and development. As to how this came about, see van Caenegem (2003); see also Alford (1995). One illustration is the continuous renewal in the former of ‘versions’ of some basic product. The law tends to reflect this cultural given, for instance in relation to the requirement of inventiveness in patents law: see Duffy (2007). Journals too numerous to list specialise in innovation policy and science policy. Some examples are: Innovation: Management, Policy and Practice; Creativity and Innovation Management; European Journal of Innovation Management; International Journal of Innovation Management; Prometheus; Research Policy; the Journal of Innovation Law and Technology; and various US university law and technology journals, for example the Berkeley Law and Technology Journal and the Harvard Law and Technology Journal. See Petrusson (2004). If incentives to create are pervasive, the role of patents as incentives is diminished; but what about the other main benefit claimed for the patent system, that is, publication of information about inventions that would otherwise remain secret? If we accept that there is a sufficiently strong incentive to innovate without patents, then the logical consequence is that information about inventions will be published – not in the form of patent literature, but in reified form. Generally speaking, inventions that can be exploited and kept secret at the same time are not patented. Further, there is considerable scepticism in the literature concerning the effectiveness of disclosure and diffusion by patent specifications (see for example, Macdonald (2003)). Others point to the fact that patenting requires a degree of secrecy in relation to information that may previously have been published freely in the academic disciplines; see for example, Grushcow (2004). That is why the so-called first patent statute, the Statute of Monopolies, is actually no such thing, but a statute primarily abolishing the right of the Crown to grant monopolies. The antecedent period of monopolies was marked by a failure to introduce into England competitive technologies that were used elsewhere in Europe. See text around notes 6 and 7. Furthermore, developing a taxonomy of these conditions is essential and it has not been sufficiently addressed to date. For some interesting insights in relation to monopolies and patent rights, see Economides and Herbert (2008). It is suspected that the most amenable markets are those where the technology component of the end product is highest: contrast for instance the pharma market, where the product is nearly 100% technology based, with supermarket retailing, where technology is a marginal component of the end product. Or conversely, will not be either without beneficial effect, or actually detrimental to the innovation dynamic. This has relevance for recent debates concerning technology, for example relating to non-technical inventions, computer programming and business methods; but also more broadly to creativity and innovation in design based industries such as fashion. Often competitive research initiatives also multiply because of dispersed educational infrastructure, and disciplinary specialisation which impedes effective signalling and diffusion. Some scholars emphasise that property rights such as patents perform exactly this function – enabling patentees to act as secure coordinators of the downstream, applied research process under the umbrella of a relatively broad patent monopoly. Others tend to favour

264

20. 21.

22. 23.

24.

25. 26.

27.

28. 29. 30. 31.

Intellectual property policy reform narrower patent monopolies emphasising that proprietary patent rights can be a vehicle for the structured exchange, sharing and cross-licensing of knowledge assets between dispersed patent holders. These differing approaches are canvassed in van Caenegem (2003). The scope of patents is limited: recorded knowledge, appropriable under the patent scheme, only concerns a small fraction of all the knowledge that needs to be shared or exchanged. This is not to say that only employment relationships or interactions are significant in the innovation constellation. Where there is a complex network of dispersed but interdependent actors, knowledgeable intermediaries are also essential. Newer forms of interaction between private and public sectors, between education and commerce have added further to complex patterns of production, exchange and consumption of knowledge. This complexity requires intermediaries with the expertise and accumulated knowledge to deploy sophisticated legal, management, financial and accounting tools effectively. The importance of lawyers, of accountants, of commercial banks, of financial entrepreneurs and of other knowledge brokers and facilitators increases with the complexity of the environment. They all acquire a stake and some influence in the system, which will have its effect on its internal dynamics, growth and future development. All these interactions are underpinned by law, mostly contractual provisions concerning exchanges of knowledge and information, but also default rules of equity that govern professional interactions. This is not a novel point: see for example, Merges (1999). This approach borrows from a fundamental tenet of market economics: that individuals presented with ex post alternative consumption choices generated by multiple competing providers will more efficiently maximise their own welfare than planners directing the supply of goods and services in reliance upon a priori assumptions about consumers’ preferences. For example, what research is being done/by whom/in what context/with what resources? Well-informed employees require an effective publicity regime and quick and efficient diffusion of new knowledge, ideas and inventions. Obviously the patent literature is one way in which this is supposed to happen, but scepticism has been expressed at how effective the publication compulsion actually is, given the difficulties that exist in searching patent literature and databases (see above note 12). Nonetheless, if we take both the mobility, and the access to information which makes it relevant for employees, seriously, then this is a critical aspect of the patents system. Although in the present fractured work environment contractors also play a significant role; see Arup, Gahan, Howe, Johnstone, Mitchell and O’Donnell (2006). This is not to suggest a mandatory model for every employment situation: naturally the proposed bias in favour of employees must be balanced with other policy considerations, including freedom of contract and party autonomy. There are also issues of efficiency and delay and the problem of employers being held to ransom. Most significantly, a ‘retrospective’ approach invites a focus on process (an adequate and efficient framework for negotiation once an invention arises) rather than outcome (that the employer should own everything). Although a close analysis of the cases tends to belie the general proposition that employers own IP created by employees in employment: see for example, Victoria University of Technology v Wilson [2004] VSC 33 (18 February 2004); University of Western Australia v Gray (No 20) [2008] FCA 498 (17 April 2008); and also Polwood Pty Ltd v Foxworth Pty Ltd [2008] FCAFC 9 (18 February 2008) concerning ownership of inventions. See Faccenda Chicken Limited v Fowler [1987] Ch. 117 and more recent cases, as canvassed in van Caenegem (2007). An established principle that goes back to the decision in Maxim-Nordenfeldt Gun Co. v Nordenfeldt [I893] I Ch. 630, [I894] A. C. 535. Employers try to deal with this difficulty in different ways, for instance by including stepwise or cascade clauses. They also deploy alternatives such as ‘garden leave’. This probably encourages employers to seek the inclusion of terms in many contracts of employment that are never in fact tested but if they were would be found invalid and unenforceable. There is thus a possible dichotomy between the actual conduct based on perceived law, and what the law actually is. This is a grey area about which empirical research could perhaps tell us more.

Pervasive incentives 32. 33.

265

See Callinan J’s reasons in Maggbury Pty Ltd v Hafele Aust Pty Ltd [2001] HCA 70; 210 CLR 181; 185 ALR 152; 76 ALJR 246 (13 December 2001). One additional way in which this is demonstrated is the abundance of stories about the personal benefits derived from innovation – the hero stories of wealthy inventors. Statistically however these stories may be irrelevant compared with the number of unheardof failures.

REFERENCES Abramowitz, M. (2007), ‘The Uneasy Case for Patent Races over Auctions’, Stanford Law Review, 60, 803–62. Alford, W. (1995), To Steal a Book is an Elegant Offense, Paolo Alto, California: Stanford University Press. Arup, C. , Gahran, P., Howe, J., Johnson, R., Mitchell, R. and A. O’Donnell (eds) (2006), Labour Law and Labour Market Regulation: Essays on the Construction, Constitution and Regulation of Labour Markets and Work Relationships, Sydney: Federation Press. Ayres, I. and G. Parchomovsky (2007), ‘Tradable Patent Rights: A New Approach to Innovation’, University of Pennsylvania, Institue for Law and Economics Research Paper No. 07-23, http://ssrn.com/abstract=1020276. Bessen, J. (2008), Patent Failure, Princeton: Princeton University Press. Duffy, J. (2007), ‘Inventing Invention: A Case Study of Legal Innovation’, Texas Law Review, 86, 1–72. Economides, N. and W. Hebert (2008), ‘Patents and Antitrust: Application to Adjacent Markets’, NET Institute Working Paper No. 07-07, http://ssrn.com/ abstract=1013226 (forthcoming publication in the Journal on Telecommunications and High Technology Law). Grushcow, J. (2004), ‘Measuring Secrecy: A Cost of the Patent System Revealed’, Journal of Legal Studies, 33, 59–84. Jaffe, A. and J. Lerner (2004), Innovation and Its Discontents: How Our Broken Patent System is Endangering Innovation and Progress, and What to Do About It, Princeton: Princeton University Press. Janis, M. (2002), ‘Patent Abolitionism’, Berkeley Technology Law Journal, 17, 899–952. Kingston, W. (2004), ‘Why Patent Harmonization is a Trojan Horse’, European Intellectual Property Review, 59, 441–62. Lichtman, D. and M. Lemley (2007), ‘Rethinking Patent Law’s Presumption of Validity’, Stanford Law Review, 60, 45–72. Macdonald, S. (2003), ‘Bearing the Burden: Small Firms and the Patent System’, Journal of Information, Law and Technology, 2003(1), http://www2.warwick.ac.uk/ fac/soc/law/elj/jilt/2003_1/macdonald. Merges, R. (1999), ‘The Law and Economics of Employee Inventions’, Harvard Journal of Law and Technology, 13, 1–54. Moore, K. (2005), ‘Worthless Patents’, Berkeley Technology Law Journal, 20, 1521–2. Petrusson, U. (2004), Intellectual Property & Entrepreneurship – Creating Wealth in an Intellectual Value Chain, Göteborg: CTHB. Schankerman, M. (1998), ‘How Valuable is Patent Protection?’, Rand Journal of Economics, 29, 77–107. Schankerman, M. and A. Pakes (1986), ‘Estimates of the Value of Patent Rights in

266


the European Countries during the Post-1950 Period’, Economic Journal, 96, 1051–76. Scotchmer, S. (1999), ‘On the Optimality of the Patent Renewal System’, Rand Journal of Economics, 30, 181–96. van Caenegem, W. (2003), ‘Intellectual Property and the Idea of Progress’, Intellectual Property Quarterly, 3, 237–56. van Caenegem, W. (2007), ‘Mobility of Creative Individuals, Trade Secrets and Restraints of Trade’, Murdoch eLaw Journal, 14(2), 265–79. van Ypersele, T. and S. Shavell (2001), ‘Rewards versus Intellectual Property Rights’, Journal of Law and Economics, 44, 525–47.

13. Commodifying sheer talent: perverse developments in the law’s enforcement of restrictive covenants Joellen Riley I.

INTRODUCTION

Earlier chapters in this book have examined ways in which law creates and protects certain intellectual property rights, with a view to proposing law reform that would promote innovation. This chapter shares the concern that law should support (and certainly not suppress) human ingenuity and thereby foster economic growth and development, in the broad interest of public welfare. The focus in this chapter, however, is not traditional ‘intellectual property’ law but the law of contract. More particularly, the concern here is with law relating to restrictive covenants in employment and service contracts. The claim in this chapter is not that the law itself needs to be reformed, but that the practice of law should return to the doctrinal purity of the past. This chapter argues that the law concerning the enforceability of restrictive covenants in employment contracts has slipped into serious error. In Australia at least, the enforcement of restrictive covenants is a matter of private contract law, so disputes are dealt with by the ordinary courts, exercising common law jurisdiction. Restrictive covenants in services and employment contracts are expressly excluded from federal competition laws in Part IV of the Trade Practices Act 1974 (Cth) by s. 51(2). There are only limited opportunities for review of contracts on the basis that they contain ‘unfair terms’ in Australia1 so that, by and large, arguments over the enforcement of restrictive covenants are litigated in the Supreme Courts of the States. It is to those courts that the complaints here are addressed. In particular, this chapter argues that, by adopting property terminology in their reasoning about restrictive covenants, some judges have granted former employers property-like rights to control their former employees’ exercise of talents and skill. An accretion of first instance cases generated by opportunistic and strategic litigation over these covenants has created the risk that the law (or at least, common perceptions of what the law is) will contribute to a sterilisation of creativity, productivity and innovation. 267

268


This is a travesty. After all, one of the orthodox justifications for the recognition and legal enforcement of intellectual property rights is what might be called the ‘incentive’ rationale (van Caenegem 2006, p. 8; McKeough, Stewart and Griffith 2004, pp. 23–6). We promote investment in creative and innovative activities by allowing the creator/innovator to claim and exploit an exclusive property (or property-like) right in the fruits of the creative/innovative endeavour. This chapter will not challenge the incentive rationale in any comprehensive way. The narrower concern here is to expose the perverse practical effects that can arise from extending property-like rights to the fruits of certain skills and talents, especially when it is the employer of the skilled worker who is granted the rights (van Caenegem 2007). After this exposure, the chapter reflects on practical measures which may be adopted to ensure that law and legal practice are not allowed to sterilise innovation.

II.

FOUR COMPLAINTS

The first task of exposing perversity in some current legal practice involves identifying and explaining four complaints common to a number of recent decisions on restrictive covenants (also commonly called ‘non-compete covenants’). The first and most fundamental complaint is that some dangerously flabby concepts of property are emerging because judges are adopting business accounting concepts to expand the concept of property at law. The concept of ‘goodwill’, for example, is useful when accountants want to record on a balance sheet some estimation of the past and potential economic value of an enterprise. However goodwill is not a form of property in any legal sense. Goodwill does not confer upon any legal person a ‘right in rem’, that is, a right enforceable against the whole world, to ensure that the business keeps its customers. It is nothing more than an accountant’s estimation of the value of an aspect of business operations. Goodwill becomes a dangerous concept when it is translated into judicial statements to the effect that a stable workforce is also part of a firm’s goodwill and is therefore an ‘asset’ which can be protected from erosion by injunctions restricting people’s liberty to compete in a labour market. See for example the statement of Brereton J. in Cactus Imaging Pty Ltd v Peters [2006] p. 55, that ‘staff connection constitutes part of the intangible benefits which may give a business value over and above the value of the assets employed in it, and thus comprises part of its goodwill’. The characterisation of ‘staff connection’ (in other words, the fruitful relationships between people) as part of the ‘goodwill’ of an employer’s business can hinder productivity and innovation if it leads to granting a former employer an entitlement to prevent workers

Commodifying sheer talent

269

from taking other employment for a period well beyond the notice period in their contracts of employment. The second complaint is that by framing the rights of employers as property-like rights, these cases are effectively defeating the common law’s natural defence against anti-competitive conduct. By casting employers’ claims as legitimate and vested interests in a form of property rather than merely as bald attempts to enforce promises not to compete, these cases circumvent the common law doctrine making illegal restraints of trade. This ancient doctrine derives from a similar commitment to that expressed in the incentive rationale for recognising intellectual property rights: a liberal society ensures the public interest by allowing and encouraging individuals to exploit their productive talents and energies in their own, and hence the broader public interest. The rationale of this doctrine, as expressed by Lord Atkinson in Herbert Morris Ltd v Saxelby (1916, p. 699), is explained more fully below. Essentially, it is that the sterilisation of productive labour is so antithetical to the public interest that the courts of law serving the public interest will not enforce a private bargain to restrain trade. Under the common law, such a private bargain is void for illegality, no matter how commercially savvy the parties were who made it (see Maggbury v Hafele 2001). Third, some of the worst decisions ignore even the doctrine of privity of contract, which holds that only parties to a contract may be burdened by the contract. A restrictive covenant purports to bind strangers to a contract by restricting the labour market choices of persons who were never party to the service contract being enforced. This occurs whenever customers are told they may not deal with a certain ex-employee; whenever existing staff are told they may not accept job offers from a departing colleague; and indeed, whenever a rival enterprise is told that it is not permitted to recruit the services of the worker bound by the covenant. Fourth, the effect of many of the judicial decisions criticised here is to undermine a fundamental principle, enshrined in the International Labour Organisation’s Declaration of Fundamental Rights, that ‘labour is not a commodity’ (article 1(a)) (ILO 1944; O’Higgins 1997, pp. 225–34). At the core of the reasoning in these cases is the implicit assumption that the value of labour can be extracted from the labourer and treated as an asset on the employer’s balance sheet. In reality, this value cannot be captured and used by the employer without the willing participation of the worker. Employers cannot force cooperation. The most they can do (and only with the support of a complicit legal system) is to sterilise the former employee’s talents for a period, and so protect themselves from free competition for that time. Presently workers in many western industrialised economies are being asked to embrace reduced job security in the interest of more flexible and competitive labour markets. Individual job security is to be sacrificed in the

270


interest of greater overall labour market participation (Collins 1991, p. 229). The trade-off for workers is supposed to be greater ‘career’ security, gained through the promise of improved opportunities to enhance skills, experience and networks in order to navigate successfully from one enterprise to another in a fluid labour market (Stone 2001; 2002; 2004). At a time when this so-called ‘deregulation’ of labour markets is creating pressures on job security, it is perverse for the law to legitimate employers’ tactics to discourage labour market competition by interfering in the mobility of talent from one enterprise to another. The remainder of this chapter interrogates each of these four complaints in turn, before speculating on some strategies to overcome them. In summary, the complaints are that contemporary developments in the law evidence: 1. 2. 3. 4.

A.

inept adoption of accounting language in legal analysis; corruption of the common law doctrine against illegal restraints of trade; ignorance of the privity doctrine which protects the liberty of individuals from the imposition of burdens by others; and departure from respect for a fundamental human right, that is, that people should be free from coercion in respect of their labour. Inept Adoption of Accounting-Speak

A number of Australian decisions have strayed into error because, in the course of reasoning, judges have translated accounting concepts that have been devised for predictive purposes in the discipline of financial business analysis into legal rights that can be marshalled to constrain the freedoms of others. Accounting is descriptive and predictive but law is coercive. Law’s power of coercion must be carefully constrained. It is not to be wielded simply because one party can see a self-interested economic advantage in securing a right against others. There are two kinds of cases that require analysis to unpack this argument. First there are the cases which stretch the notion of ‘confidential information’ so that what in orthodox legal doctrine was purely a personal right to keep one’s own secrets has become a way of claiming an exclusive, proprietary-like dominion over the ideas of others. It has long been held that certain kinds of commercially valuable information can be protected as a kind of property by the enterprise generating that information. This is the protection of the ‘trade secret’ (by the common or judge-made law, outside of statutory regimes for the protection of patented inventions) by way of enforcement of an equitable duty of confidence. The common law, in its wisdom, has never been prepared to treat any idea – no matter how unique and valuable – as a form of property, subject to the exclu-


271

sive control of a private owner. A person who wishes to restrict knowledge of an idea must do so by keeping it secret, and by imposing personal obligations of confidence on any person to whom the secret is disclosed. As the High Court of Australia explained in Breen v Williams 1996, ‘information is not property at all’ (p. 80). Whether the disclosure of information can be restrained depends upon the circumstances in which the information was acquired. ‘If it has been acquired in such circumstances that it would be a breach of confidence to disclose it to another, then courts of equity will restrain the recipient from communicating it to another’ (p. 80). Although confidential information is often referred to as the ‘property’ of the donor of the information, ‘in the end the real truth is that it is not property in any normal sense, but equity will restrain its transmission’ (p. 80). The action for breach of confidence is often mistaken as a proprietary claim because the remedies that flow from breach are similar to proprietary remedies for breach of trust (see also Moorgate Tobacco Co. Ltd v Philip Morris Ltd (No 2) 1984, p. 438). For instance, in cases of breach of confidence, equity may grant an injunction to restrain disclosure of information, or may order that the secret-breaker must account back to the donor for any profit made from the disclosure. Nevertheless, these remedies are personal. They run only against the person who owes the equitable obligation of confidence. They cannot run in rem, against the world. This point is particularly important in the context of the discussion below (see C. Ignorance of privity) of the doctrine of privity of contract. For a long time, the courts have been careful to restrict the expansion of the categories of information that can be protected as genuinely ‘confidential’. In Wright v Gasweld Pty Ltd 1991, p. 334), Kirby P. (as he then was) set out strict criteria for identifying confidential information. It needed to be the fruit of considerable skill and effort; it had to be jealously guarded by the employer; employees needed to be informed that the information was confidential; it should be the kind of information which the industry concerned customarily treated as confidential; and it should be disclosed only to high level employees on a ‘need to know’ basis. Once information became known to the public it could not be confidential (see also Maggbury v Hafele 2001), and any information that was part of the employee’s own stock of knowledge and skill could also not be confidential to the employer (see Faccenda Chicken Ltd v Fowler 1987). In recent times, however, information which is much more mundane has been treated as confidential. In Woolworths v Olson [2004], information about the best way to lay out a supermarket to maximise sales was considered to be ‘confidential’ information. In John Fairfax Publications 2006, strategies for selling media advertising to real estate agents was held to be sufficiently confidential to warrant the grant of an interlocutory injunction preventing a young

272


advertising sales manager from taking up a new job with another publishing house. In earlier times it is arguable that this kind of ‘information’ would be regarded as just part of the industry know-how and creative ideas of the employees themselves. It has now joined the category of ‘confidential’ information largely because employing enterprises have persuaded courts that this kind of information is commercially valuable and constitutes part of their ‘goodwill’. ‘Goodwill’ has been used as a rhetorical device to persuade courts that what is really nothing more than an accountant’s expectation of future value should be treated not only as an accounting ‘asset’, but as a form of legal property. The cases on confidential information demonstrate that the strategic planning skills, the general know-how and creative thinking of working people can now be treated as a resource under the control of the employing business enterprise, not only during employment but afterwards. While the former employer can still (mercifully) not enslave the former employee by requiring continued service, the employer can (with an appropriately drafted covenant in an employment contract) prevent the employee from taking that talent and exploiting it elsewhere for a time. What began in equity as an action to keep one’s own secrets has evolved into a technique for capturing the intellectual capacities of others. More insidious than the confidential information cases is a second category of cases which have held that the personal charisma of a managerial employee who is able to hold together an effective team of workers has likewise become the property of the employer, so that the employer can restrain the manager from offering alternative employment to the team members. Restraints giving effect to this type of claim have come to be known as ‘anti-poaching’ clauses. This term is itself an illustration of the invidious way that concepts of property ownership have slipped into this field. ‘Poaching’ connotes the thief who illegitimately hunts on the land of the lord of the manor, stealing the lord’s livestock. Staff ought never to be treated as livestock. This body of case law owes its existence to reasoning by analogy from the cases about ‘customer connection’ (Heydon 1999, pp. 91–101). Former employers have been held to be entitled to enforce covenants whereby departing employees agree not to solicit customers or clients, on the basis that the employer has a legitimate interest in the personal bond and loyalty that customers may develop to the employer’s staff (see Lindner v Murdoch’s Garage 1950, pp. 633–4). The employer can lay claim to that ‘customer connection’ because the employer has funded the good relationship. The employer has provided the resources and the back-up to allow the ‘front-line’ employee to develop the relationship with the customer. If the employee moves to another enterprise, there is a period of time during which the customer may be vulnerable to approaches from the familiar former employee with whom they have developed the relationship.


273

An employee who has covenanted not to take advantage of this customer connection can be held to that promise, within limits. The limits are that the former employee can only be restrained from dealing with customers with whom they had personal dealings, and not with all of the customers of the former employer. Also, the restraint will be limited in time, to a period sufficient to allow the employer time to establish new relationships between the customer and continuing staff. It is important, however, that this kind of protection of customer connection is not mistaken as legitimation of a claim that the employer ‘owns’ the customer (Riley 2003). It is nonsense to speak of customers as ‘assets’ of a business. Customers are autonomous actors, free to deal with whomever they choose. Accountants are accustomed to describing customers’ propensity to offer repeat business to a firm as ‘goodwill’ of the firm, but this term describes nothing more than an estimated value of an expectancy. The customers are, and should be, free to deal with whomever they wish. The only rationale for enforcing a non-solicitation covenant in respect of clients is to prevent the former employee from exploiting the customer’s temporary vulnerability to the pull of personal loyalty. It is the personal obligation of the former employee not to exploit that vulnerability that is enforced, not some property right in customers. Nevertheless, this accounting language has infiltrated legal reasoning with the result that goodwill is equated with a property right. For example, in Koops Martin v Dean Reeves 2006 p. 29, Brereton J. in the New South Wales Supreme Court stated: ‘Goodwill, of which customer connection is the essence, is akin to a proprietary interest.’ Later he stated that ‘customer connection belonged [my emphasis] to the employer’ (p. 52). In Hartleys Ltd v Martin 2002, the court described former employees as having taken ‘part of the plaintiff’s assets with them in the form of clients’ (p. 116). Cases which have held that an employer can enforce covenants not to ‘poach’ staff rely on an analogy with the customer connection cases. The close personal loyalty of the customer connection is likened to a similar attachment between managers and their teams. Some cases have held that this ‘team glue’ (Riley 2005, p. 18) contributes to maintaining a ‘stable workforce’. A stable trained workforce is valuable, hence the employer has a legitimate and protectable interest in preventing departing staff from disturbing that stability by offering jobs to their former colleagues. English authority on this point has been divided. Office Angels Ltd v Rainer-Thomas 1991; Dawnay, Day & Co. v De Braconier 1998; TSC Europe v Massey 1999 and Alliance Paper Group plc v Prestwich 1996 have allowed the existence of such an interest (see Brooks 2001, pp. 42–9). The English Court of Appeal in Hanover Insurance Brokers Ltd v Schapiro 1994, however, stated most robustly that the fact that a firm depends upon its staff ‘does not make the staff an asset of the company like apples or pears or other stock in trade’ (p. 16). Australian case law has

274


tended to follow Dawnay, Day & Co. Ltd (see, for example, Aussie Home Loans v X inc Services 2005; Kearney Australia Pty Ltd v Crepaldi 2006, p. 58; and Cactus Imaging v Peters 2006). The trouble with this analogous reasoning is that it has taken a quantum leap. The close relationships between staff are not created by the employer. They are due to the entirely personal charisma and relationship skills of the workers themselves. The employer has been fortunate to have employed them until now, but can lay no claim to perpetual mastery of those personal attributes. So how has this reasoning developed? Again, the inept engagement of accounting-speak is to blame. The following extract from Cactus Imaging Pty Ltd v Peters 2006 illustrates the incremental reasoning which moves from appreciating the economic value of goodwill to asserting a property-like interest of an employer to restrain the movement of staff: [E]mployees are not property, but, all else being equal, a business with a stable trained workforce will be more attractive to a purchaser and command a higher price than one with a workforce which is unstable, disruptive or poorly trained, just as a loyal and satisfied clientele makes a business more attractive and valuable. In my opinion, staff connection constitutes part of the intangible benefits which may give a business value over and above the value of the assets employed in it, and thus comprises part of its goodwill. It is amenable to protection by a covenant in a manner similar to customer connection, even in the absence of protectable confidence. (p. 55)

Everything the judge says about the economic value of a stable workforce is entirely true. From a financial business analysis point of view, a business with good staff relations is likely to be more successful. As Jenkins LJ said in an older English authority on these anti-poaching clauses, Kores Manufacturing Co. Ltd v Kolok Manufacturing Co. Ltd 1958, the value of a stable workforce is ‘an interest which employers are entitled to protect by all legitimate means, as by paying good wages and making their employment attractive’ (p. 74). But it is not legitimate to hold that because good staff relations make good business sense, an employer should be granted a legal right to prevent those staff from exercising an entirely free choice of where they will work. The law is not obliged to grant and enforce rights just because businesses would make more money if they had those rights. In summary, the problem in each of these kinds of cases is that the accounting notion of goodwill is used to justify a legal conception of property. In accounting terminology, ‘goodwill’ is a useful way of identifying the economic value that a business entity is likely to derive in the future, from the fact that it has expended resources on things like research and development, marketing, advertising and customer service. Since future income flows can have a present value it makes sense that estimations of the goodwill in a busi-


275

ness should appear on a balance sheet. But this does not mean that the business has a legal property right in any of the components making up the goodwill. The propensity of customers to provide repeat business, for example, is not a property right in a legal sense. No one can force the customers to keep coming. No one can stop the customers from taking their business elsewhere. The value of the goodwill depends on what the law would describe as a ‘mere expectancy’ – a hope that an event will happen, but a hope unsupported by any legal entitlement to force that result. A legal property right is very different. A legal property right is a right to claim a certain benefit, to claim it against the whole world, and to have the support of the machinery of the State’s legal system in enforcing that claim. At this point, some readers may raise an objection. The cases described above all concerned the enforcement of contractual promises made by former employees. Why does it matter whether the employer’s interest was proprietary or not, when the cases ultimately concerned the enforcement of a personal obligation under a contract? Did the employee in each case not voluntarily surrender their liberty? It matters because such contracts ought not to be enforceable, if they purport to enforce an illegal restraint on trade. This brings us to the second complaint about the development of case law in this area. B.

Corruption of the Doctrine Making Illegal Restraints of Trade

The reason that it is dangerous to recognise that an employer has a ‘legitimate interest’, that is, a legally recognised and supported entitlement, in a stable workforce is that recognition of such an interest undermines an important and very ancient principle of the common law, making illegal restraints of trade (Blake 1960; Heydon 1999). This doctrine is underpinned by faith in the view that the public interest is best served by supporting individuals’ freedom to engage in productive activities. So in Herbert Morris Ltd v Saxelby 1916, Lord Atkinson said: ‘The general public suffers with [the worker who is restrained], for it is in the public interest that a man should be free to exercise his skill and experience to the best advantage for the benefit of himself and all those who desire to employ him’ (p. 699). The doctrine works by ensuring that courts refuse to enforce private contracts purporting to restrict those freedoms. This is an area where the public interest (justified on the basis of sound economic policy) is held to trump private arrangements. We must remember that a contract is nothing more than an agreement enforceable at law. Contracting parties have very little power to enforce their own agreements without the support of law. Nevertheless, if it is not in society’s interests to support certain private bargains, then society is under no obligation to provide the legal machinery to enforce those contracts.

276


It seems that courts are forgetting this fundamental proposition. The doctrine against restraints is a doctrine asserting that courts must refuse to enforce contracts – even voluntarily entered agreements, made by welladvised parties – if those contracts purport to restrain trade and so defeat the public interest in unhindered economic activity. Too often judicial decisions now contain statements implying that the parties’ own agreement is sufficient to justify the enforcement of a restraint. For example, in John Fairfax v Birt 2006, Brereton J upheld a restraint keeping a young man with a dependent wife and children out of work for three months (in circumstances where hardship was pleaded in the case) on the basis that the young man was the ‘author of his own misfortune’ (p. 95). The point of the restraint of trade doctrine is that it applies notwithstanding private agreement, and we don’t have to find some separate ground (such as unconscionable dealing or inadequate understanding) to set aside the contract. We do not even have to find hardship, although individual hardship is another justification for the existence of the doctrine making illegal restraints of trade. The exception to the doctrine making restraints illegal is the proposition that a restraint may be enforced if it goes no further than is reasonably necessary to protect a legitimate interest of the employer. So long as ‘legitimate interests’ continue to be confined to those proprietary rights that employers themselves already hold, so that the contractual restraint works to prevent a loss of that existing entitlement, then the exception is sound. The problem has arisen because of the practices described above. Some courts have begun to accept that employers have legitimate interests in the skills, ideas, talents and personal charisma of their staff. These things are not property and, even if they were, they should not be taken as belonging to the employer. Even the protection of ‘confidential information’ has gone too far. Too many ephemeral things, such as ‘commercial strategies’ and ‘marketing plans’ generated by the departing employees themselves, have been protected as the confidential information of the employer. These things have been held to be too vague to capture in any explicit injunction restraining their use in future employment, so instead (on the basis of the argument in Littlewoods Organisation Ltd v Harris 1977, p. 1479, that it is more convenient to just stop an employee from competing than to try to enforce an injunction preventing the use of particular information), courts have been willing to stop the employee from working at all for the duration of the restraint. It seems counter-intuitive that information can be so vague but still constitute the kind of information that the law will protect as confidential according to the principles laid out in Wright v Gasweld Pty Ltd 1991, p. 334. In any event, it seems deeply unreasonable that the employee should bear such an extreme cost (that is, the inability to work in their chosen field for a significant period of time) in the interests of protecting such vague and ephemeral information. In


277

Shakespeare’s Merchant of Venice, Shylock was forbidden by the court from taking the pound of flesh for which he had contracted because to do so would rob his victim of vital blood, which was not the subject of the bargain. Our present law turns this on its head. The employee is made to surrender up their life-blood – the very liberty to work and use their own know-how, skills and talents to the full – in order to render to the employer a mere pound of socalled ‘confidential information’. This is particularly unreasonable when the employer’s claim is not to genuinely confidential information but to certain attributes of human talent, such as the employee’s ingenuity and personal charisma. And it is doubly so, when the employer’s claim affects not only the employee giving the original covenant but others who were strangers to the initial bargain. This brings us to the third complaint. C.

Ignorance of Privity

Whenever an anti-poaching covenant is enforced (whether it concerns customers or colleagues), it affects not only the person who signed the original employment contract but also all the people with whom they are now barred from doing business. Some cases have given a very loose definition to the notion of solicitation, so that it captures circumstances where the recruit makes the initial contact with the person restrained (see, for example, Ecco Personnel Pty Ltd v Barrett 1998). This means that the enforcement of the anti-poaching covenant limits the freedom of the people who would like to move with the departing manager. They were not party to the agreement, so why should their liberties be restricted in any way by that agreement? To allow such a result is to ignore the doctrine of privity. Undergraduate law courses perhaps focus too much on the aspect of the privity doctrine that precludes non-parties from enforcing the terms of a contract, even when the contract purported to confer a benefit upon that nonparty. We often gloss over the other face of privity – the aspect that ensures we cannot have a burden thrust upon us without our own agreement. We cannot be made the victims of the bargains of others. This seems to have been completely forgotten in some of the case law enforcing anti-poaching clauses. The reports of these cases, often heard as urgent interim injunction applications, rarely record any evidence being taken from the workers who are to be prevented from accepting employment with the departing contract-breaker. See for example Hartleys Ltd v Martin 2002, where a personal assistant, Ms Briggs, was prevented from taking up a position with the two stockbrokers whom she had assisted when they left to work with a rival firm. Ms Briggs had already given one month’s notice to her employer, but was effectively

278


prevented from taking up her new employment when the Martin brothers were injuncted from offering her employment. The report of this decision is completely silent about what Ms Briggs was required to do for the three months, in which she was prevented from taking up the new job offer. There is no record of her being asked to give evidence in the case. Was she obliged to suffer three months’ unemployment? Was she effectively forced to return to her post at Hartley’s? Either result is contrary to orthodox doctrine. No injunction ought to have been given which would force her to perform a contract against her will (Lumley v Wagner 1852; Meagher, Heydon and Leeming 2002, pp. 755–6). Neither should her ability to earn a living have been sterilised (even if it was her own covenant being enforced): see Page One Records Ltd v Britton 1967. What, too, of the interests of the rival broking firm? Why should its ability to engage staff be limited by an arrangement made between others? Kores Manufacturing v Kolok Manufacturing 1958, p. 54, refused to enforce a voluntary agreement not to recruit from the members of another’s staff. How then can more recent cases enforce such restrictions when they were not even voluntarily assumed? See, for example, IceTV v Duncan Ross and Ors 2007, which tells the story of two innovators in the media technology field who were exiled from this fast-moving marketplace on the basis of a former employer’s claim that they should not be allowed to work for Mobilesoft, a company which had been a ‘prospective’ (but not an actual) client of the former employer (Riley 2008). The employees were prevented from accepting an offer of employment initiated by Mobilesoft. On what basis was it at all reasonable to restrict Mobilesoft’s options for recruitment, when it had never even become a client of the plaintiff firm? Injunctions given in these circumstances clearly stifle free competition in labour markets, to the disadvantage not only of the workers affected but of the market as a whole. This is a particularly obnoxious result when parties’ choices are limited by an order to enforce an agreement to which they were complete strangers. This concern leads us to the fourth and final complaint. The privity doctrine is a private law doctrine which, in many respects, supports a fundamental human right to liberty. D.

Human Rights Dimension

These contemporary developments in the enforcement of restrictive covenants infringe a fundamental principle of international labour law that labour is not to be commodified. There are many meanings to this principle. At the most basic level, people are not to be enslaved and sold, but the principle is more subtle than that. People’s work is an intrinsic part of their identity. Work is not to be treated as if it is just like any other saleable commodity, to be priced and


279

traded according to the dictates of bloodless economic rationalism. Any legal development that countenances the separation of labour from the labourer is to be resisted because it leads down a very slippery slope towards the alienation of human beings from their own work. We must remember that the result of many of these cases is to take workers out of the labour market for a period. If there is any meaningful ‘right to work’, it must at least sound in an entitlement to pursue one’s chosen occupation by exploiting one’s own skills and talents. Certainly, a restraint is legitimate if the worker purports to steal the fruits of others’ labours – but it should not be legitimate to restrain their engagement in productive work from harnessing their own unique talents and their general know-how. Our legal imagination has taken us too far and into dangerous territory. Intellectual property law has been successful in capturing the fruits of labour and converting them into ownable, tradeable commodities. Examples are copyright works, patentable inventions and trademarks. Through the progress of laws protecting what we call intellectual property rights we have learned to regard the products of certain kinds of creative labour as property. By making these things proprietary, we permit owners to exclude other users from their benefits. Commodifying the fruits of labour may be justifiable, but commodifying labour itself is a dangerous incursion into human liberty. The contemporary practice of law in enforcing certain restrictive covenants in employment contracts has stepped into that perilous territory.

III.

PRACTICAL MEASURES

It is not enough to complain of the weaknesses of our laws and the ways in which they have developed and are enforced. This book has promised to deliver useful ideas for law reform. Fortunately, in the field of the enforcement of restrictive covenants, there are some relatively straightforward reforms which may quickly address the complaints outlined above. First, we should recognise that many of the decisions in this field are made at an interlocutory stage. Employers seek urgent injunctions, on the basis that the time necessary to prepare for a full hearing will exceed the period of the restraint. If the employee is not restrained in the meantime, the employer will be left only to a claim for damages because the period of the restraint will have expired by the time of hearing. In any other kind of case, the plaintiff would have to establish that damages were an inadequate remedy to succeed in claiming an injunction on this basis. However, cases about the enforcement of negative covenants are treated differently on the basis of the doctrine in Lumley v Wagner 1852 (see Spry 1990, pp. 561ff.). (This was the case of the opera singer who was held to a promise not to sing in rival establishments for

280


the duration of her contract.) On this basis courts are persuaded to grant injunctions without a full hearing of the merits of the case, so long as there is a serious question to be tried and the balance of convenience favours granting the injunction. While this balance of convenience is weighted in favour of the employer seeking the injunction, there is a serious risk of injustice to the worker, and sterilisation of innovation and productivity. If the injunction is granted, there is rarely any call for a final hearing. The interim injunction will have achieved the employer’s objective of keeping the worker out of the marketplace for a sufficient period to ruin the worker’s prospects of offering any effective competition. In IceTV v Duncan Ross and Ors 2007 (described above), an injunction was granted to prevent an inventor and a business manager from taking up employment offers in the media technology industry, notwithstanding that the former employer had itself dismissed the two innovators. It was in financial straits and was not in a position to engage their services any longer. The sheer injustice of enforcing a restraint when the employer had no use for the labour itself is breathtaking. A year later, these men were still unemployed. It is unlikely that this result could have occurred if the courts in New South Wales were to adopt the approach taken by the Supreme Court of New York in Jacobi Tool & Die v Mondi and Ors 2007: A party seeking the drastic remedy that a preliminary injunction confers must establish a clear legal right to that relief under the law and upon undisputed facts set forth in the record [citations omitted]. To prevail on a motion for preliminary injunction relief, the movant must clearly demonstrate a likelihood of success on the merits, the prospect of irreparable harm or injury if the relief is withheld and that a balance of the equities favours the movant’s position. (p. 3)

Unfortunately, the contrary approach has been taken in New South Wales in recent times. As an extreme example, take the statement of Brereton J in the Supreme Court of New South Wales in Otis Elevator Co. Pty Ltd v Nolan 2007: I am of the view that the mere fact that the injury to the plaintiff is slight or nonexistent is insufficient to justify declining an injunction on discretionary grounds; so also is the mere fact that enforcement of the injunction would occasion considerable hardship to the defendant. (p. 30)

Otis Elevator Co. involved a claim that a brand new recruit, who had attended no more than an initial orientation seminar given by the employer, should be prevented from changing his mind promptly and taking up an alternative offer of employment in the industry. Fortunately, Mr Nolan succeeded in demonstrating to the court that he had acquired absolutely no valuable information about the employer’s operations at this training session. Otherwise, the harsh


281

approach taken by the judge may have rendered him unemployed for six months following this abortive job-start with Otis. Statements such as the one cited above (from Otis) embolden employers in seeking to enforce restraints. According to this formula, they do not have to demonstrate any injury at all from breach of the covenant. How can an interest be sufficiently ‘legitimate’ to warrant the protection of the exception to the doctrine of illegal restraint of trade, if the plaintiff cannot demonstrate that the interest has a value? It would be far better if Australian courts adopted the New York approach in Jacobi.2 If an employer knew that to succeed in obtaining an injunction it must prove serious and irreparable harm, fewer of these claims may be brought. Presently, the balance is tipped in favour of the employer plaintiff, who needs only to demonstrate the existence of a restrictive covenant, and steps taken by the departing employee that may breach it, to succeed in thwarting the employee’s liberty to work in the industry. A second practical step for reducing the harm caused by these developments would be to reinvigorate proper understanding of the common law doctrine making illegal restraints of trade. Employers should be required to demonstrate that an employee is taking away something more than the employee’s own accumulated skill, experience and personal attributes before the employer is entitled to establish a ‘legitimate’ interest sufficient to support enforcement of a restrictive covenant. The law should return to the concept of a genuine ‘trade secret’, and resist the present trend towards allowing any ephemeral business plan or marketing strategy to constitute sufficiently ‘confidential’ information to justify restrictions on labour market movements. Judges in particular should be wary of arguments purporting to bootstrap accounting concepts, like ‘goodwill’, into legal property claims.

IV.

CONCLUSION

There will be critics of these views, who would prefer to focus on the employing business’ side of this competition of interests. They will argue that businesses will not invest resources in marketing strategies, customer relationships and personnel development if they cannot protect that investment from competition. This is a spurious argument. If no-one has the benefit of legal rights preventing the departure of their best staff, then everyone will have to continue to invest more to keep ahead of the game. Silicon Valley developed a software industry without the benefit of copyright protection in computer programs, and in a labour market where employment was ‘at will’. Business and industry can thrive without legal protection from competition. Some may also complain that this argument strains at gnats. The enforcement of these contractual restraints only affects seriously high flyers, doesn’t

282


it? Not any more. These kinds of clauses are appearing in the contracts of salespeople and a whole range of knowledge workers. This is not just a matter concerning Chief Executive Officers. It has the potential to seriously impact markets for skilled labour. It may also be argued that these restraints do little harm because they are enforced for a limited period, not forever. Even three months is too long to keep a person out of productive work, especially in a labour market with skill shortages. However, we are seeing longer and longer periods being enforced. It is not uncommon now to see restraints claimed for six months or a whole year. The risk is that a long restraint may dampen the enthusiasm of a new employer to make the offer of a job. An employee faced with the prospect of remaining out of the labour market in his or her field of expertise for many months may behave more conservatively in investigating more productive uses of their talents. It is also unfair that a practice of demanding long restraints in employment contracts, and in vigorously pursuing them, has developed in an environment of greater job insecurity for employees. We must also remember that some of these restraints have been enforced in circumstances where the former employer terminated the employment (see IceTV v Duncan Ross 2007). So for reasons of market efficiency, and of fairness to employees, the law in this field needs careful revision, and a return to the more robust pro-competition doctrines of the past. The optimisation of creative endeavour requires a freely competitive labour market (van Caenegem 2007, p. 278). Laws which restrain the mobility of workers also restrict creativity and innovation.

NOTES 1.

The New South Wales Industrial Relations Commission exercises a jurisdiction to review contracts under which work is performed in industry in NSW (see Industrial Relations Act 1996 (NSW) ss 105–6). However, the operation of this jurisdiction was severely curtailed by the Workplace Relations Amendment (Work Choices) Act 2005 (Cth) and Independent Contracts Act 2006 (Cth), which overrode the jurisdiction of State tribunals over any workplace relations matters. Before the enactment of these statutes, some cases involving restrictive covenants were heard by the Commission, and resolved according to a discretion to vary contracts. See, for example, Haddad v S & T Income Tax Aid Specialists Pty Ltd (1985) 13 IR 16. 2. For a general explanation of the United States’ approach to these kinds of cases, see Hogan (2006).

REFERENCES Blake, Harlan M. (1960), ‘Employee Agreements Not to Compete’, Harvard Law Review, 73, 625–91.


283

Brooks, Adrian (2001), ‘The Limits of Competition: Restraint of Trade in the Context of Employment Contracts’, University of New South Wales Law Journal, 2001, 27. Collins, Hugh (1991), ‘The Meaning of Job Security’, Industrial Law Journal, 20(4), 227–39. Heydon, J.D. (1999), The Restraint of Trade Doctrine, second edition, Sydney: Butterworths. Hogan, Thomas M. (2006), ‘Uncertainty in the Employment Context: Which Types of Restrictive Covenants are Enforceable?’, St John’s Law Review, 80, 429–54. ILO (1944), International Labour Organisation Declaration of Philadelphia. McKeough, J., A. Stewart and P. Griffith (2004), Intellectual Property in Australia, third edition, Sydney: Butterworths. Meagher, R., D. Heydon and M. Leeming (2002), Meagher Gummow & Lehane’s Equity Doctrines and Remedies, fourth edition, Sydney: Lexisnexis Butterworths. O’Higgins, Paul (1997), ‘“Labour is Not a Commodity” – an Irish Contribution to International Labour Law’, Industrial Law Journal, 26(3), 225–34. Riley, Joellen (2003), ‘Who Owns the Customers? A Reflection on Recent Cases on Post Employment Restraint Clauses’, Commercial Law Quarterly, 17(3), 3–11. Riley, Joellen (2005), ‘Who Owns Human Capital? A Critical Appraisal of Legal Techniques for Capturing the Value of Work’, Australian Journal of Labour Law, 18(1), 1–25. Riley, Joellen (2008), ‘Innovation Put on Ice? How Overly Jealous Intellectual Property Protection Discourages Creativity and Productivity’, Australian Intellectual Property Law Bulletin, 20(7), 102–4. Spry, I.C.F. (1990), The Principles of Equitable Remedies, fourth edition, Sydney, Australia: Law Books Co. Stone, Katherine V.W. (2001), ‘The New Psychological Contract: Implications of the Changing Workplace for Labor and Employment Law’, UCLA Law Review, 48, 519–659. Stone, Katherine V.W. (2002), ‘Knowledge at Work: Disputes over the Ownership of Human Capital in the Changing Workplace’, Connecticut Law Review, 34, 721–64. Stone, Katherine V.W. (2004), From Widgets to Digits: Employment Regulation for the Changing Workplace, Cambridge, UK: Cambridge University Press, ch. 7. van Caenegem, W. (2006), Intellectual Property, second edition, Sydney: Butterworths. van Caenegem, W. (2007), ‘The Mobility of Creative Individuals, Trade Secrets and Restraints of Trade’, Murdoch University E Law Journal, 14(2), 265–79.

CASES Alliance Paper Group plc v Prestwich [1996] IRLR 25. Aussie Home Loans v X inc Services [2005] NSWSC 285. Breen v Williams (1996) 186 CLR 71. Cactus Imaging Pty Ltd v Peters [2006] NSWSC 717. Dawnay, Day & Co. Ltd v De Braconier d’Alphen [1998] ICR 1068. Ecco Personnel Pty Ltd v Barrett [1998] NSWSC 545. Faccenda Chicken Ltd v Fowler [1987] Ch 117. Haddad v S & T Income Tax Aid Specialists Pty Ltd (1985) 13 IR 16. Hanover Insurance Brokers Ltd v Schapiro [1994] IRLR 82.

284


Hartleys Ltd v Martin [2002] VSC 301. Herbert Morris Ltd v Saxelby [1916] AC 688. IceTV v Duncan Ross and Ors [2007] NSWSC 635. Jacobi Tool & Die Manufacturing Inc. and Jacobi v Mondi and Ors 2007 WL 3325854 (NY Supp). John Fairfax Publications Pty Ltd v Birt [2006] NSWSC 995. Kearney Australia Pty Ltd v Crepaldi [2006] NSWSC 23. Koops Martin v Dean Reeves [2006] NSWSC 449. Kores Manufacturing Co. Ltd v Kolok Manufacturing Co. Ltd [1958] 2 All ER 65. Lindner v Murdoch’s Garage (1950) 83 CLR 628. Littlewoods Organisation Ltd v Harris [1977] 1 WLR 1,472. Lumley v Wagner (1852) 42 ER 687. Maggbury Pty Ltd v Hafele Australia Pty Ltd (2001) 210 CLR 181. Moorgate Tobacco Co. Ltd v Philip Morris Ltd (No 2) (1984) 156 CLR 414. Office Angels Ltd v Rainer-Thomas [1991] IRLR 214. Otis Elevator Co. Pty Ltd v Nolan [2007] NSWSC 593. Page One Records Ltd v Britton [1967] 2 All ER 822; [1968] 1 WLR 157. TSC Europe (UK) Ltd v Massey [1999] IRLR 22. Woolworths Ltd v Olson [2004] NSWCA 372. Wright v Gasweld Pty Ltd (1991) 22 NSWLR 317.

14. Split entitlements? Intellectual property policy for clusters and networks Christopher Arup I.

INTRODUCTION

Between the employer and the employee, or the principal and the contractor, the law largely leaves the disposition of intellectual property use rights and use of information to freedom of contract. Innovation policy has focused on the incentives the firm receives to innovate. If the position of the inventive or creative worker is considered, the focus is on the incentives that are provided within the firm. This chapter engages with cluster theory to argue that a proinnovation policy should ensure that the worker is free to use intellectual property with new firms. The focus is on mobility and the circulation of knowledge. The original firm should only hold the rights for the purposes for which it directed or commissioned work. Non-disclosure and non-competition clauses should be invalid, except for protection of a carefully circumscribed category of trade secrets. This policy would require that freedom of contract was checked. Binding legislative provisions would be needed, not merely default rules; the courts could not be expected to further this policy through the common law.

II. A.

POLICY CONTEXT Cluster Theory

The best intellectual property focuses on the context in which rights apply. Crucial to the impact of policy is interaction with the process of innovation. Characterised as an instrumental approach, it has to be grounded. Yet there is still a tendency for policy making to proceed from basic principles. Commonly, the view is that firms will not invest in the production, application and diffusion of knowledge without the security of property rights. The policy may include an assumption about the kind of firms that can do so, for the 285

286


modelling runs beyond appropriation to the allocation and distribution of rights between participants in the field – employers and employees, commissioners and consultants, for instance. Conventionally, the view is that large private firms should have the benefit of intellectual property endowments or the freedom of contract to assume them from others. This policy signifies the era of corporate capitalism (Cornish and Llewelyn 2003; Noble 1977), just as an earlier period of liberal capitalism championed individual inventors and authors. Now might be a good time to revisit the assumptions when new technologies, work practices, industrial organisations, business models and social mores seem to be outmoding the conditions that might properly have informed the old policy (Suthersanen 2006). It might even be argued that these changes are undermining the conditions that the old policy itself helped to bring about, such as the notion of ‘copyright industries’ implies, for instance (Towse 2001). Hence the strong interest at the moment in the phenomenon of amateur peer to peer production. Production in the traditional industrial sectors seems to be changing too. The changes stem from management and employment practices such as the externalisation of production functions and the disintegration of the large, hierarchical firm. In its most extended form, the ‘firm’ becomes largely a coordination centre; even design functions are outsourced to an army of casual researchers, consultants, freelance agencies and small enterprises. Short-term and single-task engagements are the lot of many young writers in the media industries (McRobbie 2002); post-doctoral researchers in the biomedical sciences too (Gaughan and Robin 2004). Such workers can be engaged on several projects concurrently or at least on a quick succession of projects with different organisations, leading to a blurring of the boundaries between firms. However, these arrangements do not dissipate entirely into a series of arm’s-length exchanges, each a simple exchange of wages for work. Between hierarchies and markets exists the hybrid relationships created within clusters and networks (Thompson 2003; Tsang 2006). Such clusters are mixes of large established firms, small and medium-sized enterprises (SMEs) including startups, universities, research institutes and clinics, government instrumentalities, publishing and broadcasting houses, professional associations, business and financial services, banks, and client and customer operations. The common element is the circulation of people, so that Hyde (2003) suggests that labour markets are markets in information. The disintegration of the large firm can have harmful social effects. Labour law is concerned about the impact of job insecurity and intermittency on people’s ability to plan and safeguard their lives against risk. However, government policy has recently been aiming to make a virtue of these developments. The thinking is that small countries and regional locations (as often

Split entitlements?

287

the policy is geographically based) can become innovation hubs or creativity nodes – if they get the policy settings right to foster these clusters. This approach stresses: (a) the social nature of knowledge production; (b) the importance of tacit knowledge, skills and sensibilities and other complementary assets as well as codified knowledge; (c) the value of transfers, spillovers and unexpected applications; and (d) the incremental, experimental, interactive, open-ended process of innovation (Powell 2001; Flew 2005). Benkler (2006) shows how people produce and disseminate knowledge in different ways. They do so non-commercially and this mode of innovation is enjoying resurgence now, not only through traditional practices such as publication but new practices such as informal professional clubs and open source research bases. Within commercial approaches, the motivation might be to trade on rights, but knowledge is also ‘given away’ in order to circulate a product quickly and widely or to attract patronage to a service being offered. The work of academics varies like this. While they work within large ‘firms’, increasingly they move between them and into industry organisations. At any one time, they are forging links beyond the boundaries of the firm with colleagues, sponsors, partners, clients and publishers. Academics compete but they also combine in groups and cooperate. It remains debatable how crucial clusters are to a strategy to become innovative and creative (De La Mothe and Mallory 2006). They do not take hold everywhere and, where they do, on the ground, they take different forms, variously mixed with hierarchies and markets (Casper 2007). On some conceptualisations, clusters can assume an international or even virtual form, but here my focus is on clusters that are spatially or geographically located within a conurbation or region. Such clusters have attractions for a small economy like Australia that has few large firms engaged in research and development locally, which relies instead on public sector institutes, an educated workforce and some SMEs for incremental innovations and local adaptations. But there always seems to come a point, at least with any major development, when intellectual property is handed to a large (foreign) company (Berry 2005). Is a more realistic strategy for Australian workers and SMEs to depend on those firms?1 The policy need not be one or the other. Clusters could be productive in certain sectors or at certain stages of innovation. The key is the freedom to move around. So, for instance, the experience identifies an early, risky stage, at which the large foreign firms are not willing to get involved, even with innovations that could be commercially large scale in the end. This lesson can be taken further: the new modes of production (such as low infrastructure barriers to entry, flexible specialisation in production, labour mobility, inter-firm collaboration and product customisation) mean that success does not have to depend on economies of scale or scope and centralised distribution channels.

288


Today, the production of medicines is being customised to different personal profiles, not just media entertainments.2 B.

Cluster Policy

What sort of intellectual policy would be conducive to clusters? The large firm policy has the virtue of concentrating property rights. Cluster policy is likely to require a more discriminating approach, which means split entitlements, yet something stricter than freedom of contract because that also leads to concentration. That policy would begin with the treatment of codified knowledge, the subject of the intellectual property categories of patents, copyright and trade secrets. Codification depends primarily on technical and social processes, the ability for example to capture in computer programs the artistry of craft work or in shop manuals the psychology of customer relations. Still, law contributes to codification; it is worth thinking of the kinds of methods and techniques which recently have been ruled patentable. Too much codification can be detrimental to innovation (Roberts 2001). A pro-innovation policy might limit the march of appropriation. For instance, it might: (a) keep certain resources from appropriation altogether (for example, basic ideas); (b) restrict the scope of individual grants (for example, to specific industrial applications, modes of expression or items of information); (c) limit the uses that the property owner may control; or (d) grant access rights through statutory licences, fair uses (for example, for research and experimentation), public interest exceptions and the remedy of anti-competitive practices. A very real issue for all researchers and creators is freedom to operate with basic information resources, the fundamental methods and building blocks for new products and services. Even large firms are dualistic these days about control. Depending on where they are in the production cycle, sometimes they wish to capture the control of the resource upstream; other times they want it to be available for downstream uses (Pisano 2006). Yet rights exist, so where knowledge is made the subject of intellectual property, it is necessary for the policy to regulate the disposition of control. The focus shifts to the transactional field, that is, how the parties use contracts to allocate rights. Conventionally, the policy leaves the participants free to determine the allocation themselves. The law (legislature or courts) provides default rules for some situations but they are in their nature not binding or immutable. They are ultimately a matter of choice, not the parameters within which the participants must operate. Contract is also crucial where information cannot be codified and control depends on a relational form – the control placed on people where tacit knowledge resides. Contracts may describe the kind of information that must

Split entitlements?

289

not be disclosed to others or that may not be used in competition with the firm. Default rules describe these kinds of obligations as well, but contract may extend the control, particularly from the time when the worker leaves the firm. Looking at the field this way, we rate intellectual property, in conjunction with contract, as an instrument for the coordination and governance of innovation in clusters (for example, Brousseau and Bessy 2006). Pro-cluster, the concern is that these information resources do not get locked up, that they are leveraged to their maximum productive potential (Cunningham 2006). The current policy favours the large firms. There is an argument for this setting, as we shall see, but it is also important to ensure that the firms cannot assert ambit claims and blanket reservations. For instance, the Australian Government’s Creative Cluster Study suggests that a large proportion of copyright content is controlled but is not in active commercialisation or use. Often it is bundled – unnecessarily or inappropriately – with the matter of protection of corporate or commercial information (Cutler & Company 2003). Likewise, recent audits of government agencies reveal that they are clawing back intellectual property from employees and consultants but not exploiting it (for example, Office of the Victorian Auditor-General 2005). Likewise, employers are pursuing non-disclosure and non-competition clauses more avidly. The focus of the policy is on control over secondary or future uses of codified or tacit knowledge. The principle is that the firm should only have control for the purposes for which it directed or commissioned the work. Otherwise, where workers want to transfer to other firms or operate from their own business, they should be free to put the knowledge to further and separate uses. When entitlements are split like this, the firm is still given incentives, but the worker is also regarded as an innovator and the policy takes into consideration the contribution to innovation overall outside the relationship. How is the policy of split entitlements represented? Default rules do apportion the rights but they give way (frequently) to assignments under contract. Instead, the rules (as far as they go) need to be non-negotiable if the outcome is to be assured. Even then, the practice might depart from the prescription. Who can implement such a policy? I argue that it has to be the legislature – it is not right to expect the courts to stick to such a policy. There are established examples of split entitlements. Most are in the form of default rules, which means that they can and are overridden by pre-assignment and other express terms. Immutable rules are scarcer. The United Kingdom policy for patents limits the employer to the inventions that might reasonably be expected to result from the duties specifically assigned to the employee.3 Australian policy limits the principal’s copyright in certain works to the purpose for which the work was commissioned.4 The Californian policy defines trade secrets narrowly and strikes down non-disclosure and non-competition clauses.5 Each

290


gives the firm something but leaves the worker free to use the knowledge in other ways – with the firm if they can reach agreement but elsewhere if not. The policy of split entitlements provides a particular focus. It is not as radically reformist as other policies to free up use, such as public domain and creative commons approaches or arrangements for pooling and licensing (see, for example, Merges 2001; Nicol and Hope 2006; Benkler 2006). We should expect greater circulation and sharing to result but it will be through the network of individual and commercial transactions rather than the public and collective bazaar (see Hope 2008).

III.

POLICY RATIONALES

This section considers some arguments why a cluster policy would mean a gain in innovation. A.

Production Incentives

Intellectual property policy is commonly characterised as the search for the right balance between the incentive that exclusivity gives the firm to invest in innovation (production, exploitation and diffusion) – and the access that enables others readily to use the resources too. Here, though, the question is more complicated because the balance has to be struck between two or more investors: the employer and employee or the principal and consultant.6 Traditionally, the concern has been that the firm has sufficient incentive to invest. Merges (1999) makes a good case for the firm to obtain control. His model is very much one of corporate research. Firms take the risk, they provide the resources, they manage the process, and staff work collectively in teams each contributing bits to the final product. So the contribution of the firm is the key to success. It exceeds the contribution of any worker, so the firm is the best place for the incentive to be applied. Letting firms establish pre-assignment clauses in individual contracts is the way to ensure this priority, with back-up from default rules. If employers had to obtain agreement only after a use had been identified, the employees might hold up the bargaining over the assignment, obstructing a whole innovation. Faced with this prospect, employers would be less likely to embark on innovation in the first place. Merges wrote his appraisal in 1999. Lately, the concern has been whether the worker is getting an appropriate return on her investment, especially where the employers are no longer offering any guarantee that they can stay in the firm through security of employment and a career path or partnership, share ownership and profit sharing (Arup 2006). The worker invests in firm-specific

Split entitlements?

291

skills, creative effort on the job, and the disclosure of innovations to colleagues and to the firm management. In the past, then, the focus has been on the incentives the workers receive inside the firm. Where the attachment to the firm is no longer guaranteed, the argument is that the worker receives her return in the employment wage or contract price that is paid. In the case of high fliers with bargaining power, that remuneration may include bonus and royalty payments for intellectual property production. Yet workers are still often paid little when they are starting out; in some industries they do not get paid at all. The rationale is that they are learning knowledge and skills that will be transferable through the industry. In the operation of quality circles and the like, the principal is doing the same with satellite sub-contractors. So, workers seek a new implicit psychological contract. Instead of security and longevity, they contract for capability and employability (Stone 2004). Such contracts should allow workers freedom to utilise intellectual assets elsewhere, while providing firms with sufficient return on their own investments. Now, though, if they cannot expect to stay with the one firm, they should be free to put that investment to other uses, beyond the firm-specific or transaction-specific uses the firm appropriates (Brooks 2002; Riley 2005). What then of the firm’s investments in these circumstances? With split entitlements, the firm still receives intellectual property in the knowledge it has directed or commissioned to be produced. Thus, it is awarded control of the primary use right or the trade secret when the intellectual resource is codified. But it is hard to say that it invested in further uses, certainly if they were unanticipated uses. It should honour the contract and leave the worker free to operate. If it really valued the further uses, it would find a way to get the worker to stay – to develop new uses with it rather than somewhere else (Hyde 2003). Merges sees that this is appropriate where the employee is not contracted to research or she uses her own time and resources to do so. But he supports the employer’s shop right (a non-exclusive, royalty-free licence) and he is happy for the employer to use freedom of contract to obtain full ownership. Likewise, he is happy for freedom of contract to override the default position that the principal only gets the use he commissioned. Pro-cluster policy would not allow such pre-assignments. B.

Exploitation Incentives

In a well-regarded piece, Hardy (1988) makes the case for the ‘work-for-hire’ doctrine which gives copyright to the employer. The ‘work-for-hire’ doctrine is a default principle that cuts in if the individual contract does not have a preassignment clause. For secondary uses, Hardy puts aside production incentives. He also puts aside the contention that the employer is better able than

292


the employee to identify further uses. This makes sense: in most of Hardy’s cases, the employer is taking the worker to court. So it is only after the worker identifies another use and exploits it, that the employer tries to get on board. Instead, the justification is that the employer is almost always the party better able to exploit the further uses. This sounds like the reason why young musicians assign all future use rights to the record company: the musicians need the company’s technical facilities and distribution channels if they are to be reach their audiences (Kretschmer, Klimis and Wallis 1999). There would appear to be an assumption behind this theory. If it is instrumentally minded, it is that the employer is more likely to command the necessary economies of scale and economies of scope for success in innovation. It is here that Merges’ argument is strongest too (Merges 1999). Innovation does not spring from one invention or creation. Achieving fruition depends on combining invention or creation with complementary assets. Giving control to the firm ensures the ‘common ownership of complementary assets’. Yet innovation does not always favour the incumbents. Today it is argued that the large bureaucratic firm is too inflexible to recognise and implement new uses. It is more likely to try to fit the new idea to its existing model (in which it has heavily invested), if not to suppress it altogether (Suthersanen 2006). So its assets will not necessarily be complementary. Large and resourceful firms still play a role in some sectors – a case in point is the product development stage of pharmaceuticals, where the (government imposed) health and safety product testing regulations are so demanding. However, if barriers to entry are coming down, new firms can be entrusted with exploitation. Workers are better able to seek out the right complementary assets (van Caenegem 2005). They may choose to stay with the original firm if it is prepared to invest but they need to be free to shift to another firm or start up on their own. Blanket reservations would inhibit the kind of mix and match patterns that production assumes in clusters – the ‘roundabout’ nature of production (Saxenian 2006). C. Mobility Incentives In clusters, the control exercised over the use of non-codified knowledge and creative powers could be more harmful to innovation than control in categories like patents and copyright. Recent studies indicate that transfers are promoted, not so much through the licensing of codes like patents, as when know-how is made available through contacts with other people (Mowery et al. 2004; Walsh, Cohen and Cho 2007). Workers’ mobility and freedom to operate are vital to knowledge transfers and network economies (Cunningham 2006). This is clearest for that kind of tacit knowledge that is very much bound up with the person.

Split entitlements?

293

In this vein, Gilson (1999) makes a case for California’s policy of hostility to non-competition and non-disclosure clauses. He attributes the success of its Silicon Valley cluster, at least in part, to the legal freedoms afforded its high velocity workers. It is this legal policy setting that distinguishes it from other locations, such as Massachusetts Route 128, which have not done so well. Merges (1999) recognises this local entrepreneurial spirit too and he recommends that employers who are losing workers take an investment position in their start-ups rather than trying to litigate the clauses aggressively. Gilson’s case was not based on his own empirical research but he did find support in a field study conducted of Silicon Valley success (Saxenian 1996). More recently, drawing on a large survey of US firms, Cooper (2001) confirms this intuition. Contractual clauses that restrict mobility might protect immediate research investments but they impede what are socially beneficial exchanges on all counts. The very same firms that lose workers benefit from the availability of others. Where restriction cannot be enforced (the conditions of appropriability are weak), the overall welfare effect is positive. The evidence even suggests that firms do not reduce their investment, that they still retain some (codified) intellectual property from the research that has been done, and that they avoid duplication of research if they have access to the knowledge of workers who come from other firms (Cooper 2001).

IV.

FURTHERING LEGAL POLICY

This section asks how the pro-cluster legal policy of split entitlements might be furthered. Will private contracts bring about an efficient allocation of rights or is it necessary for the courts or legislatures to set some limits? A.

Freedom of Contract

The Chicago School has been persuasive arguing that such allocations are best left to the parties and the individual contract. Following Coase’s theorem, they will strike their own mutually satisfactory bargain, unless transaction costs get in the way. Sometimes the argument is that they will make contracts even if no property rights apply, but in this field the Chicago School members recommend that intellectual property rights be established (Landes and Posner 2003). So contracts are framed by property rights. The parties are free to reallocate these rights if they think it is efficient to do so. But such bargaining can break down. Studies in several sectors show that transaction costs are an inhibiting factor but other factors come into play too. Significantly, one such factor might well be the presence of too many intellectual property rights, meaning too

294


many opportunities to hold out. There is too much transactional autonomy. One side sees no interest in contracting or wants too much from the contract. Thus, in some circumstances, too much is left to contracting over intellectual property. The best test of these propositions will come from an empirical study of contracting between firms and workers. From her experience with trade in biomedical research tools, Eisenberg (2001) indicates how a problem can arise. Conflict centres over and over again on the same handful of provisions. The most intractable disagreements are about reach-through and grant-back rights to potential future inventions that the owners of the tools demand in return for permission to use. Such conflict leads to serious bottlenecks and breakdowns. Transaction costs are an obstacle to agreement, mainly when the value of the research tool seems low. But there are other reasons why owners insist on ambitious standard form contracts. One factor Eisenberg identifies as institutional heterogeneity. In negotiations conducted across sectors, she finds that something like a cultural divide opens up, especially between university technology transfer offices and industrial companies. Another is conflicting agenda. Conflicts of interest mar contract negotiations, not just processing costs. An extra dimension to this conflict, one directly relevant to our inquiry, is the division within institutions or firms – between researchers and managers. Some researchers are inclined to communicate and cooperate with each other across firm lines, or they may be moving between the firms, but often they have signed non-disclosure and non-competition clauses which the managers can assert against them or the other firm. As a final factor, Eisenberg identifies differences in valuing the tools. Some owners are making money from selling the right to use the tools, while some are intent on preventing others from making competitive inventions with the tools. The other side wants the freedom to use the tools in their own research, often for publication, but sometimes so they can sell the results to their own customers. This finding recalls the distinction Benkler (2006) makes among the commercial and noncommercial motives of knowledge producers. B.

Contract Practices

How are these obstacles overcome in practice? In our field, the parties are making contracts of employment and commission already. Freedom of contract enables the firm to include pre-assignment clauses in advance of inventions or creations. They can also include broad non-disclosure and noncompetition clauses that apply even after the contract has ended. Hyde (2003) for example finds that often workers do not query broad retaining clauses when they first make contracts. Structured this way, the problem arises when

Split entitlements?

295

the firm wants to stop the worker from using the knowledge elsewhere. Stallberg (2007) found for instance that a number of UK university employers were willing to ‘reassign’ inventions to their employees. Issues may be sorted out when the worker is ‘exiting’ – or later. If a clause applies, the worker will then be arguing to be released from an obligation or, less commonly perhaps, contesting the validity of an obligation. Where the original contract makes no provision, then the parties will be negotiating afresh, perhaps with the aid of default principles (see below). In her field, Eisenberg (2001) notes a tendency for the parties to leave contracts open-ended (incomplete) when tools are licensed and then negotiate if there is a ‘hit’ downstream, that is, when it becomes apparent that something valuable is at stake. Possibly, another practice is for the worker to go ahead and infringe an employer’s right and then negotiate a solution with the firm. In another context, Merges (2001) recommends this ‘use now, pay later’ approach. All this sounds promising, if, in practice, the parties can negotiate. But freedom of contract has drawbacks. One is that the weaker party, through asymmetries of information, lack of legal resources or unequal bargaining power, is not able to make a contract in her best interest (Carrigan 1997). Furthermore, it is possible that the parties, left to their own devices, will not choose the contractual arrangement that releases the knowledge into the cluster. Even if they can make the decision calculated to serve their interests, they will not be concerned with the broader impact on innovation. Eisenberg (2001) suggests we cannot assume that private bargains will bring about an efficient allocation of rights. A corrective policy might concentrate on enhancing the process of contracting. However, the process of contracting is very difficult to police. Regulation can start with general standards or principles (common law or legislative fair dealing for instance), but finally the success of these disciplines relies on the parties’ preparedness to litigate. In Australia, we know enough of the small business experience with unconscionability and employee experience with duress to be sceptical of this protection having systemic effects. There is some indication that at their extremes the courts will decline to enforce pre-assignment, non-disclosure and non-competition clauses on the view that they go too far in restraint of trade. Again, to be practical, that response depends on the parties’ strength to litigate. The firm will bring an action to enforce a clause; the worker will be required to take the initiative. But she will have to be prepared to withstand the demands of litigation, usually against a heavier combatant. Those demands include legal information, legal costs upfront and the risk of losing, which runs in Australia to the prospect of serious cost penalties. The costs can be reputational and relational too, with responses from the immediate employer or commissioner and perhaps across the industry generally.

296


On this basis, such clauses could have an inhibiting effect, even if ultimately chances are they would be ruled invalid by a court (Bently 2002). Any uncertainty about the judicial response will increase the pressure on the weaker party. The worker might be unable to run the risk of an adverse ruling. Unless the worker is now backed by the resources of another large firm, which is a possibility, such clauses could have a ‘chill factor’. They discourage others from taking up the worker’s services, or the worker from striking out on her own. The serious student of such practices must map the structure of the field in which contracting will be conducted (the universe of relational contracting) (Bordieu 2005). This structure includes such attributes as the number, size and resources of the firms and the qualifications, organisation and scarcity of the talented workers. Then, as legal ‘anthropologists’, we should try to study the contract practices themselves. Socio-legal studies have methods to do this and it can be assisted by the research of organisational sociology into working conditions and identity formation among young creative workers (Westenholz 2004). It seems appropriate to bring into consideration intangible aspirations and emotional impacts to supplement the abstract modelling of the calculations of rational economic actors and the sometimes cold cynical language of the Chicago School. In such fields, intermediaries play a key role. For contracting to succeed, clusters need the presence of ‘an infrastructure of collaboration’ (Powell 2001). Among the intermediaries are employment agencies, management companies, venture capitalists, trade unions and professional associations. Lawyers are critical to the contracting process, not merely to drafting and enforcement but also to interest mediation and dispute resolution. Given our concerns above, lawyers can be a complicating factor, if they identify too rigidly with one interest, such as a large firm client, their university management or government principal. If contracts are to be mutual, all sides need access to competent legal assistance. Individual workers and small start-ups need legal support too. As Merges (1999) recognises, the workers’ lawyers could also be over-zealous. However, Bernstein’s study found that Silicon Valley lawyers saw themselves positively as brokers and value adders; they felt that they contributed to the success of the cluster overall (Bernstein 1995; see further, Suchman 2000). C.

Default Principles

The soft or light touch regulation of freedom of contract is the provision of default rules. Default rules are formulated for the convenience of the parties. They provide the parties with a fall-back position, should transaction costs discourage them from writing their own contract terms. Default rules fill gaps and resolve disputes, though they can require interpretation themselves.

Split entitlements?

297

Default rules can operate more prescriptively. For example, Hugh Collins (2001) recommends their use as a way of making employers and employees think about what is best for their particular situation and giving them a standard by which to do so. The parties must make the effort to override them with their own express clause if they think it is worthwhile to do so.7 Sometimes they will not. Cluster policy could inform default rules. In Australia, the default rules for patents and copyright favour the employer on the whole (for example, McKeough, Stewart and Griffith 2004). They are less generous to the commissioner of research or design. Copyright has legislative default provisions while the patent policy is in the common law characterisation of the contract of employment. Adopting an economic model of intellectual property, these rules are concerned with the assignment of ownership and control; they sidestep the issue of who is the inventor or author. For copyright, the Australian legislation gives the copyright with all its use rights to the employer, while, interestingly, confining the commissioner’s rights of copyright to the purpose of the contract. For patents, the legislation has no direct default rule. Instead, it permits the patent to be granted to anyone who is entitled to take an assignment of the patent from the inventor, which may be by way of a contract of employment or commission. Under contract law, the express assignment can include a pre-assignment clause, that is, a promise to assign any rights before any invention is actually made. Furthermore, under a contract of employment, any invention will by implication belong to the employer, so long as it is arrived at in the course of duties the employee is engaged to perform. In Australia, the rules for non-disclosure and non-competition are made entirely by the courts. Australian jurisdictions have not enacted trade secrets legislation like some other countries (United States, Switzerland). The courts start with individual contracts and the express terms of the parties, supplementing them with implied terms and other obligations. May we say that the common law obligations work like legislative default rules? The courts do not source these obligations solely in contract law; they say that the worker’s obligations – fidelity or loyalty, confidentiality, sometimes trust and confidence or fiduciary duties – arise as a matter of equity too, out of employment and other relationships (Dean 2002). The implied terms and other obligations work like legislative default provisions when they fill out the parties’ obligations; less dramatically they help to interpret any express terms that are uncertain in scope. Generally, these obligations impose controls on workers. They may be extended by express contract but they are substantial obligations in their own right. However, they do not work entirely as default rules. When for example the courts rule against unreasonable restraints on trade, they place some limits on the express terms of the contract and contract autonomy.

298

D.


Judicial Policy

The same virtue is the drawback with default ‘principles’: if contractual autonomy is upheld, the parties can override them wherever they see fit. If blanket reservations are not good for cluster activity, it might be necessary to go a step further in certain situations and limit the parties’ options for the greater good. That policy could for instance prevent the parties from assigning rights, certainly in advance of a use being anticipated or pursued. The courts are important to the legal policy in several ways. They are one of the key institutions ‘that manage and coordinate incomplete contracts’. They set some of the default rules and they interpret the rules, whether legislative or common law, from time to time. Whether interpretation is liberal or narrow can be important, for example to determining whether the worker’s innovation was within the terms of her employment or scope of her duties. They construe and uphold the express clauses that may go beyond the default rules. Generally, they defer to freedom of contract but they do strike down some clauses as a matter of public policy for being unreasonably in restraint of trade. Gilson (1999) and Hyde (2003) argue that the policy of the courts has been crucial to the hostility in California to non-disclosure and non-competition clauses. While the policy originates in a (nineteenth century) legislative provision, in other states that share the provisions the courts have read it down to allow reasonable restraints. The disapproving Californian judicial policy has extended to broad non-disclosure clauses that act in effect as restrictions on working elsewhere. Hyde (2003) thinks the judges have soaked up the Silicon Valley culture. However, the common law continues to evolve and Hyde found that some of the recent decisions were picking up on the notion of ‘inevitable disclosure’ to support such controls; a similar justification is that it is not possible to keep the specific information separate from the worker’s general body of knowledge. Interestingly, Merges (1999) had noted the same worrying trend. In a similar vein, commentators note the trend among recent Australian decisions. The decisions are being drawn out from more aggressive litigation by the employers. In particular, the commentators are concerned that the courts’ decisions on confidentiality are giving a proprietary legitimacy to postemployment controls, where outright restraints on competition might still be struck down. For example, van Caenegem (2007) queries the persistence of a loose concept of confidential information that extends beyond the category of concrete ‘trade secrets’ the employer can legitimately claim. Riley’s analysis of recent New South Wales decisions also identifies a willingness to extend confidentiality protection into new categories of information (2005; see further this volume).

Split entitlements?

299

If the judges are exercising their own judgements about the limits to control, they might be asked to make cluster policy their reference point. It is possible they would be receptive; the concern with restraint of trade connects with innovation policy. However, at least in Australia, the decisions in equity are redolent with notions of conscionable conduct, legitimate interests and just deserts. The reservation is the quite proper independence of the courts and the decentralised way in which they shape policy. For example, in Australia, these decisions are divided among the Federal Court and the courts of the six States. Policy coordination leads to the question of whether the decisions should be concentrated in a specialist higher court staffed by judges with expert training. But would that be at the expense of access to justice for workers and SMEs? Could such a court see the big firms as its main constituency, as its ‘heavy user’? E.

Legislative Intervention

In theory, the legislature can override the parties’ freedom and the courts’ discretion with sufficiently clear and binding rules along the lines indicated. These rules could restrict the firm to the uses of patents and copyright that they directed or commissioned to be produced. They could overrule non-disclosure and non-competition clauses except to protect a carefully circumscribed category of trade secrets. A jurisdiction that wanted to support cluster policy would be drawn to do so. While it might feel adventurous for one jurisdiction to encroach on the freedom of the large firms, legislation in Europe and the United States provides possibilities. If the legislature thinks that a rule allocating rights definitively is too obtuse, it might institute a rule of reason. In Australia, the restraints of trade in employment contracts are carved out of the competence of the national competition policy. If they were subsumed, this would allow more discriminatory styles of regulation. For example, the competition authority, the Australian Competition and Consumer Commission, would develop codes of conduct with industry. All the same, the flow of knowledge is now truly international. Researchers and writers come to work in centres like Silicon Valley; then they return home to participate in their own clusters and start up their own businesses. Large firms also establish operations off-shore, gravitating towards countries that have pools of skilled and inexpensive knowledge workers. Again, these workers break away and start their own enterprises. Some policy says these foreigners should not be educated in the west’s high technology advantages. But the benefits of circulation can accumulate on the global scale too (Saxenian 2006). The variations in national policies lead to conflict of laws (for example, D’Agostino 2005). That regulatory competition could be seen as healthy but

300


governments are asked to coordinate legislative controls. Does this impetus towards convergence restrict their autonomy to regulate for innovation? The World Trade Organisation TRIPs agreement is spreading strong intellectual property protection around the world and it supports the economic view of rights which allows their assignment to the firm. In its recent free trade agreements, such as the agreement with Australia, the United States has been urging the work-for-hire doctrine upon its partners. Before the TRIPs agreement was concluded, the United States had been pressing Japan to introduce trade secrets protection; with life-long employment, Japan had not seen the need. The TRIPs agreement also contains requirements that all member countries offer protection for trade secrets information. But it ties that protection to the information being disclosed to, acquired by or used by others in a manner contrary to honest commercial practices. It makes that concept situation-dependent when it indicates that such a manner includes breach of contract and breach of confidence. So the protection is not proprietary in the way that trade secrets legislation in some European countries establishes. The United States began to raise the issue of protection with other countries after TRIPs was adopted, but more recently, it has concentrated its efforts on protection of pharmaceutical test data (Arup 2008). Lack of uniform controls is a source of friction for firms that operate across borders. But the requirements at the international level are still so general and loose that they are not a major constraint – we can say that they allow a country to pursue the kind of regulation that suits local clusters.

V.

COMPLEMENTARY POLICIES

If we view legal policy as a component of a social project, not just some abstract principle of ownership, then we should connect legal policy with other policies for building workforce capacity and creative clusters. Within law, intellectual property policy should be connected with other legal policy, such as labour law policy, as Riley (2005) does when she argues freedom to operate is a fair trade-off for loss of employment security. Here is not the place to explore these policies in depth; but it is worth giving an indication of the wider world of policy in which intellectual property policy could be situated. Human and intellectual capital is a priority within these capacity building programmes. In cluster theory, the proportion of graduates in the local workforce is regarded as important to creative capacity (Florida 2005). This measure stresses the contribution that public education institutions make to the general knowledge and generic skills of the workforce.8 But it also looks to the opportunities to acquire industry-specific skills, both by studying with public

Split entitlements?

301

and professional organisations and by training with firms. Importantly, those skills cannot just be technical; they include design (recognised as a shortcoming of Australian industry), organisational, financial, inter-personal and crosscultural skills. Encouraging the temporary entry of people and permanent migration are strategies for human capital building too. Notably, Australia receives the benefit of skilled workers from other countries, including many students who come and stay. But Australia is also among those countries which need to send nationals overseas to learn and then to attract them back, so that they can spread their knowledge, skills and contacts locally. Social capital is vital too. Social capital policy is cluster-oriented in stressing the value of links throughout the community. It looks for means to share knowledge through civil society, educational institutions, professional associations, interest groups, informal networks and personal contacts, not just through commercial connections. These networks have been strengths in Silicon Valley too. Such social capital is inclusive and meritocratic, rather than exclusionary and elitist. If some social capital still privileges (old school tie, university prestige or family firm), then innovation policy should ensure that other groups (such as ethnic minorities) have the benefit of alternative outlets open to them to find commensurate work and start their own businesses (Hyde 2003). If clusters are to have a pool of skilled workers on which to draw, then workers need a cushion, a range of opportunities to exploit, whenever particular ventures fail. Otherwise, they will move somewhere else, possibly overseas. Again, it is not law primarily that cultivates these conditions; nevertheless, employment protection and anti-discrimination laws can make a contribution. Richard Florida (2005) argues that the social amenity of the locality is another ingredient in attracting and retaining talent. Such amenity includes the arts, the physical environment and social tolerance for diversity such as gender and sexual preference, bringing into account laws regulating expression, sex, violence and safety. Infrastructure settings can be influential too. While some high technology workers do not baulk at the high cost of urban living, many artists, designers, writers, freelancers and fledgling entrepreneurs depend on the availability of facilities like cheap working and living space, electronic communications and public transport.9 Incubators purpose-built for start-ups have their place. Yet the general environment might be more important. The role of zoning, town planning and urban design law should not be underrated. Slowly, we are realising that industry policy cannot rely on impersonal forces and brutal demands. Our socialist friends learnt that you cannot build an economy on heavy industry or even lots of electricity. One of the advantages of capitalism has been its capacity to promote small business and creative work. But capitalism has a tendency to consume its own children:

302


who can regulate it sympathetically? Some suggest that innovation clusters are entirely organic phenomena; and institutional efforts to engineer them make little difference to their success or failure. Still, city and regional governments are investing heavily in cluster strategies, trying to coordinate their regulation with other levels of government, while courting industry groups from international capital right through to small business. Cluster policy would endeavour to coordinate these policies with intellectual property policy.

NOTES 1. 2. 3. 4. 5. 6. 7. 8. 9.

For evaluation on my home region, Victoria, Australia, see Gilding (2008) and Mohannak (2004). See, for example, the report in The Economist (2007). Patents Act 1977 (UK), section 42(2). Copyright Act 1968 (Cth), section 35. Californian Labor Code. As a short-hand, we shall combine the employee and contractor in the concept of worker and the employer and principal or commissioner in the concept of the firm. Interestingly, a study finds that firms use the kind of standard boiler-plate clause (which law firms might provide) in much the same way (Hertzfeld, Link and Vonortas 2006): they are only a starting point for negotiation and customisation of the allocation. For example, Committee for Melbourne (2007). For example, City of Melbourne (2007).

REFERENCES Arup, C. (2006), ‘The Regulation of Human Capital: Public Frameworks and FirmBased Policies’, in C. Arup, P. Gahan, J. Howe, R. Johnstone, R. Mitchell and A. O’Donnell (eds), Labour Law and Labour Market Regulation, Sydney: Federation Press, pp. 637–56. Arup, C. (2008), The World Trade Organization Knowledge Agreements, second edition, Cambridge: Cambridge University Press. Benkler, Y. (2006), The Wealth of Networks: How Social Production Transforms Markets and Freedom, New Haven and London: Yale University Press. Bently, L. (2002), Between a Rock and a Hard Place: The Problems Facing Freelance Creators in the UK Media Market-Place, London: Institute for Employment Rights. Bernstein, L. (1995), ‘The Silicon Valley Lawyer as Transaction Cost Engineer’, Oregon Law Review, 74, 239–55. Berry, M. (2005), ‘Melbourne – Is There Life After Florida?’, Urban Policy and Research, 23, 381–92. Bordieu, P. (2005), The Social Structure of the Economy, Cambridge: Polity Press. Brooks, A. (2002), ‘“Fidelity” in the Post-Industrial Age: Developments in Case Law on Employee Disclosure of Confidential Information’, Monash University Law Review, 28, 126–48. Brousseau, E. and C. Bessy (2006), ‘Public and Private Institutions in the Governance

Split entitlements?

303

of Intellectual Property Rights’, in B. Anderson (ed.), Intellectual Property Rights: Innovation, Governance and the Institutional Environment, Cheltenham, UK: Edward Elgar, pp. 270–300. Carrigan, F. (1997), ‘Here Comes That Rainy Day Feeling Again: Shifts in the Ideology of Contract Law from Tony Macaulay to George Michael’, Law In Context, 15(1), 105–35. Casper, S. (2007), ‘How Do Technology Clusters Emerge and Become Sustainable? Social Network Formation and Inter-firm Mobility within the San Diego Biotechnology Cluster’, Research Policy, 36, 438–55. City of Melbourne (2007), Housing the Arts in the City of Melbourne, Melbourne: Community Services Committee. Collins, H. (2001), ‘Regulating the Employment Relation for Competitiveness’, Industrial Law Journal, 30, 17–47. Committee for Melbourne: (2007), Higher Education at a Tipping Point, Higher Education Taskforce, Melbourne: Committee for Melbourne, www.melbourne.org. Cooper, D. (2001), ‘Innovation and Reciprocal Externalities: Information Transmission via Job Mobility’, Journal of Economic Behaviour and Organization, 45, 403–25. Cornish, W. and D. Llewelyn (2003), Intellectual Property: Patents, Copyright, Trade Marks and Allied Rights, fifth edition, London: Sweet and Maxwell. Cunningham, S. (2006), What Price a Creative Economy?, Platform Papers, No. 9, Strawberry Hills, Sydney: Currency House. Cutler & Company (2003), Research and Innovation Systems in the Production of Digital Content and Applications, Report for the National Office for the Information Economy, Brisbane, http://www.cultureandrecreation.gov.au/cics. D’Agostino, G. (2005), ‘Freelance Authors for Free: Globalisation of Publishing, Convergence of Copyright, Contracts and Divergence of Judicial Reasoning’, in F. Macmillan (ed.), New Directions in Copyright Law, volume one, Cheltenham, UK: Edward Elgar, pp. 176-215. De La Mothe, J.G. and J. Mallory (2006), ‘Constructing Advantage: Distributed Innovation and the Management of Local Economic Growth’, Prometheus, 24, 2336. Dean, R. (2002), The Law of Trade Secrets and Personal Secrets, second edition, Sydney: Lawbook Co. Eisenberg, R. (2001), ‘Bargaining over the Transfer of Proprietary Research Tools: Is This Market Failing or Emerging?’, in R. Dreyfuss, D. Zimmerman and H. First (eds), Expanding the Boundaries of Intellectual Property: Innovation Policy for the Knowledge Society, New York and Oxford: Oxford University Press, pp. 223–50. Flew, T. (2005), ‘Creative Commons and the Creative Industries’, Media and Arts Law Review, 10, 257–64. Florida, R. (2005), The Flight of the Creative Class: The New Global Competition for Talent, New York: Harper-Business. Gaughan, M. and S. Robin (2004), ‘National Science Training Policy and Early Scientific Careers in France and the United States’, Research Policy, 33, 569–81. Gilding, M. (2008), ‘“The Tyranny of Distance”: Biotechnology Networks and Clusters in the Antipodes’, Research Policy, 37, 1132–44. Gilson, R. (1999), ‘The Legal Infrastructure of High Technology Industrial Districts; Silicon Valley, Route 128 and Covenants Not to Compete’, New York University Law Review, 74, 575–629. Hardy, I. (1988), ‘An Economic Understanding of Copyright Law’s Work-Made-forHire Doctrine’, Columbia–VLA Journal of Law and the Arts, 12, 181–227.

304


Hertzfeld, H., A. Link and N. Vonortas (2006), ‘Intellectual Property Protection Mechanisms in Research Partnerships’, Research Policy, 35, 825–36. Hope, J. (2008), Biobazaar: The Open Source Revolution and Biotechnology, Cambridge, Mass: Harvard University Press. Hyde, A. (2003), Working in Silicon Valley: Economic and Legal Analysis of a High Velocity Labor Market, Armonk, New York: M.E. Sharpe. Kretschmer, R., G. Klimis and R. Wallis (1999), ‘The Changing Location of Intellectual Property Rights in Music: A Study of Music Publishers, Collection Societies and Media Conglomerates’, Prometheus, 17, 163–86. Landes, W. and R. Posner (2003), The Economic Structure of Intellectual Property Law, Cambridge, Mass: The Bellnap Press of Harvard University Press. McKeough, J., A. Stewart and P. Griffith (2004), Intellectual Property in Australia, third edition, Sydney: LexisNexis. McRobbie, A. (2002), ‘From Holloway to Hollywood: Happiness at Work in the New Cultural Economy?’, in P. du Gay and M. Pryke (eds), Cultural Economy: Cultural Analysis and Commercial Life, London: Sage, pp. 97–114. Merges, R. (1999), ‘The Law and Economics of Employee Inventions’, Harvard Journal of Law and Technology, 13, 1–54. Merges, R. (2001), ‘Institutions for Intellectual Property Transactions: The Case of Patent Pools’, in R. Dreyfuss, D. Zimmerman and H. First (eds), Expanding the Boundaries of Intellectual Property: Innovation Policy for the Knowledge Society, New York and Oxford: Oxford University Press, pp. 123–66. Mohannak, K. (2004), ‘Innovation and Clustering among Information Technology Firms in Melbourne’, Prometheus, 22, 83–98. Mowery, D., R. Nelson, B. Sampat and A. Ziedonis (2004), Ivory Tower and Industrial Innovation: University–Industry Technology Transfer Before and After the Bayh–Dole Act, Stanford: Stanford Business Books. Nicol, D. and J. Hope (2006), ‘Cooperative Strategies for Facilitating the Use of Patented Inventions in Biotechnology’, in M. Rimmer (ed.), Patent Law and Biological Invention, Sydney: Federation Press, appearing also as Law In Context, 24(1), pp. 85–112. Noble, D. (1977), America by Design: Science, Technology, and the Rise of Corporate Capitalism, New York: Knopf. Office of the Victorian Auditor-General (2005), Managing Intellectual Property in Government Agencies, Melbourne, http://www.archive.audity.vic.gov.au. Pisano, G. (2006), ‘Profiting from Innovation and the Intellectual Property Revolution’, Research Policy, 35, 1122–30. Powell, W. (2001), ‘Networks of Learning in Biotechnology: Opportunities and Constraints Associated with Relational Contracting in a Knowledge-Intensive Field’, in R. Dreyfuss, D. Zimmerman and H. First (eds), Expanding the Boundaries of Intellectual Property: Innovation Policy for the Knowledge Society, Oxford: Oxford University Press, pp. 251–66. Riley, J. (2005), Employee Protection at Common Law, Sydney: Federation Press. Roberts, J. (2001), ‘The Drive to Codify: Implications for the Knowledge-Based Economy’, Prometheus, 19, 99–116. Saxenian, A. (1996), Regional Advantage: Culture and Competition in Silicon Valley and Route 128, Cambridge, Mass: Harvard University Press. Saxenian, A. (2006), The New Argonauts: Regional Advantage in a Global Economy, Cambridge, Mass: Harvard University Press. Stallberg, C. (2007), ‘The Legal Status of Academic Employees’ Inventions in Britain

Split entitlements?

305

and Germany and its Consequences for R&D-Agreements’, Intellectual Property Journal, 11, 489–530. Stone, K. (2004), From Widgets to Digits: Employment Regulation for a Changing Workplace, New York and Cambridge: Cambridge University Press. Suchman, M. (2000), ‘Dealmakers and Counselors: Law Firms as Intermediaries in the Development of Silicon Valley’, in M. Kenney (ed.), Understanding Silicon Valley: The Anatomy of an Entrepreneurial Region, Paolo Alto, Calif.: Stanford University Press, pp. 71-97. Suthersanen, U. (2006), ‘Technology, Time and Market Forces: The Stakeholders in the Kazaa Era’, in M. Pugatch (ed.), The Intellectual Property Debate: Perspectives from Law, Economics and Political Economy, Cheltenham, UK: Edward Elgar, pp. 230–67. The Economist (2007), Briefing: Pharmaceuticals, 27 January. Thompson, G. (2003), Between Hierarchies and Markets: The Logic and Limits of Network Forms of Organization, Oxford: Oxford University Press. Towse, R. (2001), Creativity, Incentive and Reward: An Economic Analysis of Copyright and Culture in the Information Age, Cheltenham, UK: Edward Elgar. Tsang, D. (2006), The Entrepreneurial Culture: Network Advantage within Chinese and Irish Software Firms, Cheltenham, UK and Northampton, New York: Edward Elgar. van Caenegem, W. (2005), ‘Inter-firm Migration of Tacit Knowledge: Law and Policy’, Prometheus, 23, 286–306. van Caenegem, W. (2007), ‘The Mobility of Creative Individuals, Trade Secrets and Restraints of Trade’, Murdoch eLaw Journal, 14(2), 265–79. Walsh, J., W. Cohen and C. Cho (2007), ‘Where Excludability Matters: Material versus Intellectual Property in Academic Biomedical Research’, Research Policy, 36, 1184–1203. Westenholz, A. (2004), ‘Emerging Identities beyond Organizational Boundaries’, in T. Jensen, A. Westenholz and P. Du Gay (eds), Identity in the Age of the New Economy: Life in Temporary and Scattered Work Practices, Cheltenham, UK and Northampton, MA, US: Edward Elgar, pp. 122–46.

15. Conclusion William van Caenegem and Christopher Arup The editors started out by articulating the sense that intellectual property (IP) law’s primary function in sustaining the realisation of creativity in society, whether in technology or the arts, was at risk of being relegated to secondary status. The growth of the edifice of IP law seemed to be casting a shadow over IP’s true goals. At the same time we were conscious of the multifarious voices in the universities which are addressing specific issues concerning IP, innovation and creativity. There appeared to be a growing divide between these voices of scholarly concern about IP and the support for expanding intellectual property rights (IPRs) evident in certain quarters of industry, government and the professions. That divide might be evidence of a malaise about IP’s true goals, though it might just reflect the different roles that academics and practitioners play. We therefore thought it would be worthwhile to invite a number of colleagues to contribute papers to this collection. The papers would be brought under the heading of contemporary critiques of IP law. The broadly unifying theme would be the contemporary role of IP in stimulating innovation and creativity. As part of this endeavour, we were also tempted to try to tease out some extra strands in IP scholarship. We wanted to show how IP law scholarship had grown beyond its preoccupation with the analytical study and representation of rules (of patents or copyright law for instance). At a high level of abstraction, the epistemological trend appeared to be towards increased contextualising. In other words, a scholarship that sets IP laws in a broader context and accepts that, for the purposes of policy development and law reform, IP should be assessed as it is likely to operate in different circumstances. We felt there was a reservoir of such scholarship that would support discussion in a shared forum. Our sense is that a broader, more contextualised scholarship will result in a more thorough understanding of IP as it contributes to the furtherance of innovation and creativity. That sensibility then has implications for the choice of reforms and application of the rules, as part of public policy, including policy for the arts and science innovation policy. When we reviewed the contributions, it was evident that contextualisation proceeds in various ways. One very significant form in which it is practised is 306

Conclusion

307

by setting legal scholarship in the context of other scholarly disciplines: economics, business management, human resource management, creative arts and innovation studies, for example. These days, legal scholars are inclined to visit a broader literature and learn from the insights in other disciplines. Another trend is to place IP in a broader context of regulation and governance, rather than to treat it as an isolated body of rules: IP links for example with labour law, with competition law, with the law of contracts and other legal dimensions of public policy. Perhaps the most important effect of contextualising is that it brings IP back down to ground. It is but a cog in the very large wheel which generates the cultural and material realities of innovation and creativity. When viewed this way, there is a sense that we are at risk of allowing IP over-regulation to stymie exactly what it is supposed to support. Creativity and innovation require room to move, so that the growth of statutory systems, the multiplication of rights in private hands, the increased risks of litigation and infringing others’ rights, all might become so great that investing in production and diffusion is regarded as too costly and risky. On the other hand, there is an emerging sense that we were not always looking in the right places. Some forms of critically important regulation receive very little attention. For example, those relating to patent administration, where particular methods for putting broader rules or principles into operation can have an important effect on the efficiency and legitimacy of intended structures. The examination standards and practices of the US Patent and Trademark Office or the technical assistance the major patent offices lend to patent examiners in developing countries work this way. Attention is drawn here also to the fact that the devil is often not in the detail of the rules, which tend to be broadly stated, but in the detail of their application: how exactly rules are interpreted and applied by IP offices and ordinary courts will have very major effects on the policy outcomes flowing from the regulatory system. In the detail of application there is considerable variation over time and space, and key battles for influence and capture are being waged. So scholars have to look ‘under the radar’ to find the hidden effects of industry and administrative practices that have a critical impact. A further consequence of this sense of over-regulation is that actors subject to regulation seek ways out. The more traditional focus here is on the fact that many statutory rules of IP are default rules and therefore the interest turns to contractual practices, in particular as they impact on employees and small and medium-sized enterprises. More radical is the creation of a universe where an easier regulatory model reigns, the universe of the creative commons and the biotechnology commons and so on. For our contributors, this is not necessarily a universe without IPRs, but where more open regulatory mechanisms enable interaction and production. In particular, this theme draws attention to

308


those areas of creativity and innovation that are not currently heavily regulated, nor the subject of great international harmonisation: for instance, employee rights, trade secrets protection, the university sector and the competition/IP law intersection. So although, in this volume, a position of working within an existing IP framework predominates, because international obligations make that inevitable, it also draws attention to the importance and potential of those areas of law and regulation that are less homogenous. Underlying many critiques is this trend towards international harmonisation. The positive aspect of harmonisation is of course that it can facilitate international trade. However, here some attention is also drawn to the detrimental impacts that may be suffered, if promoting genuine creativity and innovation is still the right goal. One pessimistic conclusion is that there is less room for regulatory experimentation and divergence in a world of homogenous regulation. Either, all jurisdictions get it wrong or get it all right – there is less opportunity for learning from the experience of other jurisdictions with different regulatory models. All mistakes are shared. This is an all the more worrying trend if, at the same time, IP intrudes into areas not traditionally touched by it – again the opportunity to learn from developments in unaffected spheres is lost. Think for instance about the patenting of business methods software, but also of public sector and university research. The other significant concern is the lack of room to adapt to particular conditions on the ground in particular jurisdictions – something that has resulted in widespread criticism of IP in developing countries in the wake of TRIPs. Different IP for different development levels and locally accountable forms of IP administration still seem to be important themes. Another risk of harmonisation of regulation is that it heightens the detrimental effects of regulatory capture and fallible policy development. There is an even greater need for protecting regulation and policy making processes from capture in these circumstances. Whoever can exercise influence over the selection process stands to gain enormous influence. It is important to recognise that harmonisation does not mean the adoption of a model that integrates all alternatives in one system; that is impossible. It is more likely to mean the cementing into place of one preference or choice over all others. Yet potential remains: the harmonised rules can leave space. The more flexible the harmonised or treaty rules, the more adaptability to local conditions and the more regulatory experimentation are likely. We argue that this space should be kept as open as possible. The difficulty with making regulatory choices is another important theme of this collection. Repeatedly stressed in the contributions is the fact that there is no single policy promise to IP. If selection of one model over another could be based on solid empirical grounds, then the adoption of a single regulatory model could hardly be faulted. However, a clear message from the book is that

Conclusion

309

the contention that IP generates more creativity and innovation must be carefully scrutinised. Then, when that scrutiny has been brought to bear, it does not provide a ringing endorsement of the need for IP, or at least for the IPRs in their present form and across all industries. IPRs can have a positive influence. But empirical evidence is necessary before the central tenets of IP are supported; as is the quest for alternatives to IPRs, or at least alternative uses of the basic forms of IP, which again implies a less than enthusiastic endorsement for IPRs to be seen as commercial propositions in their own right. A final theme that emerges from the volume is of a tension between the personal and the systemic. Systems, whether the legal system or the corporate system, generate nothing by themselves. It is the people within the system who bring about novelty and change. An emphasis on systems, on over-regulation, has a tendency to devalue the importance of the individual. Thus the contradictory nature of IP is brought to the surface. At the core of IPRs, and above all of patents, lies a notion that the immediate advantage of access to inventions must be denied for the broader good. For example, a sick person in a developing country should be refused a cheap drug because, if a patent is overridden to provide a cheap alternative, this will have a detrimental effect on future inward technology transfer, or investment in local research and development. The untested assumption is that, in the absence of IP, and possibly the centralisation of IPRs within large firms, there is a substantial reduction in the incentive to create and to innovate. This underestimates the non-proprietary incentives towards creativity and innovation throughout society; it also tends to ignore the variety of tools available to firms and individuals who want to safeguard their returns from innovation. A recurring theme of this book is that there are many strategies open to firms to appropriate more than marginal returns from investment. IP is but one element of an overall strategy. Likewise, there are many reasons for freeing up and sharing round IPRs. The carefully modulated tone of the various contributions in this book does not lend itself to unqualified precepts for the future. What does emerge clearly is that universal truths about the impact of IPRs are hard to come by. In the final analysis, the effect of IPRs is localised and environment-specific, both in a geographical and organisational sense. Yet the various IP schemes are increasingly undifferentiated and universal in their demands. This creates the difficulty of reconciling them with the particularity of the human, industrial or social context in which they operate. The answer in terms of rule formation and policy development must lie in treaty text and statutory law that allow for sufficient flexibility and adaptation to particular environments and microcosms. It also indicates that IPRs cannot exist in splendid isolation, elevated above the broader milieu of regulation in which they must operate: more specifically, the regulation of competition, labour, administrative practices,

310


industrial sector activity, and so on. It also means that the practices and sectors that have existed or emerged outside the confines of IPRs should not be coopted into mandatory legal schemes without careful consideration – there is probably good reason for their existence outside the reaches of dogmatic rule structures. The regulation of IP practices and outputs cannot benefit from universal rules that leave no room for experimentation and escape, no place for alternatives and self-regulating communities. In an era that trends towards greater harmonisation by way of international instruments, this is a significant message. The maintenance of a free zone of intellectual interaction and stimulation, of mobility, of self-regulation, becomes more critically important.

Index A & M Records Inc. v Napster Inc. 211, 215–16 AAMC 235 abuse of the patent system 47 academic freedom 221–2 accounting language 268–9, 270, 270–5 Adeney, E. 178 Administrative Judges 18 agreed patent database 20–2 agricultural biotechnology 63, 131–2 Alcazar, J.E. 145 Alkermes 66 all-species encyclopedia see Encyclopedia of Life American Association for the Advancement of Science 231 American Physical Society 179 Anderson, C. 176–7 Angiotech Pharm. Inc. v Conor MedSystems Inc. 22–3 Anglo-Australian Copyright Acts 203, 213 antibiotics 14, 82 anticommons problem 69–71, 130, 131–2, 138–9 anti-poaching covenants 272–4, 277 arbitration, technical 4, 17–18 Arora, A. 66–7, 68, 70–1 Arup, C. 290, 300 arXiv 180 Ashby, E. 221 assets, knowledge 155–7 Association of American Publishers 183 Association of American Universities (AAU) 235 Association of University Technology Managers (AUTM) 225, 233, 234 Australia 1, 63, 101–25, 267, 289 Backing Australia’s Ability (BAA) 237 biotechnology industry 74–5 clusters 287

commercialization of university research 221–49 objectives and policies 240–2 reviews and reform recommendations 237–40 Competition Principles Agreement (CPA) 103, 104–11, 122 Copyright Act 1968 108, 110, 122 default rules for patents and copyright 297 failure of small biomedical companies 67–8 human capital 301 induced patents 34–6 National Principles of Intellectual Property (NP) 238 nineteenth century newspapers 200, 201 patent review 1984 29 Patents Act 1990 101–2, 102–3, 122 amendments and new practices 111 regulatory quality and performance according to the CPA 104–11 restrictive covenants 280–1, 282 split entitlements 298–9 Therapeutic Goods Act 1989 103 Trade Practices Act 1974 101, 102, 103–4, 111, 122, 267 justification for partial exemption of the Patents Act 103–4, 112–21 transaction costs 45 Australian Code for the Responsible Conduct of Research (the Code) 239 Australian Competition and Consumer Commission (ACCC) guidelines 118, 120 Australian National University (ANU) 250

311

312


Australian Universities Quality Audit (AUQA) 250 Australian Research Council (ARC) grant conditions 239 Authors’ Guild 183 authorship collective 172–98 and ownership 177–80 Ayyangar Committee Report 82–3 background intellectual property 166–7 Backing Australia’s Ability – An Innovation Action Plan for the Future (BAA) 237 Bagchi, A.K. 87, 97 Baidu Baike 182–3 Bakels, R. 47 balance of probabilities evidence standard 111 Balkin, J. 194 ballads 199, 207, 212 Banerjee, P. 87, 97 Bar-Gill, O. 67 Barton, J. 46 Bartow, A. 32 Basheer, S. 93 basic research 55 with immediately identifiable commercial application 222–3 Bayh–Dole Act 58–9, 225, 232, 233, 234, 243 bazaar governance 132, 146–7 Beesley, A. 177 Bell v Whitehead 203, 207 Benkler, Y. 5, 62, 133, 137, 177, 287, 294 Bently, L. 296 Berger, P. 156 Berlin University 221 Bernstein, L. 296 Berry, M. 287 Bessen, J. 11, 30, 32, 33, 38, 46 ‘best endeavours’ clause 114–15 Bhattacharya, U.K. 87, 97 Bienstock, A. 233 bilateral agreements 16 BioBricks Foundation 142 biodiversity Encyclopedia of Life see Encyclopedia of Life

open source community for crop biodiversity conservation 145 Biodiversity Heritage Library 189–90 BioForge 143 bioinformatics software, open source 145–6 Biological Innovation for Open Society (BIOS) 143–4, 161–2 Biological Open Source (BiOS) licencing 63, 143–4, 161–2 biomedical research 5, 55–79, 131 changing the biomedical business model 74–6 history 57–60 strong patent rights and the small firm 63–9 strong patent rights and the university 60–3 strong patent rights and weak patent standards 68–74 follow-on innovation 69–71 twenty-first century 71–4 bioscience 154–71 assets, property and capital in knowledge 155–7 intellectual property to promote innovation 154–5 open licence models 158–63 openness under certain conditions 165–8 possibilities for openness 158 biotechnology, open source see open source biotechnology Bishop, E. 40 blocking 39 Board of Patent Appeals and Interferences 18 Bogsch, A. 15 Boldrin, M. 44 Botanicus 190 boundaries, patent 38, 46, 47 Bourdieu, P. 296 Boyce, G. 205 Boyer, H. 58, 61 Boyle, J. 130, 132, 142 Braithwaite, J. 4, 15 Brandt, D. 182 Brazil 80 BRCA2 159–61 breach of confidence cases 271

Index Breen v Williams 271 British Technology Group 42 broadness of patents 72–3 Brooks, D. 191 Bureau of Industry Economics (BIE) 29–30 Burk, D. 57, 67 C. V. Starr Virtual Herbarium 190–1 Cactus Imaging Pty Ltd v Peters 268, 274 California 289, 293 judicial policy 298 CAMBIA (Center for Application of Molecular Biology in International Agriculture) 143–4, 161–2 Cancer Research Technology Ltd (CRTL) 159–61 capability market power 12–13 capital, knowledge 155–7 capital payments 25–6 capitalism 2, 286, 301–2 Carnegie-Mellon Survey (CMS) 32, 33, 39, 40 Carrigan, F. 295 cartels 82 Casper, S. 287 CC-BY (Attribution) 188 CC-BY-NC (AttributionNonCommercial) 189 CC-BY-NC-SA (AttributionNonCommercial-ShareAlike) 189 CC-BY-SA (Attribution-ShareAlike) 188 CCST 224 Ceccagnoli, M. 68 Chand Patents Enquiry Committee 82–3 chemicals industry 33, 86, 90–1, 97 Chicago School 293 China 80 Cho, C. 71 Citizendium 174, 182, 183–4 Clarke, R. 184 click-wrap database licence 141 clusters 285–305 complementary policies 300–2 furthering legal policy of split entitlements 293–300 policy 288–90

313

policy rationales and innovation 290–3 theory 285–8 coalition of poorer countries 16–17 Coase’s theorem 293 codified knowledge/information 60–1, 288 coercion of labour, freedom from 269–70, 270, 278–9 Cohen, N. 180 Cohen, S. 58 Cohen, W. 12, 39, 44, 68, 70–1 Cole, J. 38, 44 collaborations 189–91 collective action theory 26 collective authorship 172–98 Encyclopedia of Life 172, 174, 184–92 Wikipedia 4–5, 20, 132, 172–4, 175–84, 187 Collins, H. 270, 297 commercial offices 238 commercialisation OSB and 134 university research 5, 221–49 areas of concern 226–7 benefits 228 defining commercialisation activity 224–5 international context 229–36 policy and development in Australia 237–42 practical problems in rejecting commercialisation 227–8 some degree of commercialisation as a sensible strategy 225–9 Commission of Intellectual Property Rights 131 common law doctrine making illegal restraints of trade 269, 270, 275–7, 281 common law obligations 297 commons-based peer production 132, 146–7 communities interpretive communities 93–4 openness within 166–7 universities’ role in economic and social renewal 228 Community Patent Review project 20

314


competition Australian competition law and patents 6, 101–25 competitors’ access to inventions 24–5 impacts of patent system 35, 42–5 innovation as competitive strategy 254–5 OSB and restructuring 137–8 patents regulating 256 problems with 255–6 Wikipedia’s imitators, competitors and rivals 182–4 Competition Principles Agreement (CPA) 103, 122 regulatory quality and performance of the Australian Patents Act 104–11 complementary proprietary products 137 complexity 91–2 compulsory licensing 24–5, 82–3, 111 Confederation of Indian Industry 85 confidence, equitable obligation of 260 confidential information 270–2, 276–7, 281, 298 confidentiality agreements 226 confidentiality clauses 260, 261, 262 conflicts of interest 226, 235 Congress Party 85 Conservapedia 182 conserved varieties register 145 Considine, M. 228 consumer surpluses 36–7 contextualisation 2, 306–7 contracts 307 contract law and restrictive covenants 6, 267–84 innovation by individuals and 257, 258–9, 259–62 psychological contracts 291 split entitlements 288–9, 293–7 contract practices 294–6 default rules 296–7 freedom of contract 293–4 control mechanisms, licences 160–1, 162 Controller of Patents and Designs 81 Cooper, D. 293 cooperative mechanisms 75 Copinger and Skone James on the Law of Copyright 203, 205

copyleft licences 134, 139, 147 copyright 3 default rules in Australia 297 and encyclopedias 175–6 online media 176, 178–82, 188–9, 192–4 and parallel importing 108–9, 110 and street literature 4, 199–217 marginal effect 200–1 new street literature 205–12 old street literature 199–205, 207 co-regulation 4 corporate capitalism 2, 286 costs and benefits of patent systems 29–54 competition and resource allocation 35, 42–5 dissemination of information 35, 40–2 impacts on innovation 35, 36–40 induced patents 31–6 priorities 47–8 transaction/regulatory costs 36, 45–7 used patents 31–6, 43 Council on Governmental Relations (COGR) 236 Court of Appeals of the Federal Circuit (CAFC) 59 covenants 121 restrictive see restrictive covenants Creative Cluster Study 289 Creative Commons 132 licences 188–9 creativity 250–1, 253–4, 307–8 optimal conditions for 257–9 Cukier, K. 141 Cunningham, S. 292 Cunningham, W. 172 Curran, J. 205 customer connection cases 272–3 DArT network 143 databases 89–90 agreed patent database 20–2 David, P. 168 Dean, R. 297 default rules 289, 296–7, 307 defensive patenting 32, 39, 68 deferred examination 90 Demil, B. 133

Index Department of Education, Science and Training (DEST) (Australia) 224 deregulation of labour markets 269–70 developing countries coalition of 16–17 and patents 93–4, 96–7 Diamond v Chakrabarty 58 direct protection of innovation 22–6 disclosure 65–6 disparate innovation 251, 257 dispute settlement 17 patent litigation 17, 38, 45–6, 95–6 technical arbitration 4, 17–18 dissemination 5–6, 25–6 American universities and effectiveness of 235–6 Australian universities 239–40 impacts of patent system 35, 40–2 Diversity Arrays Technology (DArT) 142–3 doctrine against restraints 269, 270, 275–7, 281 DNA patents 72 Doha Round 94 Dormer, M. 37 double patenting 92 Drahos, P. 15, 93 Drews, J. 135 Dreyfuss, R. 5 Drugs for Neglected Diseases Initiative 142 dual licensing 148 dynamic efficiency 107 Edison, T. 39 Edwards, C. 33 Edwards, J. 187 Eisenberg, R. 56, 69–70, 130, 294, 295 Electronic Frontier Foundation 132 Eli Lilly 58 embryo-free human stem cells 250 employee entitlements to statutory IPRs 259 employment contracts see contracts Encarta 176 encoded capacity building 94 Encyclopedia Britannica 176 Encyclopedia of Life 172, 174, 184–92 challenges 191–2 collaborations 189–91 legal dangers 188–9

315

Endeshaw, A. 39 Ensembl Project 185 entry of firms 44 environment 301, 309–10 Equitable Access licensing 144–5 equity 257, 259–62 Ergas Report 105–11, 116–19, 122 interim report 113–14 European Patent Observatory 47 European Patent Office (EPO) 41, 45, 93–4 Memorandum of Understanding with India 89–90 Research Fund 21 evergreening 92 examination of patents 73–4, 90 examiners, patent 3, 81–2, 89, 90, 93–4 excessive proprietary control 131, 138–9 exclusionary provisions 120, 121 exclusive marketing rights 85–6 exit of firms 44 exploitation incentives 291–2 expressed sequence tags (ESTs) 59 fair use and free speech 203–4 new street literature 4, 208, 209–10, 211–12 rise of 202–3 wiki-based sites 180–2, 193 Fanton, J.F. 187 fast-track patent process 95 Ferry, G. 141 filing costs 45 financial conflicts of interest 226, 235 Firestone, O. 31 firms entry and exit 44 incentives for innovation and the need for patents 251, 251–7, 262 large 286, 292 small see small firms First, H. 5 ‘flash of creative genius’ 14 Florida, R. 300, 301 follow-on innovation 69–71 Folsom v Marsh 206, 214–15 Food and Drug Administration (FDA) 21

316


foreign companies’ patent applications 87 foreign direct investment (FDI) 43 foreground intellectual property 166–7 François, D. 45 free/libre and open source software (FLOSS) 130, 132, 135, 147–8 programs for use in biotechnology settings 145–6 free speech 203–4, 210–12 Free Republic case 206–12 implications for free speech 210–12 freedom of contract 293–4 freedom to operate 73 Gaughan, M. 286 Genentech 58, 61, 66 General Agreement on Tariffs and Trade (GATT) 15, 43, 84 General Public Licence for Plant Germplasm (GPLPG) 140 Geneva Declaration on the Future of the World Intellectual Property Organisation 194 Germany 13 Ghosh, R.A. 135, 178 Gilson, R. 293, 298 Glavå, M. 156 Gleevec 92–3 GNU Free Documentation License 178, 183 Godwin, M. 180 Goldilocks hypothesis 67 Gomory, R. 186 goodwill 268–9, 272, 273, 274–5 Google 177, 183, 184 Google Book Search 183 Gowers Report 20 grace period 111 Grayling, A.C. 204 Griliches, Z. 44 Group of Eight (Australian universities) 224 Growing Esteem 240 Grushcow, J. 42 Guaragna, M. 144 Guellec, D. 41 Hall, B. 65, 68–9 Hamied, Y.K. 82

Hanover Insurance Brokers Ltd v Schapiro 273 HapMap Project 76, 141 Hardy, I. 291–2 Harman, G. 227 harmonisation 22, 308, 310 Hartleys Ltd v Martin 273, 277–8 Heald, P. 43 Heller, M. 69–70, 130 Herbert Morris Ltd v Saxelby 269, 275 Heydon, J.D. 272 Higley, G. 188, 189 Hilmer Report 101, 122 hold-ups 39, 69–71 Hope, J. 4, 61, 63, 65, 133, 144, 146, 147, 148 Hopkins, M. 72, 73 Horn, L. 63 House of Representatives Standing Committee on Science and Innovation (HRSCSI) 225 Howard Report 224 Hugenholtz, P. 47 human capital policy 300–1 human genome project 59, 140–1, 185 human rights 269–70, 270, 278–9 humanitarian purposes 163 Humboldt, W. von 221 Hunt, R. 32 Hunter, D. 178, 205–6, 210, 211, 212 Hyde, A. 286, 291, 294, 298, 301 IBM 137 IceTV v Duncan Ross and Ors 278, 280, 282 immutable rules 289 Imperial College London 231 Imperial Innovations 231 improvements, right to 161–2 In the Public Interest: Nine Points to Consider in Licensing University Technology 233–4 inadvertent infringement 38 incentive rationale 268 incentives 250–66, 309 cluster policy and innovation 290–3 firms, innovation and the need for patents 251, 251–7, 262 individual innovation 251, 257–62, 262

Index to participate in OSB projects 134–8 pervasive 253–5 income from research, universities’ 227 incomplete contracts 295 incomplete production (wikis) 193 incontestability period 20–2 incremental innovation 24–5 independent invention 39 India 15 encouragement of R&D outsourcing to 90–1 Patents Act 1970 83–4, 86, 87, 88, 91–2, 95, 97 2005 amendment 92 Patents (Amendment) Act 2002 86 Indian Patent Office 3, 80–100 modernization 88–90 re-design of colonial institution 81–5 patent interpretive community 93–4 TRIPs 85–6, 94, 97 individuals creativity 253, 257–9 incentives to innovate 251, 257–62, 262 legal mechanisms 259–62 optimal conditions for creativity 257–9 tension between systems and 309 induced patents 31–6 inducement doctrine 181 inherited values 221–2 injunctions 279–81 innovation 2–3, 307–8 clusters and 287, 290–3 direct protection of 22–6 disparate 251, 257 impacts of patent system 35, 36–40 incentives and 250–66 firms 251, 251–7, 262 individuals 251, 257–62, 262 intellectual property, innovation and openness 154–71 and patents in biomedical research 55–79 patents and investment in 11–12 Innovation Authority 23 innovation networks 229 innovation patents 23 innovation policies 230 innovation warrants 23

317

institutional conflicts of interest 226 insulin 58 intellectual property assets, property and capital in knowledge 155–6 background, foreground and sideground 166–7 and establishing openness 157–8 promotion of innovation 154–5 Intellectual Property Committee (IPC) 16 Intellectual Property and Competition Review Committee (IPCRC) 30 intellectual property policies, universities’ 238, 240–2 interference procedure 18 intermediaries 296 International HapMap Project 76, 141 international harmonisation 22, 308, 310 International Labour Organisation (ILO) Declaration of Fundamental Rights 269 International Patent Classification (IPC) system 95 International Searching Authority (ISA) status 80, 87, 89 international standardisation 7 International Treaty on Plant Genetic Resources for Food and Agriculture (ITPGRFA) 140 Internet 146–7 collective authorship 172–98 new street literature 205–12 Internet Corporation for Assigning Names and Numbers (ICANN) 184 interoperability 136, 192 interpretive communities 93–4 inventing around a patent 38–9 invention 22 inventive step requirement 72 inventiveness 253–4 investment in innovation 11–12 IPAC 40, 43 Italy 33 Jackson, J. 144 Jacobi Tool and Die v Mondi and Ors 280, 281 Japan 300

318


Jefferson, R. 144 Jensen, K. 70 job security 269–70 Joerges, C. 4 John Fairfax Publications v Birt 271–2, 276 judicial policy 298–9 Kahin, B. 47–8 Kamm, O. 174 Kapcynski, A. 144 Keen, A. 174 Kew Royal Botanical Gardens 191 Keynes, J.M. 16 Kilian, A. 142 Kingston, W. 23, 64 Kipling, R. 202 Klimis, G. 292 Knol 174, 182, 184 knowledge assets, property and capital in 155–7 politics of 172–98 spillovers 36–7 knowledge production model 222, 224 Kondo, E. 43 Koons, J. 215 Koops Martin v Dean Reeves 273 Kores Manufacturing Co. Ltd v Kolok Manufacturing Co. Ltd 274, 278 Kretschmer, R. 292 Kronz, H. 23 Labich, K. 41 labour, freedom from coercion 269–70, 270, 278–9 labour law 300 and individual creativity 259–62 labour markets, deregulation of 269–70 Lambert Review 230 Lamberton, D. 40 Landes, W. 293 Lanier, J. 174 Lanjouw, J. 45–6 large firms 292 disintegration of the large firm 286 Lastowka, G. 178 Latour, B. 169 Lecocq, X. 133 legal aid 17 legal services 96–7, 296

legislative intervention 299–300 Lemley, M. 57, 73 Lerner, J. 44 Lessig, L. 192–3 Levine, D. 44 liberal capitalism 286 licences of right 83, 84, 92 licensing 40, 71 compulsory 24–5, 82–3, 111 open licensing systems 4–5, 158–63 open source licensing 132–3, 139, 147–9, 194 US universities 232, 233–4 licensor obligations 160, 162 Lindner v Murdoch’s Garage 272 litigation, patent 17, 38, 95–6 costs 45–6 Littlewoods Organisation Ltd v Harris 276 local environment 301, 309–10 London newspapers 200, 201 Los Angeles Times v Free Republic 205–12 Luckmann, T. 156 Luhmann, N. 169 Lumley v Wagner 279–80 Macdonald, S. 29, 41, 42, 61, 65–6 Machlup, F. 29, 31, 38–9, 40, 47 Macquarie University 241 Maggbury v Hafele 269 managers, and researchers 294 Manber, U. 184 Mandeville, T. 38, 40 Mansfield, E. 32, 36, 43 Marginson, S. 228 market 6 market-based exchanges of information 64–8 technology markets 39–40 market harm 211 market positioning 136–7 market power 12 types of 12–13 market pull innovation 222 Martinez, C. 41 Maskus, K. 1 Massachusetts Route 128 293 Mathews, J. 16 Maurer, S. 141, 142, 144

Index May, C. 225, 227 McDonnell, B. 67 McRobbie, A. 286 Melbourne University 240–1 Merchant of Venice (Shakespeare) 277 Merck Gene Index 185 Merges, R. 44, 64–5, 66–7, 69, 290, 291, 292, 293, 295, 296, 298 Metro-Goldwyn-Mayer Studios Inc. v Grokster Ltd 181 Meurer, M. 11, 30, 33, 38, 46 Michaels, T. 140 Mill, J.S. 204, 214 Mischlewski, D. 37 Missouri Botanical Garden 190 Mithal, V.P. 81, 82, 86–7 Mobilesoft 278 mobility 5–6, 254, 257, 259–60, 261, 262 anti-poaching clauses 272–4 incentives 292–3 model law of intellectual property 15 Mohan, R. 90 Moir, H. 32 Monash University 241 money, measurement by 23–5 monopoly power 42 Monotti, A.L. 228, 230, 238, 239, 240, 242 Monthly Chronicle 203 Mooney, B. 45 Mowery, D. 59, 62 multiple 24, 25 Murray, F. 70 Myers, K. 180, 182 Myriad Genetics 159 national innovation surveys 43, 47 National Institutes of Health (NIH) (USA) effectiveness of diffusion 236 Grants Policy 232 principles for patenting 234–5 National Library of Medicine (USA) 236 National Museum of Natural History at the Smithsonian Institute 191 National Principles of Intellectual Property (NP) (Australia) 238 National Research Development Corporation (NRDC) (UK) 230

319

National Survey of Research Commercialisation Years 2003 and 2004 (NSRC) (Australia) 225, 227 negative externalities 30 Neglected Disease licensing 144–5 Nelson, R. 39, 69 Netanel, N.W. 193 network effects 136 Network for Open Scientific Innovation (NOSI) 144 networks 229, 301 see also clusters new street literature 205–12 New York Botanical Garden 190–1 newspapers 200–5 nineteenth century 200–1 Nibart-Devouard, F. 183 Nicol, D. 4, 60, 62, 63, 64, 65, 67–8, 72–3 Nielsen, J. 60, 62, 64, 65, 67–8, 72–3 non-competition clauses see restrictive covenants non-disclosure clauses 294–6, 298 non-exclusive licences 159–60 non-proprietary business opportunities 136–7 non-proprietary tool kit 139 Nordhaus, W. 11 Novartis 92–3 novelty 254 Nupedia 176 objectives of research universities 240–1 Office of Best Practice Regulation 111 Olson, M. 26 online media 172–98 Encyclopedia of Life 172, 174, 184–92 Wikipedia 4–5, 20, 132, 172–4, 175–84, 187 open access 75–6, 139, 147, 190 open arenas 158, 167–8 Open Hardware 132 open licensing systems 4–5, 158–63 BiOS 63, 143–4, 161–2 Cancer Research UK 159–61 PIPRA 63, 162–3 open review 19–20 open science 164

320


open source biotechnology (OSB) 5, 129–53 anticommons problem 130, 131–2, 138–9 challenges in implementing 146–50 excessive proprietary control problems 131, 138–9 incentives for participation 134–8 key elements of approach 132–4 nature of 129–30 real-world examples 140–6 Open Source Definition (OSD) 148 Open Source Initiative (OSI) 148 open source licensing 132–3, 139, 147–9, 194 open source model 166 open source research platforms 166–7 openness 5, 154–71 intellectual property-based structural arenas for 167–8 possibilities for establishing 158 with as few restrictions as possible 165 within a community 166–7 within the research group 165 within the research project 165–6 Oppenheim, C. 41, 45 Oppenheim, J. 179–80 originality 253–4 orphan drugs 21, 23 orphan works 189, 194 Otis Elevator Co. Pty Ltd v Nolan 280–1 outsourcing to India 90–1 over-regulation 307 P2P Foundation 132 Page, R. 192 pamphlets 199, 207 parallel importing 108–9, 122 Parchomovsky, G. 67 Paris Convention 15, 83, 86 1925 meeting at the Hague 13–14 Article 19 26–7 Pasquale, F. 210 passion 227–8 Patel, P. 72 patent attorneys 96–7 patent classes, searching 41 Patent Cooperation Treaty (PCT) 80, 86, 87–8 patent examination/assessment 73–4, 90

patent examiners 3, 81–2, 89, 90, 93–4 patent interpretive community 93–4 Patent Lens 143 patent offices 80–1 India 80–100 patent portfolio races 44–5, 71, 72 patent searches 111 patent secretaries 81 patents 3–4, 11–28, 307 abuse of the patent system 47 agreed patent database 20–2 Australian competition law and 6, 101–25 broadened scope of patentable subject matter 223 commercialisation of university research 228, 234–5 costs and benefits of patent systems 29–54 decline in quality 18–19 default rules in Australia 297 evolution of 13–14 incentives for innovation in firms and 251, 250–7, 262 patents as incentives to invent 251–3 patents regulating competition 256 problems with competition 255–6 India’s strategy on 90–1 and innovation in biomedicine 55–79 strong patent rights and small firms 63–9 strong patent rights and universities 60–3 strong patent rights and weak patent standards 68–74 need for reform 11–17 reform proposals 17–26 penicillin 14 personal conflicts of interest 226 pesticides 84 persuasion market power 12–13 pervasive incentives 253–5 see also incentives Petrusson, U. 155, 156, 157, 158 pharmaceuticals industry 14, 15, 33 exploitation incentives 292 India 82, 84, 85, 86, 92–3, 97 encouragement of outsourcing R&D to 90–1 patents as incentives to invent 252–3

Index Physical Review Letters 179 pipeline protection 85, 87 Pisano, G. 75, 288 plagiarism 180–2 Plant, A. 38 politics of knowledge 172–98 portfolio patent races 44–5, 71, 72 positive externalities 30, 36–7 Posner, R. 293 post-Chandlerian school 64, 76 post-grant opposition 95 Powell, W. 135, 296 Prakash, V.V. 135 pre-assignment clauses 294–6 pre-grant opposition 95 pre-incubators 167–8 Pressman, L. 60 price fixing 121 prior art searches 19–20 prior use defence 111 privity doctrine, ignorance of 269, 270, 277–8 process benefits 135 production incentives 290–1 productivity 44 property, knowledge 155–7 prospect theory 108 provincial newspapers 201 psychological contract 291 public choice theory 46 Public Intellectual Property Resource for Agriculture (PIPRA) 63, 162–3 Public Library of Science 132, 185 public research organizations (PROs) 157–8 Public Resource.org 191 publication 230 PubMed Central 236 pure research see basic research quality of patents 18–19 Quantiki 185 Queensland University 241 Quiggin, J. 205–6, 210, 211, 212 Qwiki 185 Rai, A. 141, 142 Railway Times 203 Ralston, L. 240 Rao, P. Narasimha 85

321

rational ignorance 73 Raymond, E. 174 Re Fisher 72 recombinant DNA technology 57–60, 61 re-contextualisation 211 Red Hat 142 regulation costs 36, 45–7 regulatory capture 46, 308 regulatory competition 299–300 Reichman, J. 4, 15 ‘relates to’, interpretation of 115–16 renewal fees 111 rent extraction 91 representative democracy 204 reputation 12 Research Councils UK 230 research and development, outsourcing to India 90–1 research directions 38 research groups, openness within 165 research platforms for openness 164–8 research projects, openness within 165–6 researchers, and managers 294 resource allocation 35, 42–5 Responsible Use of Public Research programme 157 restrictions justification of restrictions on competition 109–10 openness with as few as possible 165 restrictive covenants 6, 260, 261–2, 267–84 complaints common to recent decisions 268–79 corruption of common law doctrine 269, 270, 275–7 human rights dimension 269–70, 270, 278–9 ignorance of privity doctrine 269, 270, 277–8 inept adoption of accounting language 268–9, 270, 270–5 practical measures 279–81 split entitlements 294–6, 298 Reuveni, E. 178 reward theory 108 Ricketson, S. 228, 230, 238, 242 rigour 1, 3–4 Riley, J. 273, 278, 298, 300 Rimmer, M. 175

322


Rio Convention on Biological Diversity 185, 192 Roberts, J. 288 Robin, S. 286 Robinson, J. 206, 208 royalties 25 safe harbours 180–2, 193–4 Sainsbury Review 231 Sali, A. 141 Samuel Report 116, 117 Sand, I.-J. 4 Sanger, L. 173, 174, 176, 183 Sanger Institute 141 Saxenian, A. 292, 293, 299 scale 12 Schankerman, M. 45–6 Scherer, F. 32, 33 Schollin, K. 164 Scotchmer, S. 69 science, public perceptions of 253 Science Commons 132 second enclosure movement 132 secrecy 41–2 see also trade secrets Sell, S. 15, 16 semiconductor industry 65, 68–9 service provision 136 Shakespeare, W. 277 Shaw, H. 190 Shawn Bentley Orphan Works Act 189 Sheehan, J. 41 Shepard, L. 199, 212, 214 Shiva, V. 145 sideground intellectual property 166–7 Silberston, Z. 32 Silicon Valley 281, 293, 296 simultaneous invention 44 Sjoberg, L. 175 SklogWiki 185 Slow Food project 145 small firms 61–2, 74–6 strong patent rights and 63–9 Smithsonian Institute 191 SNP Consortium 76, 137, 185 social amenity of the locality 301 social capital policy 301 social constructions 156–7, 169 software programs, open source 145–6 solicitation 277

Sparling, D. 75 specific market power 13 ‘specific, substantial and credible’ standard 111 Spekkens, R.W. 179 Spender, J.-C. 169 spin-off companies 60–1 split entitlements 6, 285–305 cluster policy and innovation 290–3 complementary policies 300–2 furthering legal policy 293–300 policy context 285–90 policy rationales 290–3 Sprouse, G. 179 St James’ Gazette 202 stable workforce 273–4 Stallberg, C. 295 Stallman, R. 41, 139 standardisation, international 7 Stern, S. 70 Stokes, D. 56 Stone, K. 270, 291 strategic patenting 32 street literature 4, 199–217 new 205–12 old 199–205, 207 strong patent rights 3, 56–7 positives of 63–4 and small firms 63–9 and universities 60–3 and weak patent standards 68–74 and follow-on innovation 69–71 twenty-first century 71–4 structural transformation 155, 168 sub-licensing 162 Sulston, J. 141 Suthersanen, U. 286, 292 Sykes, Sir R. 231 systems 4, 309 tacit knowledge 3, 60–1, 260, 292 Tapscott, D. 173, 184–5 Taylor, C. 32 Taylor, G. 142 technical arbitration 4, 17–18 technology 254 development and OSB 133 markets 39–40 technological hold-ups 39, 69–71 technology push innovation 222, 230

Index Teubner, G. 4, 183–4 The Synaptic Leap (TSL) 142 third-line forcing 120, 121 Thomson, J. 250 threshold tests 107, 108 Thurbon, J. 16 Times, The 202, 204 Tkacz, N. 174 total factor productivity 44 Towse, R. 286 Trade Related Aspects of Intellectual Property (TRIPs) 7, 16–17, 22, 26, 74, 154, 300 India and 84–6, 94, 97 role of corporations 15–16 transitional provisions 85 trade secrets 260–1, 270–2, 281, 300 trademarks 47 trading of patents 40 Traditional Knowledge Digital Library 95 Traditional Knowledge Resources Classification System 95 training of patent examiners 93–4 transaction costs 117 impacts of patent system 36, 45–7 transactions 1 Transfield Pty Ltd v Arlo International Ltd 114–15 transformation 208, 211 transitional arrangements 119 translational research 55 Treaty on Access to Knowledge (A2K) 194 trilateral 26 trolling 18–19 Tropical Diseases Initiative (TDI) 141–2 United Kingdom (UK) 95, 289 commercialisation of university research 223, 230–1 licences of right 84 Patents and Designs Act 1919 84 United States of America (USA) 300 Bayh–Dole Act 58–9, 225, 232, 233, 234, 243 biotechnology companies 63 commercialisation of university research 231–6, 243 reviews and reform recommendations 232–6

323

Constitution 13 Digital Millennium Copyright Act 1998 181, 182 fair use doctrine 180–1 history of biomedical research 57–60 induced patents 34, 36 interference procedure 18 Orphan Drug Act 1983 21, 23 Patent Act 1952 14–15, 19 Shawn Bentley Orphan Works Act 2008 189 WTO action against India 85, 87 United States Patent and Trademark Office (USPTO) capacity building in developing countries 93–4 interference procedure 18 universal rules 309–10 universities 167–8 commercialisation of research 5, 221–49 strong patent rights and 60–3 University Companies Association (UNICO) 225, 231 Uruguay Round 43, 84 use of patents 31–6, 43 user benefits 135–6 uses of intellectual property rights 2, 4–5 valley of death 167 Van Caenegem, W. 68, 282, 292, 298 Van Doren, C. 172 Van Pottelsberghe, B. 45 Venter, C. 59 virtual herbarium 190–1 Vitale, M. 75 voluntarism 6 voluntary obligations to disseminate research 236, 239 Wadman, M. 57 Wales, J. 176, 177 Wallis, R. 292 Walsh, J. 39, 70–1 Walter v Howe 202 Walter v Lane 204 Walter v Steinkopff 202 Washington Post 208 Watt, J. 39

324


weak patent standards 3–4, 68–74 follow-on innovation 69–71 twenty-first century 71–4 Weisburst, S. 33 Weiss, L. 16 Westenholz, A. 296 wiki communities 177 Wikia Inc. 177 Wikia Search 177 Wikibooks 177 Wikileaks 182 Wikimedia Commons 177 WikiMedia Foundation 177, 181–2, 182–3 Wikinews 177 Wikipedia 4–5, 20, 132, 172–4, 175–84, 187 authorship and ownership 177–80 imitators, rivals and competitors 182–4 plagiarism, fair use and safe harbours 180–2 Wikiquote 177 wikis 172 WikiScanner 182 Wikisource 177 Wikispecies 177

Wilkins v Aikin 202–3, 207–8 Williams, A. 173, 184–5 Williams, R. 200 Wilson, E.O. 185–6, 191, 192 Wingate, P. 205 Winter, A. 179 Wookieepedia 182 Woolworths v Olson 271 work-for-hire doctrine 291–2, 300 World Fairs 38 World Trade Organization (WTO) 84–5 dispute resolution between India and USA 85, 87 India 85, 87, 94 TRIPs see Trade Related Aspects of Intellectual Property World War I 13 Wright brothers 39 Wright v Gasweld Pty Ltd 271, 276 Yencken, J. 240 Yeo, R. 175 Ziedonis, R. 65, 68–9 Zimmer, C. 184 Zimmerman, D. 5 Zittrain, J. 173–4, 176, 180

Intellectual Property Policy Reform: Fostering Innovation and Development

Intellectual property law and policy

Innovation, Intellectual Property, and Economic Growth

Public Health: Innovation and Intellectual Property Rights

US Intellectual Property Law And Policy