Fostering Dynamic Innovation, Development and Trade: Intellectual Property as a Case Study in Global Administrative Law

The World Trade Organisation (WTO) is currently suffering from a lawmaking deficit. Reliance on consensus-based decision-making has stymied the political process. At the same time, lawmaking through the Understanding on Dispute Settlement (DSU) 1 suffers from concerns about the legitimacy of law created through 'adjudication' rather than party agreement. 2 Less well understood are the special problems that this deficit poses for TRIPS, the intellectual property portion of the WTO Agreement. 3 Admittedly, a great deal of attention has focused on the question of making essential medicines available in the developing world. 4 However, that issue barely scratches the surface of the problems that the creative community is facing.

As Andreas Lowenfeld and I suggested when the Uruguay Round ended, fitting intellectual property into a trade agenda distorted perceptions about the structure of intellectual property law. 5 While domestic regimes strive to strike a balance between the interests of producers in earning a return from their intellectual investments and the interests of users in accessing new knowledge, the WTO's concern was with creating 'trading chips'with commodifying information streams and turning knowledge products into articles of commerce. As a result, the TRIPS Agreement deals almost exclusively with proprietary interests. It obliges members to recognise a series of intellectual property rights (patents, trademarks, copyrights, as well as rights over trade secrets, geographic indications, layouts of integrated circuits, and industrial designs); it specifies the minimum contents of these protective regimes (the duration of protection, the scope of exclusivity, limits on defences to infringement, and such); and sets out required methods of enforcement (including specific judicial procedures and forms of relief). Unlike other intellectual property instruments, which similarly focused on right holders' interests, TRIPS makes these obligations subject to international dispute resolution. While the Agreement also claims that its goal is to balance proprietary and access interests and includes 'flexibilities' that give members authority to support the public interest, 6 these provisions are highly circumscribed. 7 In disputes over their scope, WTO adjudicators have proved wary of expansive interpretations. 8 As a consequence, WTO members at all points along the development spectrum are now concerned that they lack the leeway they need to support creative production, keep the law responsive to a changing technological landscape, and safeguard the interests of consumers of intellectual goods.

In truth, the deficit in lawmaking is not only a function of WTO involvement in the intellectual property system. Because intellectual property law is expressly aimed at fostering dynamic innovation, the responsiveness of lawmakers to the information sector is an issue at the national level as well. For example, the Internet has revolutionised how copyrighted material is distributed and, in some cases, produced. Where patents were once associated with individual products, they map onto information technology in entirely new ways; manufacturers must now Round's public health legacy: strategies for the production and diffusion of patented medicines under the amended TRIPS provisions' (2007) lOJournal of International Economic Law 921. R R Dreyfuss & A Lowenfeld 'Two achievements of the Uruguay Round: putting TRIPS and dispute settlement together' (1997) 37 irginiaJournal ofituenational Law 275. assemble a multiplicity of patents before they can bring products to market. By the same token, international travel, global distribution, and Internet marketing broaden consumers' knowledge base and call the territoriality of trademark protection into question. Equally important, the differential impact of new technological opportunities on various creative sectors has altered the political economy. As demonstrated by the continuing difficulty in adopting an EC-wide patent system and reforming the US Patent Act, domestic lawmaking is also at something of an impasse. 9 Just as the WTO has, essentially, deflected lawmaking to DSU adjudicators, several countries have tried to solve their various lawmaking problems with specialised courts. The process is particularly well advanced for patents, where something of a crisis has been produced by the critical nature of the subject matter (medicines are a good example), abstruse science (e.g. biotechnology), shifts in the organisation of the technological enterprise (including university spin-offs and private/ public joint ventures), and changes in the way that patents are utilised (e.g. defensive patenting and trolling). 10 Trial-level patent courts have enjoyed considerable success in practitioner circles, in large part because expertjudges are better at understanding the technological complexity of the facts with which they are presented. But trial courts do not make, or definitively interpret, law. In the United States, the decision was therefore made to experiment with an appellate court with exclusive authority over patent law. Unfortunately, in its 25 years of existence, the Federal Circuit has proved somewhat disappointing. Although it has managed to make patent law more uniforn and predictable, the substance of its lawmaking has been heavily criticised."1 To many, this experience suggests that new lawmaking mechanisms are needed. Because patents are issued only after applications are reviewed by technically trained examiners presumably familiar with both the applicable law and relevant scientific developments, there is growing interest in placing primary responsibility for administering patent law in the agencies that issue patents. For example, the US Patent and Trademark Office (US PTO) has enacted a series of guidelines on both procedural and substantive issues.' 2 Similarly, the European Patent Office (EPO) 13 I Cf D R Cahoy 'An incrementalist approach to patent reform policy' (2005-2006) has become a focal point for studying the impact of new technologies and business trends. 14 Indeed, in a move that might ultimately lead to an alternative international patent law regime, the US PTO, the EPO, and the Japanese Patent Office GPO) meet regularly to harmonise their approaches to patentability determinations, with an eye toward building worksharing arrangements.' 5 There is now a growing literature on whether these agencies are capable of becoming the architects of patent jurisprudence; on the kinds of changes that would have to be made to give them the right mix of expertise; and most importantly, on how to ensure that they maintain a broad view of the factors that promote innovation, do not become subject to capture, and act in accordance with recoguised standards of procedural and substantive fairness. 16 Given these developments on the national and regional stage, the time is ripe to consider whether patents and other intellectual property should also be regulated in this manner at the international levelby a specialised body capable of tracking developing technologies, recognising emerging problems, and marshalling economic, scientific, and legal expertise. Expertise at the global stage would not only enhance the flexibility member states enjoy; it could also provide them with guidance on how to structure their domestic regimes. Two candidates recommend themselves as potential regulators. One is the World Intellectual Property Organisation (WIPO). WIPO administers the patent, trademark, and copyright conventions on which the TRIPS Agreement is based and it provides support to the WTO and its members.' 7 The other is the Council for (Fed. Cir.) (agreeing, but not deferring to, utility guidelines); Tafas v Dudas 2007 (511) F. Supp. 2d 652 (E.D. Va.) (preliminarily enjoining continuation rules on the ground that there is substantial likelihood that their promulgation exceeded the US PTO's authority).

13 The EPO administers the European Patent Convention (EPC). The EPC is not related to the European Communities, although all of the members of the European Union are members of the EPC. 14 See, for example, EPO Conference on the Patentability of Biotechnology, available at http://www.epo.org/about-us/events/biotechnology.html; European Patent Office (EPO) Scenarios for the Future -How Might IP Regimes Evolve by 2025? What Global Legitimacy Might Such Regimes Have? (2007), available at http://www.epo.otg/topics/patent-system/scenarios-for-the-future. html [hereinafter: EPO Scenarios].

I-See, for example, the website of the Trilateral Cooperation at http://www.trilateral.net/. See generally, J G Mills III 'A transnational patent convention for the acquisition and enforcement of international patent rights' (2006) 88 Journal of the Patent & Trademark Office Society 958. See also P Drahos Trust Me: Patent Offices in Developing Countries (2007) available at http://ssm.com/abstract=1028676 (noting the influence of the trilaterals on other nations' patent offices and suggesting that patent offices are playing increasingly important roles in negotiating certain treaties). 16 See, for example, C A Nard & J F Duffy 'Rethinking patent law's uniformity principle' (2007) 101 Northwestern University Law Review 1619; A K Rai 'Engaging Facts and Policy: A Multi-institutional Approach to Patent System Reform' (2003) 103 Columbia Law Review 1035. 17 See, for example, Articles 63.2 & 68 ofTRIPS Agreement; Agreement between the World Intellectual Property Organisation and the World Trade Organisation (22 December 1995), available at http://www.wipo.int/treaties/en/agreement/pdf/trtdocs-woO30.pdf WIPO TRIPS, which is charged with monitoring the operation of the Agreement.' 8 Among other things, the Council was entrusted in the Doha Round with finding a solution to the essential medicines problem. 19 Delegating authority to either of these 'agencies' raises many questions. The consensus rulemaking of the WTO, while ineffectual at keeping law current, does assure that its actions are aligned with the interests of its members. A move to agency rulemaking would need to be accompanied by procedures that protect the sovereign interests of the parties. Furthermore, while the confluence of interests in the North and the South suggests that the interest in flexible intellectual property laws are wellrepresented in these fora, both WIPO and the TRIPS Council have been criticised for a lack of transparency and participation. 20 Thus, as with national systems, the institutional design must also include assurances of good administrative practicethat the agency has the requisite expertise, is open to input from all relevant stakeholders, reaches reasoned decisions, and is free of corruption. 21 Given the multiplicity and overlapping interests of intergovernmental organisations, new rules will also be needed to allocate jurisdiction both horizontally (between agencies) and vertically (between agencies and members) and to protect these lawmaking authorities from opportunistic practices, particularly forum shopping.

Using patent law as its focus, Part II provides an overview of the problems confronting producers and utilisers of protected materials, and demonstrates that the debates over international intellectual property protection should not be viewed simply as a fight between the North and the Souththat in fact, there are many commonalities between the interests of developed and developing countries. This part then moves on to examine the ways that the WTO system currently permits nations across the development spectrum to accommodate their interests. Part III assesses the relative advantages of channelling lawmaking to WIPO or residing greater authority in the TRIPS Council. Part IV concludes with the lessons that the intellectual property story teaches about global governance and the need for international administrative norms.

Although much of the critique of the TRIPS Agreement focuses on its differential impact on developing countries, it is becoming increasingly clear that the enforceable obligations that it imposes have negative effects in the developed world as well. Although the problems are hardly comparable in terms of their magnitude and impact on human welfare, the alignments and cross linkages among interests at both ends of the development spectrum are nonetheless worth exploring. They demonstrate where modifications in the international regulation of intellectual property are needed, point to the forces that can be harnessed to effectuate change, and suggest how the international intellectual property system might be reconfigured.

The problems that developing countries are experiencing with TRIPS are well recognised. As with the WTO as a whole, one set of problems centres on voiceon the difficulty that countries with fewer resources and poorer markets encounter when they seek to participate in international trade negotiations, and on the failure of the negotiators to take account of their perceptions, traditions, and experience. Thus, during the TRIPS drafting process, compromises were made to assuage strong concerns by the South about the value of intellectual property protection to their economies. The TRIPS Agreement begins with statements about 'Objectives' and 'Principles,' which acknowledge the goal of'promoting technological innovation.., to the mutual advantage of producers and users' 22 and an understanding that members can modify their laws to 'promote the public interest in sectors of vital importance.' 23 In addition, the Agreement includes its vaunted 'flexibilities', along with transition provisions and promises of technical assistance and technology transfer. 24 Furthermore, representations were made that TRIPS would act, at least in certain respects, as a ceiling to demands by the Norththat TRIPS would, in effect, replace the unilateral trade sanctions previously used by the United States to 'encourage' countries strongly to protect intellectual property within their borders. 25 In reality, however, these assurances have come to very little. The transition provisions, which were meant to be temporary, existed for far too short a period. 26 Of the provisions that were intended to be permanent, many are proving illusory. 27 And as the essential medicines debate showed, TRIPS was crafted without sufficient understanding of the challenges developing countries face. Thus, while the negotiators recognised that there would be occasions when compulsory licensing would be necessary, the essential medicines crisis arose, in part, because they overlooked the possibility that some countries would have such inadequate manufacturing capacity that they would be unable to satisfy their own needs. Accordingly, the Agreement permitted members to authorise compulsory licences, but only when the goal was 'predominantly for the supply of the domestic market of the Member authorising [the] use.' 28 To a great extent, however, the problems that developing countries have with the TRIPS are very much ones of substance. For patents, the TRIPS Agreement is beneficial only to countries that can either invent at world levels -and can therefore enjoy the benefits that derive from encouraging innovationor make enough use of the marketing opportunities created by the other WTO agreements to compensate for the tax that patents impose on local consumption. For truly under-developed economies, neither of these conditions is met. For these, the requirements in the Agreement largely serve to extract wealth from local populations and, to the extent that patents cover educational inputs, to make it more expensive for inhabitants to learn enough to advance to the technological frontier. 29 To be sure, there are some nations that were on the cusp of becoming inventive at the time TRIPS entered into force. TRIPS incentives may well be helping to turn these countries into innovative forces. 30 But even for these economies, there is substantial question whether TRIPS' requirements are flexible enough to foster incipient creativity.

To make matters worse, the costs associated with patenting are largely uncontrolled. Prior to TRIPS, many countries did not have a tradition of protecting knowledge products, and thus had no occasion to develop rules to safeguard the public interest. Since many continue to lack competition laws, patent holders can often wield more power in poor countries than they are permitted to exercise in rich ones. Supporting a patent system is also expensive: countries are required to establish and maintain examination officesand must use some of their best-trained scientists to staff them. 3 2

(2) TRIPS in developed countries Somewhat surprisingly, countries that are net inforniation exporters are also finding that the overall impact of the TRIPS Agreement is suboptimal. Of course, many current patent holders are thriving from the added revenue available from WTO-wide protection. However, the dynamics of information production are changing in ways that put pressure on the existing system and TRIPS makes revision difficult. Patent law was, after all, developed at the time of the Industrial Revolution, when fundamental science was perceived as very different from technological application, when patents covered end products that were generally freestanding, and when products were usually associated with only one or a few patents. In that environment, the law could distinguish between ideas (unpatentable) and their application (patentable) and the scope of patents could usually be readily defined. Inventors were protected from free-riding, but the system also fostered a rich public domain of scientific knowledge.

In the knowledge-based economy, the situation is very different. New scientific enterprises emerged and patent law grew to encompass them. In many of these fields, the old rules no longer protect the public interest effectively. In biotechnology and computer science, for example, science and technology have coalesced, so that a patent on a product will tend to cover the scientific advance that made the product possible. 34 If the scope of infringement is broad, then second-generation innovators cannot freely make further use of the underlying idea to push the knowledge frontier forward. Products in the information sector are often composed of many separately patented features and are meant to interoperate with advances that are also subject to multiple patent rights. With many patents required to make novel products and pursue new technological opportunities, transaction costs rise and the probability of hold outs increases. 35 In addition, the business of science is changing. Universities are aggressively pursuing patent protection for foundational advances that, in an earlier era, would have entered the public domain and furnished researchers with a base of inventive opportunities. The shift from curiosity-to business-driven research also affects the ethos within the innovation sector. Some academic researchers have gone into business for themselves, spinning off 'boutiques' that look to patents as a way to attract venture capital and joint venturers. 36 'Open' collaborative communities, built on sharing norms, are discouraged. 37 With fewer information exchanges and less tolerance for unauthorised experimentation, there is increasing risk that the pace of innovation will slow. 38 A vicious cycle can 33 See generally, R Dreyfuss 'Protecting the public domain of science: Has the time for an experimental use defense arrived?' (2004) 46 Arizona Law Review 457.

14 Both products and processes can be patented. This article uses the term 'product' to encompass both unless otherwise indicated. 35 eventually take hold: when some participants in an industry acquire patent protection, others begin to patent defensively in order to acquire the bargaining chips they will need to settle lawsuits. With more patenting activity, new business models develop. So-called 'trolls'non-market participantsbuy up patents and, free from the constraints of the market, use their positions to threaten ongoing businesses. The explosion in patenting puts pressure on even well-financed national patent offices, leading to patents on developments that should have been freely available. These low-quality patents further increase transaction costs, create barriers to entry, and breed even more need for patent protection. 39 With higher costs, follow-on innovation is discouraged.

(3) Achieving balance While it may initially appear that countries at different ends of the development spectrum experience TRIPS in radically different ways, there is, in fact, a great deal of commonality. In both cases, the Agreement's focus on commodificationon providing a return for initial innovatorsignores the real goal: improving social welfare. Instead of providing assurances that inventions will be optimally utilised, the Agreement fosters deadweight lossfor some countries, in the form of foregone consumption and for others, in the form of lost opportunities for follow-on production. Furthermore, the ways to revise patent law to mitigate these problems are quite similar: places where patenting is particularly dysfunctional could be excluded from protection (medicines in the South; fundamental discoveries in the North); new privileges could be recognised (compulsory licensing in the South, experimental usages in the North); and remedies could be altered (replacing injunctive relief with damages would reduce market manipulation in both the South and the North). Finally, striking a new balance in any of these (or other) ways is proving equally difficult for both the South and the North, and at both the national and international levels.

In some ways, developing countries are in a good position to maintain laws that are responsive to the needs of both information producers and users. For many, TRIPS was a hard sell to begin with and TRIPS sceptics still tend to outstrip its supporters. Thus, there is considerable political will declining willingness to transfer materials); J P Walsh, A Arora & W M Cohen 'Effects of research tool patents and licensing on biomedical innovation in patents in the knowledge-based economy' in W M Cohen & S A Merrill Patents in the Knowledge-Based Economy (2003).

to respond to local needs by adopting the lowest level of protectionism 40 that TRIPS allows. Working through the TRIPS flexibilities is not, however, an easy task. Countries, such as India, that were at the cusp of development at the time of the Uruguay Round, may be in a good position to accomplish it,41 but many WTO members lack the incentive to put intellectual property lawmaking at the top of their legislative agendas. For those new to an intellectual property regime, there may also be inadequate legal infrastructure to accomplish the task effectively. To be sure, TRIPS promised developing countries technical assistance. 42 But some countries have complained that what they mainly receive is advice to adopt the same laws found in the country providing the help. 43 In developed countries, the situation is somewhat reversed. Legal talent is readily available and there is adequate legislative interest; however, the political economy is vastly more complicated. Traditionally, intellectual property users have been badly organised and, relative to right holders, poorly financed. 44 Technology producers are in an almost equally bad position to institute needed legal revisions. The impact of change is far from uniform across fields. 45 Accordingly, while some sectors (e.g. information technology) feel strongly that patent reform is necessary, others (the pharmaceutical industry) vigorously defend the current system. Although a mixed system is the obvious solution, TRIPS includes a ban on discrimination by field of technology, 46 and high protectionists claim that any deviation from a 'one size fits all' patent system would violate that provision. 47 Even at the theoretical level, there is sharp protectionist measures. However, to TRIPS negotiators, intellectual property protection is good because it commodities information and makes it tradeable. In this article 'protectionism' is used in the intellectual property sense of the term. disagreement. For example, the EPO recently published a report projecting future developments in the patent industries. It came in the form of four scenarios that reflect the uncertainty in how recent trends will play out and disagreement about what legal actions are needed to ameliorate the problems that are likely to emerge. 48 The United States has also experienced a problem that other democracies may well encounter. Intellectual property legislation tends to proceed through a series of compromises. In a challenge (if one is possible at all), domestic courts will tend to review the resulting legislation on a deferential standard. In the WTO, however, impingements on exclusive rights receive fairly strict scrutiny. Because of the dichotomous standard of review, deals are vulnerable to unravellingalways in the direction of stronger protection. For example, in 1998 US copyright holders succeeded in extending the term of copyright protection, but at the same time, relinquished the right to control musical performances in certain settings (the so-called 'Irish bar' exemption, named after one of the locations covered by the measure). Both parts of the legislation were challenged: term extension went to the Supreme Court, which upheld the statute, while a WTO panel decided that the Irish bar exemption violated the TRIPS Agreement. 49 Were the United States to comply with the panel report, the effect would be to increase the level of protection and upset the legislatively-set balance between right holders and the public.

To some extent, the problems encountered at the national stage might be solved at the international level. Indeed, the WTO system has been described as democracy-enhancing because it operates as a pre-comirrutment strategy that reduces rent seeking by powerful interest groups. 50 In the context of TRIPS, the argument is that by reducing the force of short-term concerns about the availability of intellectual goods for both consumption and production, the Agreement allows members to capture the long-term benefits of fostering the discovery of new knowledge. 5 ' If there is anything to this argument, then one way to overcome legal infrastructure deficits in the South and political economy problems in the 48 EPO Scenarios (n 14). 41 North would be to revise the TRIPS Agreement to better reflect the need for balance and to impose balancing obligations (or obligations that achieve balance) on member states. At the very least, the DSU could be producing interpretations of the Agreement that lay out the normative options and clarify the scope of existing flexibilities. Armed with a better understanding of permissible alternatives, reformers would have an easier time in the domestic arena.

Somewhat paradoxically, it is likely that the problems of developing countries are more amenable to accommodation through TRIPS revision than are the difficulties encountered in developed countries. One development issueaccess to essential medicinesis well understood and has attracted both attention and sympathy. Although it took rather longer than expected, the TRIPS Agreement is being amended to permit nations that lack manufacturing capacity to look elsewhere for the products they need. 52 This is not to say that in the patent area, all development problems have been solved. Medicines are not the only patented products where access is an issue and copyrighted goods can raise equally strong concerns. Moreover, as India is demonstrating, the road to technological capability passes through a stage of 'fair following', where skills are learned by adapting the work of others. 53 To accommodate that kind of activity, nations must be permitted to set either the scope of infringement liability low enough to allow adaptation or the standard of patentability high enough to put some inventions that are commercially worth fair following into the public domain. Alternatively, they need authority to award compulsory licences to local industry.

Developing nations appear to have some capacity to instigate these sorts of changes. Individually, they may have less leverage in international intellectual property negotiations than developed countries, but their clear preferences for low levels of protection allow them to take unified positions. In WIPO, for example, they succeeded in diminishing the protectionist bite of various intellectual property instruments and in promoting a development agenda. 5 4 Indeed, as Susan Sell tells it, one 52 See Doha Declaration (n 4) and Declaration on Public Health (n 4). s3 'Fair following' refers to utilisation of prior advances in ways that do not entail free riding, and which therefore do not threaten incentives to innovate. reason international intellectual property lawmaking shifted from its traditional home at WIPO to the General Agreement on Tariffs and Trade (GATT) negotiations, and then to the WTO, was because attempts to raise the level of international protection in WIPO had, essentially, stalled.

Refashioning TRIPS to accommodate the new needs of developed countries is likely to prove even more difficult. Those who are interested in revising the terms of protection must contend with the same highly organised and well-heeled rent seeking forces that pushed intellectual property negotiations to the WTO. Furthermore, governments in the North have reasons of their own to prefer strong worldwide protection: intellectual property exports bring in revenue, furnish a basis for income taxation, and offset trade deficits. For example, knowledge goods are a key contributor to the US balance of trade. 5 6 Even if developed countries were to recognise that the problems they are experiencing in fostering creative production require amendment of the TRIPS Agreement, it would be difficult to implement the insight. True, the technological neutrality principle, which is complicating domestic negotiations in the newly-complex patent sector, could be modified. However, the critical problem for developed economies is the Agreement's failure to balance the interests of the first generation of producers against those of follow-on creators. Rectifying this problem would require the addition of 'substantive maxima' or 'user rights' that would prevent members from raising the level of protection to the point where it threatens robust creative development. 5 7 But much as such a Convention (also known as the Berne Appendix) is an example of international lawmaking specifically directed at problems of development. showing that the US, Germany, France, Italy, Sweden, and Switzerland are the only beneficiaries of the TRIPS Agreement and that significant losers include Canada, Brazil, UK, India, Mexico,Japan, Spain, and Korea), available at http://www.sciencedirect.com/science?_ob=ArticleURL& udi=B6V6D-43F70T3-8&_user=30681 &_coverDate = 10%2F31%2F2001 &_rdoc= l&_ fnt=fufll&_orig=na&_cdi=5812& docanchor=&_acct=C00000333&_version

= 1 &_ urlVersion=0& userid=30681&md5=3055d43b997a8827179a8fd862dI 03f6&artImgPref=F. See also CIPR Report (n 27) 24: 'Between 1991 and 2001, the net US surplus of royalties and fees (which mainly relate to IP transactions) increased from $14 billion to over $22 billion'.

-1 See G B Dinwoodie 'The international intellectual property law system: new actors, new institutions, new sources' ( 2004 system would add a pleasing symmetry to the Agreement, a working group at the Max Planck Institute has identified a key structural problem. TRIPS is formulated to safeguard the interests of foreign traders; it does not protect domestic consumers and creative users from their own governments. Thus, while user rights might be useful indirectly to inform the scope of members' flexibilities, 5 8 they could not be enforced directly. 59 (ii) Interpretive guidance through the DSU Given the remote prospect of TRIPS revision, a more likely vehicle for accommodating emerging concerns is through dispute settlement. If DSU panels and the Appellate Body were to clarify the choices available through the TRIPS flexibilities, members would understand their options. For countries where legal advice is deficient, a roadmap would be highly desirable. In countries with complex political economies, clarification might simplify the political debate. It would make it clear where TRIPS roadblocks lie, eliminate rhetorical appeals to supposed TRIPS prohibitions, and provide some assurance that political compromises will not unravel. 60 In fact, the TRIPS Agreement would appear to leave members in both the South and the North with sufficient room to respond to domestic concerns. As noted earlier, the Objectives and Principles that introduce TRIPS obligations explicitly acknowledge the trade-offs inherent in protecting intellectual property. 61 For example, the patent provisions in the Agreement include a host of flexibilities. The size of the inventive step '8 For an example of indirect enforcement, see text at note 77, infra.

Paradigms in IP Law: A Project Reform of TRIPS (5 Sept 2006) (spearheading a proposal to amend the TRIPS Agreement). To see the problem, consider, for example an amendment to the Agreement requiring members to recognise an experimental use defence peruitting universities to conduct research on patented materials without authorisation. Xandia (a WTO member) refuses to enact such a provision and Xandian research universities are successfully sued for infringement. What would be their remedy? TRIPS is not generally interpreted as selfexecuting. See, for example, TRIPS, Article 1 ('Members shall give effect to the provisions of this Agreement'). Since the universities' quarrel is with Xandia, there is nothing in the DSU portion of the WTO Agreement that provides it with recourse. Other nations may complain that they are deprived of the fruits of the Xandian research. While a slowdown in world-wide innovation is certainly a global concern, it would likely be considered too speculative (or insufficiently related to trade) to furnish the basis for a complaint in the WTO.

Of course, the DSU could be amended to allow non-government entities in WTO countries to bring challenges. However, scholars who have studied the idea tend to reject it, in part because it would exacerbate the asymmetry between the lawmaking capacity of the dispute settlement procedure and the legislative capacity of the WTO. is not set; 6 2 presumably countries (such as India) with an interest in fair following can define the step to be quite high. Similarly, the Agreement does not specify what it means by 'making, using, offering for sale, selling or importing', 63 which should give nations some leeway in defining the scope of infringement. Members can also exclude from patentability specific classes of inventions, 64 as well as any advance that is needed to protect public order or morality. 65 Under highly specified circumstances, including blocking patent problems, members can permit unauthorised (but usually, compensated) uses. 66 And most importantly, a three-part 'exceptions' provision permits members to enact (1) 'limited' exemptions so long as they (2) 'do not unreasonably conflict with a normal exploitation' and (3) 'do not unreasonably prejudice the legitimate interest of the patent owner, taking account of the legitimate interests of third parties'. 67 Unfortunately, however, the prospects forjurisprudential development do not appear encouraging. Graeme Dinwoodie and I studied the six intellectual property disputes that have been taken as far as a panel decision (half ultimately resolved by the Appellate Body) 6 8 and we found little sensitivity to the need for balance, for law that reflects the dynamics of intellectual production, or for the demands of the political process in which national lawmaking proceeds. 69 The three panel reports interpreting the 'exceptions' tests are by far the most troubling. These include US-110(5), the above-mentioned 'Irish bar' case, which interpreted the copyright analogue to the three-part patent provision; Canada-Pharmaceuticals, which applied the patent test to a Canadian law permitting generic drug manufacturers to test drugs for regulatory review purposes and to stockpile inventory prior to patent expiration; and EC-GI, which applied a somewhat different trademark 'exceptions' test to an EC law recognising geographic indications (GIs) that impinge on trademark exclusivity. Of the three challenged statutes, only the EC's was found filly compliant with the restrictions in an exceptions provision. In the other two cases, the panels appeared to see the issues through the lens that TRIPS createsas purely a problem ofcommodification.

To start, the panels essentially decided that the reasons underlying the challenged legislation are of little relevance. On the copyright side, the three-part test refers to 'special cases,' which suggests that a member could defend its action with a compellingjustification. But the US-110(5) panel did not see things that way: it held that 'special' means 'clearly defined'. 70 Under that view, even the US fair use provision, 7 ' which has strong constitutional underpinnings, may not be defensible: free speech may feel special to the United States, but it is hardly clearly defined. For patents and trademarks, the situation may appear somewhat better because the exceptions tests make express reference to the interests of third partiespresumably, in the case of patents, these would include user groups such as patients interested in therapeutic use of medical inventions and research scientists interested in experimentation involving patented inventions. 72 The panels, however, held that the individual parts of the test were to be considered cumulatively. 73 For patents, this means that if a measure runs afoul of part (1) ('limited' exemption) or part (2) ('unreasonably conflict with normal exploitation'), the user-based justification will never come into play. In Canada-Pharmaceuticals, for example, the panel did not reach the question whether patients would benefit from the stockpiling exemption. Rather, it held the provision, which permitted stockpiling for only the last six months of the patent term, was nonetheless not 'limited' because during that time, three of the five exclusive rights associated with a patent were affected.

Admittedly, this interpretation of the exceptions tests would not be critical if adjudicators looked at the concepts 'limited', 'normal exploitation', 'legitimate interest,' and 'unreasonable prejudice' in a nuanced way. After all, these terms appear to lend themselves to normative interpretation. Adjudicators could construe them in light of the core WTO goal of maintaining an international trading system for intellectual products or in view of the stated objectives of the TRIPS Agreement. 74 But the panels 70 US-110(5) (n 49), paras 6.107-6.110. 71 Section 107 of17 U.S.C. 72 Canada-Pharmaceuticals (n 68) para 7.69. 73 See, for example, ibid at para 7.20. 74 See, for example, J C Ginsburg 'Toward supranational copyright law? The WTO panel decision and the 'three step test' for copyright exemptions' (2001) 187 Revue Internationale du Droit d'Auteur 17.

have not been willing to venture beyond the literal wording of the tests. Indeed, the Canada-Pharmaceuticals panel was wary of considering the Objectives of the Agreement, arguing that it amounted to renegotiating TRIPS' 'basic balance'.7 5 Mostly, the panels interpreted the terms by counting the number of rights within the patent or copyright bundle affected by the challenged provision, by considering the number of situations where the exception was applicable, and by examining potential economic losses. Every right received equal weight, no matter how small the impact on the right holders' market; markets the right holder had never utilised were counted equivalently to those that it had. Monetary estimates were extremely generous. 76 On this approach, only two provisions passed muster. In EC-GI, where the trademark exceptions provision gave a specific example of an acceptable exemption ('fair use of descriptive terms'), the panel found for the EC. 77 The Canadian law permitting generic drug testing for regulatory review purposes was upheld, largely because the effect of that exemption was to reduce the amount of lead time that patent holders would otherwise enjoy after patent expiration, and the panel considered lead-time exclusivity peripheral to the patent rights secured by TRIPS . 78 The rigorous review of these exemptions leaves nations with little room to accommodate their varied interests. The cumulative nature of the tests poses a high bar, hampering legal evolution. Safeguarding public interests will be difficult because it is the rare case where there is room to manoeuvre at the periphery of the patent right. 79 Developing countries cannot defend exemptions that intrude upon a substantial number of protected rights, no matter how little profit the right holder could hope to extract from their limited markets. Nor can they use the compelling nature of their justifications to offset intrusions into intellectual property prerogatives.

76 For example, the US-110(5) panel found that EC copyright holders could lose as much as $53.65 million per year. When the United States failed to conform its law to the panel report, the case went to arbitration and the arbitral award was only C1,219,900 per year (this at a time when a dollar and a euro were close to parity). See Recourse to Arbitration under Article 25 of the DSU, Award of the Arbitrators, United States -Section 110(5) of the US Copyright Act, 5.1, WT/DS160/ARB25/1 (11 November 2001).

EC-GI (n 68) para 7.654. 78 Canada-Pharmaceuticals (n 68) para 7.57 ('The additional period of market exclusivity in this situation is not a natural or normal consequence of enforcing patent rights').

71 In this regard, there is irony in the Canada-Pharniaceuticals decision. While the panel failed to understand members' need to accommodate future developments, it stressed the need of patent holders to alter their strategies to evolving market conditions. See (n 68) at para 7.55: 'The specific forms of patent exploitation are not static, of course, for to be effective exploitation must adapt to changing forms of competition due to technological development and the evolution of marketing practices'.

The patent case added yet another constraint on members' freedom of action. As noted earlier, patents are also subject to a requirement of non-discrimination by field of technology. 80 While it could be argued that a member enacting a technology-specific measure can rely on the exceptions provision to defend against a TRIPS challenge, the Canada-Pharmaceuticals panel treated non-discrimination as structural, as an issue that is not subject to exceptions. 8 1 In other words, it interpreted the provisionwhich is only found in the patent section of TRIPSas the equivalent of the commitments to most favoured nation and national treatment, which infuse the entire WTO framework. This interpretation is especially problematical because a law that is broad enough to encompass all fields of technology is unlikely to simultaneously meet the 'limited' requirement of the exceptions test. Accordingly, it may be difficult to devise patent regimes that solve complicated domestic economy issues or meet the needs of both traditional and emerging technologies.

The TRIPS cases are also troubling for things the TRIPS adjudicators did not do. They gave no consideration to the political dimension of the problem of complying with TRIPS. For example, members in transition were required to ensure that pharmaceuticals invented after TRIPS came into force would be protected once compliant patent law was enacted. 83 To circumvent problems associated with adopting protectionist legislation in a low-protectionist climate, India fulfilled this obligation with an administrative ('municipal'), rather than a legislative, scheme. In India-Pharmaceuticals, however, that approach was successfully challenged. Although the Appellate Body acknowledged that members are permitted to choose their methods of compliance, it nonetheless retained the power to make the ultimate decision and held that India's method fell short. 84 As noted earlier, the US-110(5) case displayed a similar disregard for the political compromise that enabled Congress to extend the copyright term.

The adjudicators were equally unwilling to consider the trade implications of the challenged legislation. Thus, while some of the challenged national measures could have undermined international trade in intellectual goods because they presented strong arbitrage possibilities, others were so purely consumptive that their threat to trade was minimal. Yet none of the reports drew this distinction. For example, because the drugs stockpiled in Canada could have been exported and affected pharmaceutical prices across the WTO, the Canada-Pharmaceuticals result makes sense as a matter of protecting the trading environment. In contrast, musical performances in US Irish bars are entirely consumptive and thus pose much less danger to trade. Certainly, they potentially reduce the rewards that EC right holders can expect in the US market, and can therefore erode the comparative advantage that the EC enjoys in producing music. The EC was thus not wrong to challenge the US measure. But the US-110(5) panel failed to even note that the harm was confined to US territory. This omission suggests that the South could not enact laws that, say, permit translations of copyrighted materials into local dialects, and defend the legislation on the ground that it has minimal implications for international trade.

The adjudicators also failed to notice important differences between TRIPS and the rest of the WTO Agreement. In the GATT, members benefit from cushioning provisions that permit them to deal specially with issues of overarching national importance; 86 reduce trade concessions in certain circumstances; 87 and suspend obligations or withdraw concessions on a temporary basis. 88 Had adjudicators considered the inapplicability of these provisions to TRIPS, they might have taken a more generous view of the flexibilities that TRIPS does provide. Admittedly, the GATT measures have less significance now that the GATT portion of the Agreement is firmly established. But TRIPS is not established, at least not in developing countries. And as we have seen, the dynamic nature of intellectual property production means that its obligations will raise continuous problems even in developed economies. 89 As the locus of international intellectual property lawmaking, dispute settlement has other limitations. Despite its 'legalisation', there is still a diplomatic flavour to the proceedings that is somewhat inconsistent with the notion of expounding law. For example, aside from India-Pharmaceuticals, all of the intellectual property cases decided to date were, in a sense, a draw: the complainant won on one issue and the respondent won on 86 Article XX of GATT. See P Lamy 'The place of the WTO and its law in the international legal order' (2006) 17 European Journal of International Law 969 at 978 (noting that Article XX permits nations to set aside market access regulations when 'considerations other than those of trade must prevail').

8 7 Article XXVII of GATT. 88 Article XIX of GATT and its associated Agreement on Safeguards. 89 Esty suggests that nations are cushioned because they can pay compensation in lieu of compliance, Esty (n 2) 1536. Indeed, the US-110(5) case ended in exactly this way, with the US paying the EC yearly to compensate for unauthorised performances. However, this approach works only for countries that have the means to pay. Thus, it tends to exacerbate tension among countries at different stages of development.

another. To produce these outcomes, the reports took some fairly far-fetched positions. In US-110(5), for example, the EC had also challenged an exemption on musical performances that pre-dated the Irish bar provision. That exemption was upheld on the ground that it was confined to performances of dramatic musical works (for example, opera). 90 Yet there is nothing in US copyright law suggesting this limitation. Indeed, the provision was regarded in the United States as so open-ended that the Irish bar exemption was arguably added to provide the establishments it covers with a safe harbour. Similarly, at the same time that the EC-GI panel found the conflict between geographic indications and trademarks acceptable, it invalidated a part of the GI legislation on national treatment grounds. In order to decide for the EC on the trademark issue, it interpreted the scope of the GI right far more narrowly than EC lawmakers likely envisioned. 91 Because diplomacy-oriented decisions are highly contingent on framing, they cannot offer domestic policymakers the kind of guidance they need. 92 DSU adjudication also suffers from a problem of participation. As Eyal Benvenisti has noted, WTO adjudicators have been diligent about developing administrative law that assures fair process. 93 In one way, the intellectual property cases are no exception. In EC-GI, the panel found that the EC violated its national treatment obligations because it did not provide a registration system for foreign geographic indications that is as easily utilised as the one set up for registering European indications. Similarly, in Havana Club, the Appellate Body acknowledged that ownership of trademarks was not covered by the TRIPS Agreement, yet it required the United States to furnish holders of Cuban marks with the same procedural advantages that other right holders enjoy.

At the same time, however, the cumulative effect of the DSU process is to truncate participation. In all of the cases decided to date with the exception of India-Pharmaceuticals, the respondent has been a developed country and the complainants were all developed countries. Developing countries (particularly members that are still in the transitional phase) have, in short, played a very limited role in developing TRIPS jurisprudence. Although the DSU permits non-party members to make their 9 US-110(5) (n 49) para 2.8. 9' EC-GI (n 68) para 7.659 ('the positive right to use the GI extends only to the linguistic versions that have been entered in the register and not to other names or signs which have not been registered').

92 A cynic might also wonder whether panels produce split reports to avoid appeals to the Appellate Body.

3 E Benvenisti 'The interplay between actors as a determinant of the evolution of administrative law in international institutions' (2005) 68 Law & Contemporary Problenis 319 at 332. views known, 94 and the Appellate Body famously permitted NGOs to file amicus briefs in US-Shrimp-Turtle, 95 the role of non-parties is highly circumscribed. Reduced participation by nations and entities with strong user-sympathies makes it more likely that decision-makers will interpret the Agreement with right holders' interests at the fore. While DSU reports do not strictly speaking, have stare decisis effect, 96 they are regularly cited as precedent (which, indeed, is why there are concerns about the legitimacy of WTO lawmaking). 97 Furthermore, to the extent that these decisions are not precedential (except with respect to the parties in the dispute), their force in domestic political debates is diminished.

Of course, it may be wrong to draw pessimistic conclusions from only six cases, half of which were not appealed. Much would be learned if, for example, the Appellate Body were to consider a challenge to India's position on the patentability of new uses of known substances because that legislation raises an issue concerning follow-on invention in the context of a patent law crafted to promote both innovation and development. 98 But the paucity of cases raises another significant question: will DSU jurisprudence be varied enough to provide sufficient guidance? There is a continuing moratorium on non-violation complaints regarding intellectual property, which means that none of the subtler international intellectual property issues raised by TRIPS will be expounded upon in the near future. 99 Because the dispute settlement system lacks an analogue to declaratory judgment actions, a member interested in responding to an " Articles 10 & 17.5 of DSU. 98 See section 3(d) of Indian Patents Act (barring a patent on 'the mere discovery of a new form of a known substance which does not result in the enhancement of the known efficacy of that substance or the mere discovery of any new property or new use for a known substance or of the mere use of a known process, machine or apparatus unless such known process results in a new product or employs at least one new reactant'). India's denial of a patent on Novartis' drug Glivec (Gleevac) may set the stage for such a challenge. See R Daniel 'Novartis: India High Court denies a Glivec-related challenge' (6 Aug 2007) Market Watch, available at http://www. marketwatch.com/news/story/novartis-india-high-court-denies/story.aspx?guid= %7B28DBBD2F-95A0-434B-B73A-E2066046C3D7%7D.

99 Article 64.2 of TRIPS; Ministerial Declaration, WT/MIN(05)/DEC, Article 45 (18 December 2005) (directing the TRIPS Council to consider the conditions for lifting the moratorium). The issue remains on the agenda of the TRIPS Council (see Annual Report (2006) of the Councilfor TRIPS, IP/C/44 (4 December 2006)), but as of mid-2007 no further action on this issue had been undertaken. For a discussion of the role nonviolation complaints could play in TRIPS, see Dinwoodie & Dreyfuss (n 49). Nonviolation complaints arise when a member asserts that an objective of the Agreement is being impeded as the result of any measure applied by another member state. In contrast, in a violation complaint, the claim is that a measure taken by another member nullifies or impairs a benefit accruing under the Agreement. emerging problem in a pioneering way cannot get an advanced read on whether the proposed approach is compatible with TRIPS. Most importantly, the number of intellectual property cases reaching disposition is declining. This phenomenon is particularly troubling, for it speaks not only to the sufficiency of interpretive guidance, but also suggests that the WTO may no longer occupy a central role in international intellectual property lawmaking. Unless the WTO system can become more responsive to members' interests, problems will be increasingly settled in other fora or bilaterally. Given the current tendency of bilateral agreements to increase levels of intellectual property protection, 100 the latter approach bodes poorly for the development of a well-functioning multilateral mechanism for regulating rights in knowledge products.

In some ways, the TRIPS story is a classic example of a regime in need of administration: a treaty with unanticipated consequences, entered into by nations with widely diverse needs, affecting groups whose interests are not always aligned with those of their governments, and without an effective legislative capacityli It has been suggested that in the WTO system as a whole, the Appellate Body is taking on the role of mediating among different interests, constraining the range of outcomes, and providing interpretive guidance. 10 2 But as the last section suggests, the DSU system (whatever one may think of its legitimacy) has so far proved ineffectual at overcoming the deficiencies in TRIPS. The drafters of the TRIPS Agreement incompletely theorised the functioning of exclusive rights regimes, and created an instrument that can impair consumer access to knowledge products and impede dynamic innovation. 1 0 3 TRIPS does not even convey a sense of what 'Trade Related Aspects' of Intellectual Property Rights might mean. 0 4 It is no surprise, then, that adjudicators cannot effectively delineate the space in which WTO members may operatethe Agreement does not contain judicially-manageable stan-"' See, for example, P K Yu 'Currents and crosscurrents in the international intellectual property regime' (2004) 113 Even the overview section on the WTO website discusses only short term costs and long term benefitsthere is virtually no recognition that costs can also be an issue in the long term. See http://www.wto.org/english/thewto-e/whatis-e/tif e/agnn7_e.htm.

104 Indeed, it is not clear that the drafters meant to limit the Agreement to trade-related matters. See Croome (n 25) at 215. dards relevant to the question. Nor can much insight be expected from the adjudicators themselves: although intellectual property expertise qualifies a candidate for an appointment on the Appellate Body, none of the current members of the Appellate Body has experience in intellectual property law. 105 Further, many cases are decided by panels and few individuals versed in intellectual property have been chosen to serve. 106 TRIPS does, however, include two notable features. It contemplates close cooperation with WIPO, which has considerable experience administering intellectual property instruments. Furthermore, the Agreement sets up a Council to oversee compliance, conduct negotiations on issues left open at the end of the Uruguay Round, and 'carry out such other responsibilities as assigned to it by the Members'. 0 7 The issues, then, are whether and under what conditions the expertise of either (or both) of these organisations could be exploited to rectify the deficits in TRIPS.

The problem of administering international intellectual property instruments is not of recent vintage. The earliest multinational agreements, the Paris Convention (on patents and trademarks) and the Berne Convention (on copyright), were negotiated in the late nineteenth century. Each began as a separately administered 'Union' of its members, but it soon became clear that there was enough overlap of issues (or gaps in the coverage of the conventions) to make joint administration desirable. To that end, the United International Bureaux for the Protection of Intellectual Property (BIRPI) was founded in 1893. As international intellectual property instruments and inter-government organisations interested in intellectual property proliferated during the course of the next century, BIRPI eventually found itself in a situation not too different from the Paris and Berne Unions. In 1967, it decided that a struggle for supremacy could be avoided by affiliating with the United Nations. In 1970, WIPO was established as a specialised agency of the United Nations and charged with the responsibility of administering multilateral intellectual property instruments. 1 0 8 WIPO is composed of members of the conventions it administers, other UN members that care to join and invitees. The Organisation operates through a General Assembly, a Conference (consisting of all of the members of the WIPO Convention), a Coordination Committee (composed of the Executive Committees of the principal conventions), an International Bureau (which functions as the secretariat) and a series of provisional, special, expert, and standing committees. In 1995, pursuant to provisions in the TRIPS Agreement, WIPO entered into an agreement with the WTO 'to establish a mutually supportive relationship'. 109

In some ways, WIPO is an ideal locus of international intellectual property lawmaking. Putting aside TRIPS, it now administers virtually all of the major intellectual property conventions (upwards of 20 instruments), which gives it considerable expertise in every facet of knowledge production and protection. More importantly, it regularly grapples with the kinds of problems the intellectual property community is facing. In response to technological changes, WIPO presided over the successful completion of two new agreements, the WIPO Performances and Phonograms Treaty (WPPT), and the WIPO Copyright Treaty (WCT), 110 both of which enable the creators of various intellectual goods to benefit from distribution in the markets that modem technologies make possible. Furthermore, its standing committees routinely monitor the creative environment, identify emerging issues, and make concrete proposals. For instance, soon after the Internet went into wide use, the WIPO Standing Committee on the Law of Trademarks, Industrial Designs and Geographical Indications began to consider the problem of localising trademark infringements and eventually proposed criteria helpful in deciding jurisdictional and choice of law issues in the digital environment."' WIPO is also an appropriate forum for considering the needs of developing countries. As Debora Halbert puts it, 'WIPO was born into the controversy of how intellectual property would impact the developing world': at the time of the 1967 Stockholm Conference at which the WIPO Convention was approved, a Protocol Regarding Developing Countries was adopted to broker the deal.11 2 In exchange for agreeing to WIPO control, the South was assured access to educational and cultural 5 As originally conceived, the plan was to consider the relationship between intellectual property protection and the UN Millennium Development Goals and to help implement the development-oriented provisions of the TRIPS Agreement, including operationalising TRIPS' Objectives and Principles. To these ends, WIPO created a Provisional Comnmittee on Proposals Related to a WIPO Development Agenda (PCDA). In its 2007 session, the PCDA recommended that the WIPO General Assembly set up a permanent committee, open to participation by WIPO member states as well as interested intergovernmental organisations and NGOs. 116 Not surprisingly, many of the PCDA's proposals deal specifically with the problems of development: technical assistance to craft and implement intellectual property and competition law appropriate for developing economies; aid in developing scientific and cultural institutions; and facilitation of technology transfer (both direct and in the form of exchange programmes). Significantly, however, the proposals also go one step further. Recognising that the South and the North are not necessarily antagonists in debates over intellectual property, the PCDA also includes recommendations that deal with concerns about the quality of the creative environment. Thus, the proposals speak of 'making national IP institutions more efficient' and promoting 'fair balance between IP protection and the public interest"' 1 7goals that could certainly include finding a better accommodation between first and later-generation innovators. The PCDA also recommends that WIPO undertake studies on the impact of strong intellectual property rights and the role of open collaborative projects in promoting innovation. 11 8 Even more provocatively, it proposes 'norm setting activities' that promote 'the preservation of the public domain within WIPO's normative processes and deepen the analysis of the implications and benefits of a rich and accessible public domain'."1 9 Admittedly, there is a great deal of room for scepticism regarding the future of this agenda. First, several of the items look eerily similar to the activities that WIPO undertook at the Stockholm Conference. As with the current Development Agenda, NGOs were permitted to attend the deliberations. However, many of the NGOs that participated in negotiating the Stockholm Protocol turned out to be high protectionists (broadcasting unions, industrial property organisations, publisher groups, confederations of authors and composers). 20 At the end of the day, the developed countries prevented the Protocol from becoming a part of the Berne Convention. A replacement was adopted, but it has proved too unwieldy for developing countries to utilise effectively. 121 The fact that the current Development Agenda has been in the planning stage for three years strongly suggests that it too may falter.

Second, some of the PCDA's 'development' goals are, in reality, efforts to move to even higher levels ofcommodification. One agenda item talks about protecting developing countries' 'domestic creations, innovations and inventions', 122 another calls for recognising rights over wholly new classes of information, including genetic resources, traditional knowledge, and folklore.' 23 Protecting local creativity and knowledge is, of course, an attractive idea in that it would allow the South to turn the tables and extract returns from the North. On the whole, however, these materials are not protected by existing regimes because they are not new, inventive, orin the case of genetic resourcesthe product of human ingenuity. Classifying them as intellectual property strongly suggests that WIPO's real commitment is to the complete commodification of the knowledge base. Recognising these claims as intellectual property rights would also co-opt developing countries (or, in some cases, indigenous groups within these countries) to the protectionist cause and further increase the cost of the inputs that are needed to create the next generation of innovation. 124 Finally, interspersed among the promises to safeguard access interests and the public domain is a worrisome trope. The Convention establishing WIPO states that the organisation's primary goal is to 'promote the protection of intellectual property throughout the world'. 1 25 Since WIPO was partially created to administer the Berne Convention, which was formed by authors and composers for the express purpose of safeguarding their rights, 126 it is to be expected that WIPO would start by conceptualising its role in this manner. However, it is troubling that, despite its 40-year involvement in issues of development, WIPO continues to recommend that the South adopt laws similar to those found in the North, 127 and that it persists in equating the human rights side of the intellectual property problem with the liberty interests of creators rather than with users' interests in health, culture, education, or scientific progress. 28 The commitment to intellectual property promotion also influences WIPO's work in other arenas. For example, it encourages ever-stronger protection for well known trademarks; 29 it is working on a substantive patent law treaty (SPLT) that many observers fear will further complicate the innovation landscape; 130 and its initial impulse was to create an extraordinarily protective copyright regime for the digital environment.13

This scepticism is, however, likely unwarranted. It is misleading to compare WIPO now with its early incarnations. The shift in international intellectual property protection to the WTO has left WIPO somewhat 124 See, for example, S Safrin 'Hyperownership in a time of biotechnological promise: the international conflict to control the building blocks of life' (2004) 98 American Journal of International Law 641.

12 WIPO Convention (n 108), see especially Preamble & Articles 3-4. 126 Significantly, the French term for copyright is 'droit d'auteur'. 127 See Halbert (n 43) 263-64 (citing US attempts to impose its law); 270 (citing an argument the Swiss have made suggesting that in the absence of patent protection, technology transfer would be ineffectual because the information would be transferred without the know-how to use it and noting that 'transfer' appears to refer to the transfer of the protectionist legal system).

.2 Halbert, ibid at 257. To be sure, authors' rights are also mentioned in such instruments as the Universal Declaration of Human Rights, Article 27(2). It is not, however, clear that this view should be transferred to patent rights, ormore broadlyto instruments aimed at promoting the interests of developing economies. bereft of purpose. Embracing the Development Agenda can be understood as a search for a new role. More importantly, where developing countries were once largely alone in opposing strong protection, the changes in the patenting environment have led many in the patent sectors in the North to join their cause. This trend is equally evident on the copyright side. As the cost of creating and distributing intellectual goods decreases, activists increasingly question the need for strong copyright protection. 132 Open collaboration and information exchange communities have proliferated (Wikipedia, the Creative Commons, and Linux furnish three examples). 133 Further, thanks (in part) to the aggressive enforcement tactics of the recording industry, there are now millions of consumers who have developed a visceral understanding of the problems associated with strong rights regimes. 134 The NGOs at the table are thus no longer just largely representatives of film companies, book publishers, and artists-rights groups. There are many low-protectionist advocates involved as well. 135 Finally, and perhaps most significantly, the stark dichotomy among members is changing: there are now emerging economiesa 'Middle' composed of nations like India and Brazilthat are reaping benefits from protecting intellectual property, yet continue to deal with many of the problems strong protection poses to development.

With a new coalition among the politically-solidified South, the technically-expert, legally-skilled, low protectionist reformers in the North and the emerging Middle, there is reason to believe that WIPO will actively engage in its new Agenda and that it will succeed in reconceiving intellectual property norms in light of development issues and an appreciation for the public domain. Indeed, the power of such an alignment is already evident. As Pamela Samuelson's account of the negotiations leading up to the WCT demonstrates, the high protectionist regime originally proposed by the United States and the EC to deal with the digital challenge was modified through the combined efforts of developing countries and low protectionist activists. 1 37

Given the WTO's failure to fully comprehend the impact of intellectual property on innovation and the unwillingness of adjudicators to rely on the Objectives and Principles of the TRIPS Agreement interpretively or to provide content for terms such as 'limited', 'normal exploitation', 'legitimate interest' and 'unreasonable prejudice', WIPO's new efforts could certainly be a significant help in fleshing out the TRIPS Agreement. Resort to WIPO's administrative guidance would also appear consistent with the approach taken in other WTO agreements, where principles set out by expert international bodies such as the Codex Alimentarius Comnmission or the International Standards Organisation have been incorporated and utilised by DSU adjudicators. 138 Indeed, WIPO participation may, in the end, be the only way to cope with the dynamic and increasingly complex nature of the intellectual property sector.

There is, however, a question about the legitimacy of relying on standards generated by WIPO. The membership of the WTO and WIPO are not coextensive. If, for example, WIPO's Development Agenda were to result in a new convention, a separate vote in the WTO would likely be needed to make it a part of the TRIPS Agreement. Furthermore, TRIPS' references to WIPO are not on the same footing as references to non-WTO intergovernmental organisations found in other agreements. The WTO Agreement on Technical Barriers to Trade, for example, explicitly envisions the development of international standards by non-WTO bodies and even directs members to use them.1 39 In contrast, the TRIPS Agreement is highly specific about which parts of the WIPO instruments are incorporated into the Agreement and which are subject to the DSU. 140 Significantly, the agreement between WIPO and the WTO is largely confined to securing transparency and providing technical assistance to WTO members. Although there may be instances in which WIPO developments would be considered illustrative of the terms of the TRIPS Agreement, 141 139 Article 2.4 ofTBT. 140 See, for example, Articles 9, 14, 16, 22 & 39 ofTRIPS. 141 See, for example, para 6.69 of US-110(5) (n 49):'[S]ubsequent developments [such as the WCT] may be of rather limited relevance in the light of the general rules of interpretation as embodied in the Vienna Convention. However, in our view, the wording of the WCT ... could be thought to 'diminish the rights and obligations provided in the covered agreements' in violation of the DSU.1 42 Nonetheless, there are arguably several ways that WIPO interpretations could be interpolated into the TRIPS Agreement. One method was proposed by Neil Netanel in connection with the question of how subsequent interpretations of the Berne Convention would apply to TRIPS. He suggested that TRIPS ought to be viewed as continually evolving. According to Netanel: TRIPS drafters must have been well aware [that] the Berne Convention is a dynamic instrument.., and [that] the rapid development of copyright-related technology require[s] an ongoing process of interpretation and reinterpretation within the framework that Berne sets forth. 143 Under his approach, WIPO's elucidations of the terms of the instruments it administers would be immediately incorporated into TRIPS, either because WTO members should be regarded as having agreed to an evolving interpretation of the Agreement, or because each new interpretation represents a 'subsequent agreement between the parties' within the meaning of the Vienna Convention.1 44 Of course, this approach can only go so far: it works fairly well in the context for which it was proposed because TRIPS largely incorporates the Berne Convention. 45 However, TRIPS obligations regarding patents, trade secrets, and trademarks go well beyond the provisions of the Paris Convention. Since the problems in all of the intellectual property regimes are somewhat similar, it will often be comparatively straightforward to take an approach WIPO suggests for copyright and apply it in the other areas. However, it would be hard to consider an interpretation developed in this way as representing a genuine agreement among WTO members (and especially not among members of the WTO who have not nonetheless supports ... that the Berne Union members are permitted to provide minor exceptions to the rights provided.

142 Article 3.2 of DSU. joined the WIPO conventions). Indeed, if the panel report in Canada-Pharmaceuticals provides any guidance, adjudicators are likely to take a very hard line on what constitutes the negotiated understanding or a subsequent agreement.1 46 An alternative approach would be to wait. WIPO identification of TRIPS flexibilities could be regarded as guidance to WTO members on what laws they can enact to accommodate their local situations. 147 Once members act on these recommendations, the legislation would arguably qualify as 'a subsequent practice' that establishes the meaning of the Agreement. 148 Canada-Pharmaceuticals may, however, cast doubt on that argument as well, for the panel rejected Canada's claim that other members' exemptions for regulatory review drug testing constituted a subsequent practice for interpretive purposes.1 49 Because the prospect for both of these approaches is dim, a more formal procedure for importing WIPO guidance into TRIPS is likely needed. One idea would be to amend TRIPS to expand WIPO's role in interpreting the Agreement. The system as a whole would then retain its basic structure, but (so long as WIPO continues to utilise a majority vote procedure) it would be more capable of responding to changing circumstances. Adopting this approach would, however, be very much a decision from 'behind the veil of iguorance'as the discussion ofWIPO suggests, it remains unclear whether WIPO will retain its traditional protectionist stance or become a venue for accommodating access interests. Since it is unlikely that WTO members will be willing to roll the dice and adopt such an open-ended approach, a less formal relationship would likely be found preferable. For example, the Agreement between WIPO and the TRIPS Council could be altered to expand the range of issues they consider jointly. The TRIPS Council could then be given the task of bringing desirable modifications to the attention of the WTO. i50 Indeed, some role for the TRIPS Council would be desirable in any event, for the interests of WTO and WIPO are imperfectly aligned. While there are some decisions about the creative environment that http://www.wipo.int/ip-development/en/legislative assistance/advice-trips.html.

14' Article 31(3) (b) of the Vienna Convention (n 144).

149 Canada-Pharmaceuticals (n 68) para 7.42 ('the subsequent acts by individual countries did not constitute "practice in the application of the treaty which establishes the agreement of the parties regarding its interpretation" within the meaning of Article 31(3)(b) of the Vienna Convention').

s' A similar suggestion was proposed by E Petersmann 'Challenges to the legitimacy and efficiency of the world trading system: democratic governance and competition culture in the WTO' (2004) 7Journal of International Economic Law 585 at 601. would also improve tradefor example, were WIPO to help developing countries commodify traditional knowledge there would be even more trading chips to exchangeother decisions about intellectual property could be inimical to trade. For instance, if the Development Agenda led WIPO to change course on the issue of well known marks, traders engaged in global advertising would suffer. By the same token, there are problems that could be solved in TRIPS without any need to alter the WIPO agreements. For example, the limitations imposed by the US-lO(5) case is a TRIPS, not a Berne Convention, problem. 15 1 If the WTO were to adopt the consumption/arbitrage distinction suggested earlier, there would be uses freed for public use by virtue of TRIPS that would not run foul of any WIPO instrument. Similarly, the North might be more willing to relinquish strong rights in the South ifWTO members reached an accord on parallel imports.' 52 In sum, there are problems in intellectual property where WIPO expertise would be useful, but there are also issues on which WIPO is likely to be insensitive. Building expertise in the WTO is therefore crucial.

The TRIPS Council was created when the Uruguay Round established the WTO. 153 Like the other WTO Councils, its functions are determined by the covered agreements as well as by special assignments from the General Council. 154 Under the TRIPS Agreement, the Council's duties include affording members 'the opportunity of consulting on matters relating to the trade-related aspects of intellectual property rights', furnishing assistance in the context of dispute resolution, and setting up a cooperative arrangement with WIPO.1 55 In addition, TRIPS requires the Council to accept notifications from members regarding their laws 156 and to review periodically the implementation of the Agreement. 57 At the conclusion of the Uruguay Round, several important issues were left 151 The US-110(5) case (n 68) interpreted Article 13 ofTRIPS, which is based on Article 9 of the Berne Convention. However, the three-part test of Article 9 applies only to reproductions of protected works, not to public perfornances. Article 11 of Berne protects 'public performances' and leaves it to national law to decide such issues as whether performances in Irish bars are 'public'. See S Ricketson & J C Ginsburg International Copyright and Neighboring Rights: The Berne Convention and Beyond (2005) § 12.02.

12 See Article 6 of TRIPS. This is not to say that a total ban on parallel importation is required. An agreement could, for example, take the form of prohibiting parallel importations between countries when the patentee is obliged by law in the country of export to offer its products at a price that is a specified percentage (say, 25 per cent) below the market price in the country of import.

153 Article IV(5) of the WTO Agreement. open, and the Council was charged with the responsibility of overseeing their continued consideration. These include added protection for geographic indications for wines and spirits;1 5 8 questions concerning protection for living plants and animals;' 5 9 the applicability of nonviolation complaints to the TRIPS Agreement; 160 and extra assistance for leastdeveloped countries.' 61 In 1998, the Council was also asked to consider electronic commerce in information products. 162 In the Doha Round, the TRIPS Council was assigned several additional tasks, most significantly, embarking on a development agenda.' 63 Thus, the Doha Declaration stressed the importance of interpreting TRIPS in a manner 'supportive of public health', and asked the Council to explore the relationship between the TRIPS Agreement and the Convention on Biological Diversity (CBD)(which protects genetic resources), 164 to consider the protection of traditional knowledge and folklore, to study 'relevant new developments raised by members pursuant to Article 71.1' (which requires periodic review of the Agreement), andsignificantlyto 'be guided by the objectives and principles set out in Articles 7 and 8 of the TRIPS Agreement'. In a separate statement on public health, the Council was instructed to find a solution to the essential medicines problem and to extend the transition periods for least-developed countries. 165 Membership in the Council is open to all WTO members and individual members (or, more often delegations of members) are free to bring an item up for the Council's consideration. The Council meets regularly throughout the year and its deliberations have a degree of both transparency and outside participation. Thus, it is required to issue Annual Reports and these, along with meeting minutes, working documents, and decisions are available on the WTO website. 1 66 The 2006 Annual Report indicates that the Council has granted regular observer status to the OECD, UNCTAD, WIPO and to other intergovernmental organisations with expertise on the issues the Council has been asked to consider. In addition, the Council grants ad hoc observer status to intergovernmen- tal organisations with special interest in the topics discussed at specific meetings.

The Council has made moderate headway on its assignments. As noted earlier, it has entered into a cooperative agreement with WIPO and the minutes of its meetings indicate that it has actively undertaken its oversight obligations. Notably, it has defined these obligations to include not only monitoring compliance with TRIPS' minimum standards, but also ensuring that developed countries are fulfilling their technology transfer obligations. 168 It has supervised the negotiations for a solution to the essential medicines problem and extended the relevant times for compliance with the Agreement. Its other tasksthe matter of nonviolation complaints, the G1 questions, life patenting, electronic commerce, and the broader issues in the development agendaremain under consideration. 1 69 Although the Council has generally been viewed as, essentially, the WTO's agent for carrying out specifically assigned tasks, Kal Raustiala has made the interesting argument that it is, in fact, the ideal forum for generating the guiding principles that domestic policymakers and international adjudicators currently lack. Raustiala points out that during implementation reviews, members challenge one another's legislation, consider the problems challenged measures are intended to address, argue about the meaning of TRIPS obligations, and examine ambiguities and lacunae within the Agreement.1 7 0 According to Raustiala: Interactions like these are useful because they inform other member states about substantive national law, signal potential areas of conflict and dispute, alert the queried member states to weaknesses of their national laws vis-a-vis TRIPS, and begin to create the norms and standards that will guide the work of the TRIPS Council and subsequent disputes and settlements. ' 7 ' Raustiala's vision has many attractive features. The Council's monitoring obligations and other assignments give it an important perspective on intellectual property issues in general, and, in particular, on how they play out in trade. Its authority appears to be extensive enough to deal with many of the problems members are facing. For example, its charge to 167 See review implementation of the Agreement presumably includes the power to consider whether competition (antitrust) standards should be added to the WTO framework and to study the problem of parallel importation. Its obligation to consider nonviolation complaints could enable it to find ways to shield national accommodations from ready challenge while preserving an opportunity for clarification through dispute resolution. 172 Council deliberations could also be used to compensate for the absence of declaratory judgment actions: a nation intent on adopting a new approach could vet it in the Council before paying the political price of enacting new legislation or suffering the political consequences of unravelled compromises. And as Raustiala stresses, if disputes were resolved in the Council though diplomacy, DSU adjudicators would less often be put into the position where they must create law. Concerns about legitimacy might then abate.

Attractive as this approach appears, it is not without its problems. Foremost is the question whether WTO members could rely on the norms and standards generated in the procedure Raustiala envisions. Raustiala is certainly right that issues of compliance are ironed out at TRIPS Council meetings. However, it is unclear whether an accommodation between two parties that avoids dispute resolution should be regarded as establishing a norm that binds other members. In many cases, it may be more in the nature of a waiveran agreement not to bring a complaint and instead to await future developments. Of course, as accommodations regarding a particular issue accumulate, members may begin to perceive the accommodation as a new normas a reasonable interpretation of the TRIPS flexibilities (taking into account its Objectives and Principles), or as a subsequent agreement among the parties. Nevertheless, any member that incorporated the norm into legislation would be vulnerable to a challenge from a member that did not participate in the deliberations. The risks associated with defending the legislation could diminish the impact of these 'norms' on the thinking of policymakers or the behaviour of domestic legislatures.

Of course, the Council could reduce the risk of challenge by turning aggressive interpretations into recommendations for amending the Agreement. The problem here is that this would appear to require agreement of the TRIPS Council, followed by agreement of the General Council. Since under current procedures, both Councils operate by consensus, 173 protectionist members would presumably have ample opportunity to veto proposals the Council makes to accommodate the interests of the South or the creative community. Of course, the operative word here is 'presumably'. As dissatisfaction with the current stalemate grows, there is increasing interest in moving away from consensus decision-making. Thus, Claus-Dieter Ehlermann and Loring Ehring suggest that attention be paid to other options envisioned by the WTO Agreement. 174 They point out that when consensus cannot be achieved, the Agreement permits voting and that obligations can be waived by three-fourths vote of the Ministerial Conference. 175 Further, they note that upon recommendation of a Council, an interpretation can be adopted by a three-fourths vote of the members of the Ministerial Conference or the General Council. 176 The latter routethe so-called 'authoritative' or 'definitive' interpretation mechanism -is particularly interesting in the context of TRIPS, where one of the major problems is that members and adjudicators lack standards for engaging in interest-balancing.

The Ministerial Conference and the General Council's authority to adopt authoritative interpretations is, however, limited. The WTO Agreement provides that this mechanism cannot be utilised in a manner that would undermine the amendment provisions of the Agreement. 77 Because the WTO has never pursued this route, it is unclear what this means: where interpretation ends and amendment begins, whether authoritative interpretations can alter the rights or obligations of the parties, and even whether these are questions the Appellate Body is empowered to decide. 17 8 Ehlermann and Ehring suggest that the better view is that an authoritative interpretation must 'somehow be linked to the pre-existing rules and cannot create self-standing rules'. 17 9 If they are right, that should be more than sufficient for an agreement like TRIPS, which uses highly capacious terms like 'legitimate', 'unreasonable' and 'normal.' Indeed, much could be accomplished by simply clarifying the relationships between various provisions, such as the operation of the steps in the exceptions tests and the applicability of the patent exceptions test to the provision barring technological discrimination.

Nonetheless, questions would remain. The TRIPS Council's shifting membership is unlikely to develop the kind of expertise necessary to fully grasp the complex technical questions that intellectual property presents, 177 Ibid.

'78 Ehlermann & Ehring 'Authoritative interpretation' (n 2) (also pointing out some highly technical questions). There is also a question whether this route is any more practicable than the current method of decision-making. The usual consensus rule is, in reality, a negative consensus rule: because abstentions are permissible, a high level of participation is not required. In contrast, authoritative interpretations require positive action from three-quarters of the membership of the General Council or the Ministerial Conference. the distortions that were created when intellectual property was shoehomed into a trade agreement, or the complicated political economy in which the system currently operates. Nor is it presently situated to get the input it needs. An early draft of TRIPS contemplated the creation of an Expert Group to advise what was then called the 'Committee on TRIPS'. However, that proposal was rejected. 180 Instead, the Agreement permits the Council to 'consult with and seek information from any source it deems appropriate' to carry out its obligations. 181 Ad hoc advice is not, however, as valuable as internal expertise because it is not available until the Council recognises that advice is needed. Here, then, is a place where closer relations with WIPO could help. The agendas of the two organisations are similar: both are considering protection for the knowledge products of the South and putting teeth into the Objectives and Principles that introduce the TRIPS Agreement. As noted earlier, WIPO already has considerable experience regarding many of the key issues. At present, however, WIPO enjoys only observer status at Council meetings, except in the specific areas where it has entered into a working relationship with the Council.

But even with WIPO's participation, there will be lingering concerns about voice. The outcomes of TRIPS Council deliberations are reported, but the meetings are held in secret. 182 Although the Council regularly gives observer status to intergovernmental organisations, none is permitted to participate in deliberations. And NGOs are not even allowed to attend meetings. They often learn from participants what will be covered and put their views into open letters, but that process is highly informal and episodic. 183 While it is may be true that low protectionists can sometimes count on the South to pursue their interests, this is unlikely to be sufficient in all situations. Developing countries have complex motivations in WTO negotiations, and those that lack creative sectors are unlikely to be familiar with the impact of strong protection on innovation or to have reason to fight for innovator-regarding safeguards. Again, some of these problems could be ameliorated by a carefully structured relationship with WIPO, where it appears that a more open process is developing and a coalition among low protectionists, developing countries, and emerging economies is building. But until WIPO actually acts on its ambitious new agenda, many observers will remain sceptical about the adequacy of its representation.

Possibly, dispute resolution could be a legitimating mechanism in this regard. After all, a case challenging a norm developed in the TRIPS Council or the legitimacy of an authoritative interpretation, would presumably raise these governance issues. 1 84 As previously suggested, adjudicators have been fairly attentive to issues of process. 18 5 Were the Appellate Body to decide that norm-creation and authoritative interpretations are valid only if they are arrived at through reasoned decisionmaking in the context of a procedure that is fair to all stakeholders, the TRIPS Council would have incentives to develop the appropriate procedures. Council-developed intellectual property norms and authoritative interpretations could, in short, not only bring the TRIPS Agreement into better alignment with the public interest, they could provide the WTO with an occasion to consider the institutions needed for effective global governance and the opportunity to promulgate administrative principles for these institutions to follow.

This account of developments in international intellectual property can be viewed as a case study in global governance. Centuries of trade in intellectual goods revealed a need for intergovernmental coordination of the worldwide marketplace. As trade increased, the value of minimum standards of protection also became apparent. 18 6 The WTO was an attractive option for continued international lawmaking, first because a protectionist agenda could be more easily pursued in a context in which trade offs and side payments could be made, and second because its obligations are enforceable through the DSU. 187 But the shift to the WTO has taken its toll both substantively and procedurally. Classically, the private returns from patenting represented only a portion of the social benefits of innovation. 188 Commodification was imperfect, so advances leaked into a publicly-accessible space where they modulated the effect of supra-competitive pricing and nourished future generations of innovation. TRIPS facilitated world trade by expanding the geographic scope ofcommodification. However, it also went one step further: it generated a one-way ratchet. States cannot reduce protection below a minimum level, but they are free to increase protection. Furthermore, through the combined effect of most favoured nation and national treatment, and (for patents) technological neutrality obligations, TRIPS is, in effect, an engine driving protection to ever-higher levels throughout the WTO. 189 Procedurally, the shift from WIPO to the WTO converted an informal system of rulemaking into a legalistic regime that tends to mute the voices of certain constituencies. The shift should have immediately raised concerns about legitimacy. However, as Daniel Esty has argued, for other kinds of tradeespecially before the Uruguay Roundlegitimacy concerns were easy to ignore. The interests ofparticipants were fairly well aligned. Once the basic policy decision to engage in freer trade was made, the remaining tariff issues were largely technical and highly susceptible to resolution by expert regulators. Besides, members reaped strong benefits from coordinating their activity. 1 90 It may have appeared straightforward to amalgamate intellectual goods into the same regime. Once again, the issue appeared to be trade, and history had already demonstrated the benefits of a single strong international organ that coordinates the global marketplace.

Events have, however, revealed this assessment of the relationship between intellectual property and trade to be almost entirely incorrect. For intellectual property, the hard questions are matters of policy: put into the vocabulary of trade, they concern the extent to which particular firms and nations should be permitted to enjoy the comparative advantages of technological superiority and lock their advantages in by preventing others from accessing their works and building on them. 191 Nor are the benefits from coordination as large as might be supposed. Developing countries are incurring high costs and developed countries are finding that TRIPS makes it difficult to respond effectively to changes in the innovation landscape. Expertise in trade is unhelpful in dealing with these concerns, and yet the negotiators of the TRIPS Agreement were largely trade, not intellectual property, specialists. 192 Many developments since Uruguay have come about through dispute resolution, but adjudicators likewise tend to have trade, rather than intellectual property, experience.

The previous discussion proposed various mechanisms for bringing together the intellectual property expertise found in WIPO with the trade expertise developing in the Council for TRIPS. However, before any of these mechanisms (or alternative procedures for making international intellectual property law responsive to changing needs) 193 could be adopted, several issues of institutional design must be resolved. The first is a method for controlling forum shopping ('regime shifting'). 1 94 Any proposal that invokes the expertise of more than one 'agency' must deal with the possibilities that the agencies will be played off against each other. Indeed, as the short history recounted above demonstrates, the creative community has long oscillated between multiple and consolidated administration of its international instruments. Currently, the situation is equally complex because WIPO and the WTO are only two of the international organisations with a colourable interest in intellectual property. Others include UPOV (which deals with protection for plants), UNAIDS and WHO (which have a strong interest in medicines), UNCTAD (which deals with development issues), and the Secretariat for the CBD (which deals with genetic resources).' 95 Even the European Court of Human Rights has weighed in (surprisingly, on the side of right holders).' 96 In addition, nations interested in finding a congenial forum have regional and bilateral possibilities. 1 97 Of course, there can be advantages to regulatory competition: overlapping agency authority can lead to better decisions, and different mixes of expertise can produce fresh perspectives. 198 At the same time, however, shared responsibility duplicates work and can also lead to inconsistent 192 See Halbert (n 43) 280 (noting that WIPO was not invited to any of the substantive meetings on TRIPS); Gervais (n 25) 13 (noting that the negotiators realised that because the subject matter was so different from that traditionally handled by the GATT, special information gathering efforts needed to be made, and that they periodically requested information from WIPO). regulations and redundant or incoherent obligations. 199 And for intellectual property, it can exacerbate the protectionist pull. For example, developing countries learned from watching the North shift negotiations from WIPO to the WTO: they have successfully pursued protection for knowledge about genetic resources in negotiations over the CBD. CBD obligations require participating states to recognise a whole new class of rights. Dealing with these new right holders is likely to raise transaction costs, increase the possibilities for holdouts, and at the end of the day, slow innovation. And CBD obligations do not even have the same user flexibilities available in TRIPS. 200 Control over forum shopping is straightforward in domestic administrative contexts because all of the potentially relevant agencies are created by the same authority and that authority can, presumably, also definitively allocate responsibility. As important, there are courts to enforce the allocation chosen. 20 ' But analogous mechanisms have yet to be developed at the international level. There are, however, several possibilities. One is for nations creating a new intergovernmental organisation to make the scope of that organisation's exclusive authority a part of their negotiations, and to agree to bring all matters within its scope only to that forum. Other intergovernmental organisations would presumably have an incentive to honour these allocations as a matter of institutional comitythat is, in the expectation that their primary jurisdictional authority will also be protected from similar manoeuvres. Alternatively, an international norm akin to resjudicata might be developed so that once an issue is resolved in one forum, inconsistent obligations imposed by another intergovernmental organisation will not be enforced (or can be ignored). In either case, however, compliance will remain a significant problemespecially in situations where heavy-handed unilateral action or bilateral negotiation is possible. 20 ping issues prevents forum shopping and, at the same time, improves decision-making without sacrificing consistency.

A second concern relates to sovereignty. In order for the WTO to make active use of the capabilities of WIPO and the Council for TRIPS, members must abandon consensus decision-making and reconceive the decision to rely on agency action as an expression ofrather than a derogation fromsovereign authority. At a minimum, that would likely require a mechanism for assuring members that WIPO or TRIPS Council decisions will be implemented only when they are within the scope of the authority that members chose to delegate. As suggested earlier, one possibility is to rely on disputes over legislation that integrates new norms or authoritative interpretationsthat is, to put the Appellate Body in charge of enforcing the limits on delegation by giving it jurisdiction to decide when the norm or authoritative interpretation is a construction of the Agreement or an amendment to it. Even with that check, the 'agencies' would retain a wide area in which to work out accommodations of various interests. After all, TRIPS leaves many terms ('technology', 'inventive step', 'selling') undefined and the flexibilities provide considerable additional leeway.

Another constellation of issues relates to questions of competence, participation, and accountability. While it is attractive to bring together the intellectual property expertise ofWIPO and the trade expertise of the TRIPS Council, it is necessary to take a closer look at the source of these institutions' knowledge bases. In both cases, there has been considerable suspicion that information is acquired selectively and mainly from strong protectionists. Both organisations tend to act in secret and neither has formal procedures for insuring broad stakeholder input. If international intellectual property law is to be developed through entities akin to administrative agencies, it is imperative to develop global administrative norms and, as important, a mechanism for insuring their application. 20 4 These issues could be addressed by the Appellate Body. Before permitting the incorporation of a new norm or an authoritative interpretation, adjudicators ought to consider the procedures that the Council (and/or WIPO) used to formulate it. The high level of confidence produced by interpretations of open questions that are arrived at by experts, based on a comprehensive study of the issues, subject to broad input, and carefully reasoned, may also compensate for the loss of formal control and thus also serve to assuage sovereignty concerns. 20 5 In addition to these governance norms, which were largely developed in domestic settings, special attention needs to be paid to the international context, where member states at very different levels of sophistication are pitted against each other. 20 6 For example, the TRIPS Council is currently responsible for assisting members engaged in dispute resolution 2 7an obligation which should be broadened to help members pursue their interests in Council deliberations during what is, essentially, agency rulemaking. Indeed, because one effect of TRIPS is to keep countries at a comparative disadvantage disadvantaged, there may be a special obligation in the intellectual property portion of the WTO Agreement to reach out to developing countries to make sure that transparency and participation objectives are measured from their perspective. 20 8 V CONCLUSION Although adding intellectual property to the Uruguay Round may have been key to the establishment of the WTO, 2 0 9 forcing intellectual property into a regime designed to promote free trade had unfortunate consequences for both the consumers and producers of innovative materials. The GATT worked well because reducing trade barriers is an unalloyed benefit: if the principle of comparative advantage is correct, then an engine that exerts continual downward pressure on tariffs will steadily improve social welfare. But intellectual property protection is much more complex. Domestic regimes have never considered maximal protection to be optimal. Instead, the level of protection has varied over timedepending on the needs of emerging and maturing industries, and over placedepending on the levels of national sophistication, access to education, and wealth. The TRIPS Agreement complicates the balancing process. In part, it was drafted in a manner conducive to strict interpretation; in part, it is largely interpreted by tradenot intellectual propertyspecialists; and in part, the WTO utilises decision-making procedures that are incapable of responding quickly to changing circumstances. At best, the TRIPS Agreement is not producing the level of welfare benefits its negotiators envisioned. But it might be doing much worse: slowing 206 CfKingsbury et al (n 21) at 52 (suggesting that global administrative law must take account of the 'imperial' dimension of intergovernmental institutions).

207 Article 68 of TRIPS. 20. Similarly, procedures are needed to develop new norms of substantive fairness in a global intellectual property system, where the incentives available through international protection weaken the case for extracting maximum returns from every local market. See, for example, M Chon 'Intellectual property "from below": copyright and capability for education' (2007) 40 U.C. Davis Law Review 803 (advocating consideration of distributive justice as a goal of international intellectual property law); G B Dinwoodie 'Private ordering and the creation of international copyright norms: the role of public structuring' (2004) 1 Journal of Institutional & Theoretical Economics 160 (suggesting ways in which the international context might shape substantive copyright law). 209 See, for example, Dreyfuss & Lowenfeld (n 5) 276.

progress, fostering deadweight loss, and preventing the poor from benefiting from the world's intellectual achievements. WIPO and the TRIPS Council are currently embarked on development agendas that promise to reshape the relationships between intellectual property and the public domain, between generations of innovators, and between producers and users of knowledge products. However, importing these developments into TRIPS presents new challenges. Although WIPO and the Council have developed, or are in the process of developing, considerable expertise in intellectual property, deliberations have been less than transparent and interest groups have experienced differing levels of access. To some extent, this may be changing: low protectionists in the North are increasingly allying with the South to improve their leverage. Moreover, the emerging Middle is nicely situated to identify and pursue new approaches. For these developments to feed into TRIPS, however, new institutional mechanisms are needed to protect sovereign interests and assure procedural and substantive fairness. The story of TRIPS is, in other words, a demonstration of the need for global administrative norms.

It may be tempting to see the problems of adapting TRIPS to the WTO as unique (as an example of what some call 'IP exceptionalism' 210 ). But it is not. TRIPS became a part of the WTO framework partly because intellectual property is traded in an international marketplace, but also because decisions that one country makes about protection spill over to other nationsif Canada allows generic drug manufacturers to stockpile pharmaceuticals prior to patent expiration, the patent holder loses lead time around the globe; if the Netherlands permits peer to peer file sharing, everyone else enjoys free music as well; 2 11 consumers who see a trademark in Paris expect it to have the same meaning in New York. 2 12 There are other fields where spill-overs exist and where minimum standards are similarly desirable. Some of these may, like trade, represent situations where there is an optimum result that does not vary greatly temporally or geographically. Labour law and civil rights are arguably examples. But there are many places where the issues are complex and both technologically and culturally contingent, and where continual monitoring and adjustments will be required. In these situations, there will inevitably be a similar move to depend on 'functional relationships'