Foong, Cheryl --- "Reconceptualising Copyright Markets: Disseminative Competition as a Key Functional Dimension" [2021] UNSWLawJl 36; (2021) 44(3) UNSW Law Journal 1014

	Home \| Databases \| WorldLII \| Search \| Feedback University of New South Wales Law Journal

The notion of ‘markets’ occupies a prominent yet ambiguous position in copyright discourse. When the term is raised, the copyright owner’s market tends to be taken as its implicit meaning, perpetuating an assumption that the market needs to be protected solely to preserve incentives to create. This dominant narrative overshadows an important dimension of copyright markets – disseminative competition, which is characterised by rival disseminators competing for inputs (copyright content) and audiences (copyright consumers). With the aid of competition law principles, this article distinguishes competition for dissemination of content from competition for the creation of content. It underscores the importance of dissemination markets to a well-functioning copyright system and shows how certain copyright doctrines substantively impact on disseminative competition. In reframing contemporary understandings of copyright markets, this article highlights the biases in copyright infringement analysis that may favour incumbent content disseminators to the detriment of a vibrant and innovative digital economy.

I INTRODUCTION

The notion of ‘markets’ is often raised in debates on the proper scope of copyright protection, without recognition of its connotations. In using the term ‘markets’, emphasis tends to be placed on the protection of owners or authors of copyright content from demand substitution (ie, encroachment upon their markets). This author-centric perspective narrowly focuses on copyright owners’ interests and overlooks the market structures and competitive pressures in copyright markets, particularly disseminative markets. The incentive side of the classical incentive-access paradigm of copyright takes priority.1

* Senior Lecturer, Curtin Law School, Curtin University. I thank Joanne Gray, Brian Fitzgerald, participants in the 5^th Annual Texas A&M Intellectual Property Scholars Roundtable (and convenor Peter Yu) and the referees for their comments and feedback. I also acknowledge the Whadjuk Noongar people as custodians of the lands on which Curtin Law School is located, and pay respect to elders past, present and emerging.

1 Under the incentive-access paradigm, the law provides protection to secure incentives to create but broadening protection may impede access to the works created: see Stephen Breyer, ‘The Uneasy Case for Copyright: A Study of Copyright in Books, Photocopies, and Computer Programs’ (1970) 84(2) Harvard

Copyright, in addition to having a prominent authorship-incentivising function, serves to disseminate content to the public.2 It has been observed that ‘[t]he development of all of the major content industries trace back to key technological innovations that made possible the instantiation and dissemination of expressive works’.3 These technologies advance copyright’s dissemination function and ensures that the law fulfils the benefits of access to content promised to the public. Furthermore, both creators and disseminators are indispensable to the advancement of copyright’s functions; without expression there is nothing to disseminate and without means for dissemination such expression would have no audience. Existing scholarship has recognised the importance of dissemination to the furtherance of copyright’s fundamental policy objectives,4 yet analysis of how dissemination markets affect the public ‘access’ side of the incentive-access paradigm remains relatively limited.

We need a more nuanced approach to analysing markets in copyright law that recognises both the expressive and disseminative dimensions of such markets. The role of disseminators should not be subsumed into that of content creators and producers, as different parties may be executing these roles and each party could be subject to different competitive pressures. Drawing inspiration from Australian competition law’s analysis of markets, this article shows how creators and disseminators of content effectively compete at different functional levels. Disproportionate focus may be placed on expressive competition when a copyright dispute has significant impact on disseminative competition. This is apparent once we consider the disparate effects certain legal doctrines have on expressive and disseminative competition. This exercise carries two key objectives: (1) recognising these trends in legal analysis and highlighting what is missing from the existing

Law Review 281. See further Glynn S Lunney Jr, ‘Reexamining Copyright’s Incentives-Access Paradigm’ (1996) 49(3) Vanderbilt Law Review 483. It should be recognised that other Intellectual Property (‘IP’) regimes such as patents and trademarks confer their own forms of market power, commensurate with the nature of the exclusive rights granted. While some of the scholarship referred to in this article consider the notion of markets and intellectual property (‘IP’) more broadly, this article focuses on copyright for two related reasons: firstly, recognising differences in the nature of rights granted under other IP regimes and secondly, to afford close consideration of copyright’s key exclusive rights.

2 Cheryl Foong, The Making Available Right: Realizing the Potential of Copyright’s Dissemination Function in the Digital Age (Edward Elgar, 2019). See also L Ray Patterson and Stanley F Birch Jr, ‘A Unified Theory of Copyright’ (2009) 46(2) Houston Law Review 215, 392, describing dissemination as one of several functions of copyright.

3 Peter S Menell, ‘This American Copyright Life: Reflections on Re-equilibrating Copyright for the Internet Age’ (2014) 61(2) Journal of the Copyright Society of the United States of America 235, 271.

4 See, eg, Timothy Wu, ‘Copyright’s Communications Policy’ (2004) 103(2) Michigan Law Review 278 (arguing that copyright has a communications policy that regulates rivalries between disseminators of content); Séverine Dusollier, ‘Realigning Economic Rights with Exploitation of Works: The Control of Authors over the Circulation of Works in the Public Sphere’ in P Bernt Hugenholtz (ed), Copyright Reconstructed: Rethinking Copyright’s Economic Rights in a Time of Highly Dynamic Technological

and Economic Change (Wolters Kluwer, 2018) 163 (asserting that copyright has a main ‘communicative function’, which is to promote and regulate ‘the circulation of works in the public sphere’). See also Randal C Picker, ‘Copyright as Entry Policy: The Case of Digital Distribution’ (2002) 47 Antitrust Bulletin 423.

narrative; and (2) moving forward, refining copyright law’s role in encouraging competition in dissemination markets.

The findings from this article are of significance to our ability to foster competitive and efficient dissemination markets. Instead of allowing incumbent disseminators in increasingly concentrated online environments to shield their interests from critique behind the rhetoric of authors’ interest, we should be able to parse the respective interests of content producers, disseminators and consumers. This nuanced understanding of market power should guide our assessment of whether contemporary copyright dissemination markets meet the needs of consumers. Notably, the need to coordinate copyright and competition policies has been explicitly recognised by the World Intellectual Property Organisation (‘WIPO’). WIPO explained in a brief note titled ‘Streaming Wars’ that ‘[t]he incentive to invest in new content will not only depend on the provisioning of rights but also on the level of market competition’.5 However, the question is not simply what level of competition should be encouraged in copyright markets, but what type of competition. Current abstract conceptions of copyright ‘markets’ act as blinders; they prevent us from seeing the different dimensions and policy tensions in copyright law. With the removal of these ‘market’ blinders encumbering copyright analysis, lawyers, judges, and scholars have new opportunities to pave a clearer path forward for competitive disseminative markets.

II ‘MARKETS’ IN COPYRIGHT RHETORIC AND THEORY

At a fundamental level, copyright may be described as a precondition to the existence of copyright markets, serving to remedy a market failure caused by the non-rivalrous and non-excludable nature of copyright as an intangible output.6 In other words, without some form of exclusivity, owners would have nothing to deal

5 ‘Streaming Wars: Creative Economy Notes’, World Intellectual Property Organization (Web Page, 2020)

<https://www.wipo.int/edocs/infogdocs/creative_industries/en/streaming-wars/> (analysing statistical data on streaming services in Brazil). In addition, inquiries by the Australian government have highlighted the importance of assessing the anti-competitive effects of IP rights: Productivity Commission, ‘Intellectual Property Arrangements’ (Inquiry Report No 78, 23 September 2016) 444 (‘A well-functioning IP system must take account of the long term effects on competition that exclusivity can create’); Ian Harper et al, ‘Competition Policy Review’ (Final Report, March 2015) 41 (‘Harper Review’) (‘IP rights can help to break down barriers to entry but, when applied inappropriately, can also reduce exposure to competition and erect long-lasting barriers to entry that fail to serve Australia’s interests over the longer term.’); Intellectual Property and Competition Review Committee, Review of Intellectual Property Legislation under the Competition Principles Agreement (Final Report, September 2000) 7 (recognising that

‘first, competition concerns must play an important part in the design of the intellectual property laws themselves, ensuring that the rights granted are not over-broad; and second, effective remedies must be available if the rights are abused’).

6 See William Fisher, ‘Theories of Intellectual Property’ in Stephen R Munzer (ed), New Essays in the Legal and Political Theory of Property (Cambridge University Press, 2001) 168, 169. See also Oren Bracha and Talha Syed, ‘Beyond the Incentive-Access Paradigm? Product Differentiation and Copyright Revisited’ (2014) 92(7) Texas Law Review 1841, 1849–50, who argue that the problem primarily is non- excludability (not non-rivalry), and that IP rights convert ‘what are relatively nonexcludable goods into relatively excludable ones’.

or bargain with. However, the property rights granted and markets they enable are a vehicle for advancing societal benefits; they do not exist for the sake of rewarding producers of content.7 Markets should be seen as a tool for realising the public benefits of copyright. Copyright is a means to an end, and society gains the utilitarian benefit from authorship so long as the authored content is disseminated to the public.8

Copyright has been described by the United States (‘US’) Supreme Court as ‘a marketable right to use ... one’s expression’,9 a statement that hints at copyright law’s role as a ‘market control mechanism’.10 Unfortunately, the market conception of copyright tends to be oversimplified, leading to an assumption ‘that a copyright exists to protect a copyright owner’s market’.11 This focus on the owner flows from the property-centric view of copyright, with the idea of property carrying ‘an established expectation ... to derive certain advantages from the object’.12 The ‘established expectation’ is that authors or producers that invest in creative activity through relevant markets will recoup their investment.13 However, a precondition

7 Cf John Locke’s labour desert theory which posits that one may legitimately acquire property rights by mixing their labour with resources held in common, provided there is ‘enough and as good left in common for others’: John Locke, ‘Second Treatise’ in Peter Laslett (ed), Locke’s Two Treatises of Government (Cambridge University Press, 1960) 285, 306.

8 See William M Landes and Richard A Posner, ‘An Economic Analysis of Copyright Law’ (1989) 18(2) Journal of Legal Studies 325. The tension between public access and authorship incentives is also known as copyright’s ‘access-incentive’ paradigm: Breyer (n 1).

9 Harper & Row Publishers Inc v Nation Enterprises, [1985] USSC 128; 471 US 539 (1985) 558 (‘Harper & Row’).

10 See Niva Elkin-Koren, ‘It’s All about Control: Rethinking Copyright in the New Information Landscape’ in Niva Elkin-Koren and Neil Weinstock Netanel (eds), The Commodification of Information (Kluwer Law International, 2002) 79, 80. Likewise, Lyman Ray Patterson has observed that ‘copyright statutes from the time of the Statute of Anne have been trade regulation statutes’: Lyman Ray Patterson, ‘Private Copyright and Public Communication: Free Speech Endangered’ (1975) 28(6) Vanderbilt Law Review 1161, 1194. See also Lyman Ray Patterson, ‘The Statute of Anne: Copyright Misconstrued’ (1966) 3(2) Harvard Journal on Legislation 223, 236.

11 Anna F Kingsbury, ‘Market Definition in Intellectual Property Law: Should Intellectual Property Courts Use an Antitrust Approach to Market Definition?’ (2004) 8(1) Marquette Intellectual Property Law Review 63, 85 (emphasis added), referring to the US Supreme Court’s approach in Harper & Row[1985] USSC 128; , 471 US 539 (1985) 1084, which in turn cites Wendy J Gordon, ‘Fair Use as Market Failure: A Structural and Economic Analysis of the Betamax Case and Its Predecessors’ (1982) 82(8) Columbia Law Review 1600, 1615.

12 Jeremy Bentham, ‘Principles of the Civil Code (1843)’ in John Bowring (ed), The Works of Jeremy Bentham (Russel and Russel, 1962) 308. The property system is said to provide independence and autonomy to authors given property rights to trade in authors are no longer beholden to monarchs, churches or other patrons for their livelihoods: Clark D Asay, ‘Copyright’s Technological Independencies’ (2015) 18(2) Stanford Technology Law Review 189, 192, citing Copyright Law Revision: Hearings

on S 1006 before the Subcommittee on Patents, Trademarks, and Copyrights of the Committee on the Judiciary, United States Senate, 89^th Congress, 1^st session: 18–20 August 1965 (United States Government Printing Office, 1967) (statement of Abraham L Kaminstein, former Register of Copyrights, that copyright provides ‘an alternative to the evils of an authorship dependent upon private or public patronage’). The market (as an alternative to patronage) is said to support the production of ‘diverse forms of expression for the purposes of advancing society’: Alina Ng, Copyright Law and the Progress of Science and the Useful Arts (Edward Elgar, 2011) 59.

13 See Jennifer Rothman’s caution against utilising expectations to drive the development of IP law: Jennifer E Rothman, ‘The Questionable Use of Custom in Intellectual Property’ (2007) 93(8) Virginia Law Review 1899, 1961–5. As Rothman rightly suggests, industry-driven solutions should not as a general matter

to market participation should not be mistaken for a guarantee of market share or market profits.14 Yet the ‘property’ label has led to a tendency to treat intellectual property (‘IP’) rights, such as copyright, as ‘absolute in the sense that they are largely invariant to market structure or market-specific incentives’.15 Once a property right is created, there is little concern about the state of markets enabled by that right.

Characterisation of copyright as property may be one explanation for the limited analysis of markets in copyright law. Another reason that Herbert Hovenkamp proffers is that ‘IP law has never produced a sufficiently robust consensus about the relationship between market structure and optimal IP protection’.16 However, as Mark Lemley and Mark McKenna argue, there is already widespread reliance on market definition in IP, without acknowledgement of the line drawing exercise being undertaken and without an apparent methodology applied.17 Likewise, Anna Kingsbury has observed that ‘market’ in IP is a flexible concept, analysed and used differently depending on the context or area of IP law.18 Courts often arrive at conclusions on the scope of rights without empirical evidence about relevant markets.19

Put simply, we have fallen into the habit of letting our assumptions about property drive the scope of copyright, rather than asking how copyright can shape exclusive rights (and the markets built upon these rights) to advance public policy.20 The existing copyright literature is not devoid of market analysis or consideration

be taken as accurate representations of optimal levels of protection, as the unequal bargaining power of the parties and underrepresentation of the public in the development of these practices tend to lead to the development of suboptimal practices and norms: at 1947. See also James Gibson, ‘Risk Aversion and Rights Accretion in Intellectual Property Law’ (2007) 116(5) Yale Law Journal 882, 932 (‘A market formed in the shadow of legal ambiguities, risk-averse actors, and strategic bargaining ... tells us little about the entitlement’s optimal coverage’).

14 Cf Elkin-Koren (n 10) 105, citing Christopher Lind, ‘The Idea of Capitalism or the Capitalism of Ideas? A Moral Critique of the Copyright Act’ (1991) 7 Intellectual Property Journal 65, 69 (‘Copyright in this sense protects a market share. It defines rights to exploit the commercial potential of the work’).

16 Herbert Hovenkamp, ‘Response: Markets in IP and Antitrust’ (2012) 100(6) Georgetown Law Journal

2133, 2158. Hovenkamp argues that this has ‘served to keep serious market power enquiries off the table’.

19 See ibid 71–2 (using US trademark cases as examples). See also the role of market definition in calculating damage awards in IP cases: Marion B Stewart, ‘Calculating Economic Damages in Intellectual Property Disputes: The Role of Market Definition’ (1995) 77(4) Journal of the Patent and Trademark Office Society 321.

20 It should be noted that the property label, while well-accepted in copyright jurisprudence and explicit in section 196 of the Copyright Act 1968 (Cth), has been strongly criticised by several scholars: see, eg,

Jessica Litman, ‘What We Don’t See When We See Copyright as Property’ (2018) 77(3) Cambridge Law Journal 536 (explaining how the property conception exacerbates disparities in wealth and bargaining power between authors and intermediaries); L Ray Patterson, ‘Copyright Overextended: A Preliminary Inquiry into the Need for a Federal Statute of Unfair Competition’ (1992) 17(2) University of Dayton Law Review 385, 403–7 (arguing that the property conception has facilitated the use of legal fictions to overextend the subject matter of copyright protection).

of dissemination markets.21 However, this debate is underappreciated and deserving of more urgent attention from lawmakers, policymakers and the judiciary.

III MARKET-BASED CALLS FOR REFORM

A number of scholars have investigated the meaning of ‘markets’ and the surrounding assumptions, and have called for copyright reforms tied to the notion of markets or commercial exploitation. Broadly, these market-based reform proposals fall into two themes, although these thematic categories are by no means mutually exclusive:

2. Interpreting the law with reference to reasonable expectations and market harms.

The first category is more drastic, proposing an overhaul of exclusive rights with the overall objective of simplifying the copyright framework.22 The second category covers a body of scholarship arguing that copyright has a role in fulfilling reasonable expectations of reward or avoiding market harm, and that this can be achieved through the interpretation of existing copyright rules.23 While insightful, these reform proposals to an extent continue to rely on abstract notions of the ‘market’ or on market expectations, which cultivates expansive approaches to copyright infringement.

A Exclusive Right of Commercial Exploitation

Scholarship within the first category envisions significant legislative change and proposes that copyright owners are granted an overarching exclusive right to exploit their content in a manner that accords with market practices. Jessica Litman, for instance, has called for a recasting of copyright as ‘an exclusive right of commercial exploitation’,24 a proposal that is said to accord with copyright practice and public

21 For example, Timothy Wu has argued that copyright has a communications policy that regulates competition between rival disseminators: Wu, ‘Copyright’s Communications Policy’ (n 4). Wu’s seminal article articulates how rivalries between disseminators are regulated by copyright law, particularly by the rules on secondary copyright liability. However, as the boundary of primary infringement of exclusive rights and secondary liability blurs, a broader reconceptualisation of copyright which considers the different forms of competition is timely.

22 See, eg, Jessica Litman, Digital Copyright (Prometheus Books, 2^nd ed, 2006) 180 (to be discussed in more detail below).

24 Litman (n 22) 180. See also Daniel J Gervais, (Re)Structuring Copyright: A Comprehensive Path to International Copyright Reform (Edward Elgar Publishing, 2017) 213 (arguing for ‘a right to prohibit uses that demonstrably interfere with actual or predictable commercial exploitation’ of copyright content); Ole-Andreas Rognstad and Joost Poort, ‘The Right to Reasonable Exploitation Concretized: An Incentive Based Approach’ in P Bernt Hugenholtz (ed), Copyright Reconstructed: Rethinking Copyright’s Economic Rights in a Time of Highly Dynamic Technological and Economic Change (Wolters Kluwer, 2018) 121 (proposing an exclusive right to ‘reasonable exploitation’, which is a right to control uses of the work to the extent necessary to secure current and future exploitation opportunities).

expectations and be likely to ease enforcement efforts.25 While the form may vary, other scholars have argued for an exclusive right to ‘reasonable exploitation’, or a right to ‘actual or predictable commercial exploitation’.26 This overarching right would replace a proliferation of overlapping rights, and entail moving away from ‘our current bundle-of-rights way of thinking about copyright infringement’.27 In evaluating a defendant’s actions, the question instead is ‘what effect those actions had on the copyright holder’s opportunities for commercial exploitation’.28

Linking infringement to the realities of markets is a step in the right direction. However, without a clearer understanding of copyright markets, a right of market exploitation may perpetuate an expansive approach to infringement. By conceptually allocating markets in the abstract to copyright owners, parties with deep pockets will be better positioned to advocate for a broader scope of exclusivity through litigation.29 Therefore, it is questionable whether, as asserted by Litman, a single right of commercial exploitation is ‘a more constrained grant than the current capacious statutory language’.30 It may be so as conceptualised by scholars advancing this approach, but in the context of prevalent interests and a property conception of rights, this perspective is likely to be lost in the process of implementation and application.31 In short, such proposals carry risks, which along with the drastic nature of the proposal, weigh against the envisaged benefits.

B Infringement Test: Reasonable Expectations of Market Harm

The second category of market-based reform proposals focuses on judicial interpretation of existing legislative frameworks. This scholarship proposes that the scope of exclusive rights should be tied to market harms.32 Again, these proposals

29 The US Supreme Court case of Herbert v Shanley Co, [1917] USSC 29; 242 US 591 (1917) illustrates this point. Here a ‘for profit’ restriction on the scope of the performance right, envisaged by some members of the

legislature as a protection of the rights of the public to enjoy copyright works, was construed broadly to include the playing of music to make a business more attractive regardless of whether money was taken at the door. For further discussion of this case, see Foong, The Making Available Right: Realizing the Potential of Copyright’s Dissemination Function in the Digital Age (n 2) 174–6.

31 It should be noted that Litman recognises that ‘[o]ne significant drawback of this sort of standard ... is that it would replace the detailed bright lines in the current statute with uncertainty’, but goes on to argue that ‘[t]he brightness of the current lines is illusory’: ibid. Litman also acknowledges that this change will not solve copyright’s problems, and calls for explicit recognition of the public’s rights in the law (among other things): at 182.

32 This article focuses on conceptual approaches to prima facie infringement, although it should be noted that there is rich scholarship on market effects or harms in other areas of copyright law, particularly the US fair use doctrine and the derivative work right: see, eg, Jeanne C Fromer, ‘Market Effects Bearing on Fair Use’ (2015) 90(2) Washington Law Review 615; Wendy J Gordon, ‘Fair Use Markets: On Weighing Potential License Fees’ (2011) 79(6) George Washington Law Review 1814; Glynn S Lunney Jr, ‘Copyright, Derivative Works, and the Economics of Complements’ (2010) 12(4) Vanderbilt Journal

of Entertainment and Technology Law 779. While the introduction of an open-ended fair use exception in Australia has been proposed and its merits debated at length, the proposal has yet to be implemented: see, eg, Productivity Commission (n 5) 165. Cf Australian Government, ‘Australian Government Response

aim to limit the reach of the current copyright regime. Sara Stadler, for example, argues that in the context of infringement we should ask ‘what creators are entitled to expect given the nature of the public interest in copyright’.33 More specifically, the argument is that infliction of competitive harm in a ‘relevant market’ ought to be defined as the ‘rights creators are entitled to expect to enjoy when they engage in the act of creation’,34 with the notion of harm tied to the public interest in access. Along similar lines, Christina Bohannan and Herbert Hovenkamp call for explicit consideration of harms to innovation (as opposed to markets).35 They propose that courts should develop a concept of ‘IP injury’ (drawing inspiration from the concept of ‘antitrust injury’ to competition), broadly linked to the underlying goals of IP laws to encourage innovation.36 While there is a greater focus on innovation, Bohannan and Hovenkamp’s doctrinal proposals align with Stadler’s as they also rely on expectations as a limitation. The authors assert that ‘[i]nfringement harms innovation when it diminishes anticipated returns, and the only effects that can be anticipated are those that are reasonably foreseeable’.37 Indeed, the foundation for these proposals based on harm, injury and expectations is the public interest in incentivising creation and promoting access to the resulting works. However, how such harms or expectations are interpreted depends on whose perspective is considered. As Wendy Gordon has explained, ‘[h]arm is usually considered a shortfall from some baseline’ and ‘there are potentially an infinite number of possible baselines, depending on one’s normative commitments’.38 She observes

to the Productivity Commission Inquiry into Intellectual Property Arrangements’ (Final Report, August 2017) 7.

33 Sara K Stadler, ‘Copyright as Trade Regulation’ (n 23), citing Sara K Stadler, ‘Incentive and Expectation in Copyright’ (2007) 58(3) Hastings Law Journal 433, 476–7. Stadler’s argument on incentives and expectation is that changes in the law condition creators to expect increasing rewards.

35 Christina Bohannan and Herbert Hovenkamp, ‘IP and Antitrust: Reformation and Harm’ (2010) 51(4) Boston College Law Review 905. In the copyright context (at 971–9), the authors discuss several fair use cases that have an impact on innovation in expression, for example, a transformative race-based critique/ parody of the novel ‘Gone with the Wind’ in Suntrust Bank v Houghton Mifflin Co, [2001] USCA11 368; 268 F 3d 1257 (11^th Cir, 2001) and innovation in dissemination, eg, time-shifting of free-to-air television programs in Sony Corporation of America v Universal City Studios Inc, [1984] USSC 14; 464 US 417 (1984). The distinction between expressive and disseminative markets is discussed further in Part V of this article.

36 Bohannan and Hovenkamp, ‘IP and Antitrust: Reformation and Harm’ (n 35). See also Christina Bohannan and Herbert Hovenkamp, Creation without Restraint: Promoting Liberty and Rivalry in Innovation (Oxford University Press, 2012) 33–59.

37 Bohannan and Hovenkamp (n 35) 989, citing Christina Bohannan, ‘Copyright Harm, Foreseeability, and Fair Use’ (2007) 85(5) Washington University Law Review 969, 988–9. See also Shyamkrishna Balganesh, ‘Foreseeability and Copyright Incentives’ (2009) 122(6) Harvard Law Review Association 1576. Focusing on incentives to create, Shyamkrishna Balganesh argues for an infringement test that refers to ‘foreseeable copying’, ie, exclusivity is limited to uses reasonably foreseeable at the time the work was created. Balganesh however distances his arguments from market-based proposals, clarifying

that foreseeability should relate ‘to the form and purpose of the defendant’s copying and not other factors, such as its magnitude or monetary consequences’: at 1605. In other words, ‘foreseeability would focus

on the defendant’s actions (that is, the copying), rather than function as an open-ended device that courts might then connect to the notions of “harm” or “market”’: at 1606.

38 Wendy J Gordon, ‘The Concept of “Harm” in Copyright’ in Shyamkrishna Balganesh (ed), Intellectual Property and the Common Law (Cambridge University Press, 2013) 452, 465. See also Shyamkrishna Balganesh, ‘Copyright as Market Prospect’ (2018) 166(2) University of Pennsylvania Law Review 443,

that ‘[t]o define any shortfall from a statutory entitlement as a “harm” just leads to an unhelpful circularity’,39 and therefore cautions ‘against rushing to impose a harm requirement before the conceptual and policy issues inherent in “harm” are fully fleshed out’.40 In essence, using measurements of harm to determine the scope of copyright protection is problematic.

Instead of focusing on market harms, one might utilise alternative terminology and consider copyright’s role in protecting copyright as market prospect. Shyamkrishna Balganesh draws parallels between copyright infringement and tortious interference with market prospect, an action that seeks to protect a zone of probabilistic market benefits from specific volitional interferences.41 Aiming to protect market prospect is arguably more grounded in commercial reality in contrast to preventing ‘harms’ in the abstract. Importantly, Balganesh highlights that ‘[w]ithout a plausible market for a work, no amount of copyright protection can induce creative production’.42 In other words, we should be cautious of a theoretical over-reliance on copyright law as a supplier of incentives.43 As Arnold Plant stated in 1934 from a demand perspective, ‘[c]opyright in a particular work cannot itself create a demand for the kind of satisfaction which that work and similar works may give, it can only make it possible to monopolise such demand as already exists’.44 A crucial point that can be drawn from Plant and Balganesh’s respective arguments is that in order to maintain copyright’s incentive function, competitive dissemination market structures must be in place to facilitate efficient access.45

Such conceptual approaches bring us closer to the more objective competition law approach to markets, whereby expectations or harms are not limited to the

447: ‘Unlike the tangible harm with which negligence commonly concerns itself, the harm that copyright seeks to prevent is a contested and multifaceted issue.’

42 Ibid 504. Balganesh further explains that ‘copyright does not create the incentive to produce work, instead it protects that incentive when generated by the market’: at 506 (emphasis in original). In other words, ‘[c]opyright shapes a putative market potential through its set of exclusive rights, by raising the prospect of its realization. But that is a far cry from creating or supplying the incentive on its own, in the way that courts caricature the justification’: at 505 (emphasis omitted).

43 See also Rebecca Tushnet’s point that ‘regardless of the strength of protection, it is the likelihood of success in the market – a highly unpredictable variable, and one that [copyright] law can do little if anything to affect – that is key to whether new authors reap rewards from creating works’: Rebecca Tushnet, ‘Economies of Desire: Fair Use and Marketplace Assumption’ (2009) 51(2) William and Mary Law Review 513, 517–18. Along a similar vein, Herbert Hovenkamp has explained that ‘an IP right may grant freedom from duplication and thus permit appropriation of whatever value an underlying asset already has. But if it is attached to something of no value, then the IP right will not confer any value’: Hovenkamp, ‘Response: Markets in IP and Antitrust’ (n 16) 2139.

44 Arnold Plant, ‘The Economic Aspects of Copyright in Books’ (1934) 1(2) Economica 167, 171 (emphasis omitted).

45 Balganesh’s broader argument is that parallels between copyright and tortious interference with market prospect ‘suggests spending more attention on the details of the underlying market for the work and

its construction of demand when speaking of copyright’s function as an inducement for creativity’: Balganesh, ‘Copyright as Market Prospect’ (n 38) 506.

perspective of particular parties.46 A consideration of market substitution has been raised in scholarship criticising the use of ‘harm’ as a benchmark for infringement. For example, Gordon argues that ‘copyright should recognize only the effects of rivalry as constituting “harm”’.47 Likewise, Balganesh argues that interpretation of ‘substantial’ similarity ‘is a stand-in for whether the substitutionary potential of the copy was real, rendering the freeriding wrongful’.48 However, we do not as yet have clear or consistent ways to analyse market substitution under copyright law. It is also questionable whether market substitution analysis is enough. The scholarship canvassed above indicates that a closer look at copyright markets is warranted, as it remains unclear how the notion of ‘markets’ or market substitution ought to be considered under copyright law.

IV LESSONS FROM COMPETITION LAW

Analytical approaches that have developed in competition law,49 an area of law that has an express role in regulating markets, may be used to strengthen our understanding of copyright markets. However, it should be recognised that key conceptual and doctrinal differences exist between the two areas of law,50 and therefore, competition law tests should not be imported blindly into copyright law. For example, a copyright doctrine may selectively call for the identification of markets but attempts to define a single product market in the copyright context is a challenge. Nevertheless, competition law’s analytical tools could support a reconception of copyright markets and bring focus to dissemination markets. Particularly relevant is Australian competition law’s purposive approach to identifying the relevant functional level of a market. The approach highlights the different types of competition that are occurring in copyright markets and prompts further consideration of the disparate impacts certain legal elements of copyright infringement have on competition.

46 See Kingsbury (n 11) 91, who argues that ‘[u]se of antitrust market definition would produce more empirically based results and more consistent outcomes’. Note however that Kingsbury limits her proposed use of antitrust market definition to areas of IP that more explicitly seek to facilitate market competition, such as trademark law. Furthermore, Herbert Hovenkamp notes that in copyright

cases ‘“harm to the market” really meant “harm to the plaintiff” – a proposition that antitrust policy categorically rejects’: Hovenkamp, ‘Response: Markets in IP and Antitrust’ (n 16) 2149.

47 Gordon, ‘The Concept of “Harm” in Copyright’ (n 38) 483 (emphasis added). This is in contrast to other ‘harms’ such as ‘revenue foregone because of a non-rivalrous defendant’s failure to pay license fees, loss of exclusivity, and subjective distress’.

49 In the United States, the term antitrust law is used, due to its original role to dismantle the ‘trusts’ controlling national industries such as railroads, oil production and steel manufacturing: Wayne D Collins, ‘Trusts and the Origins of Antitrust Legislation’ (2013) 81(5) Fordham Law Review 2279.

50 It should be noted that the aim here is not to detail the intersection between IP and competition, which has received in-depth treatment elsewhere: see, eg, Ariel Katz, ‘Making Sense of Nonsense: Intellectual Property, Antitrust, and Market Power’ (2007) 49(4) Arizona Law Review 837.

A Conceptual and Doctrinal Differences

A key conceptual difference between copyright law and competition law is that competition law ‘starts out from the position that markets generally work well and correct themselves, and that government intervention is justified only occasionally’, whereas ‘market failure is the starting point for IP laws [including copyright], and it is market failure that gives rise to the need for legal entitlements’.51 Despite these contrasting starting points, the objectives of IP and competition laws are said to align, as they share ‘a common purpose of promoting innovation and consumer welfare’.52

Methods of achieving that common purpose differs – one seeks to promote competition while the other permits limited distortions in competitive pressures to promote more long-term innovation goals. Nevertheless, consumers are ultimately said to benefit from new products (under IP law) and have access to products at the lowest price (under competition law).53

Still, tensions persist between the two areas of law – notably, ‘antitrust scepticism often aligns with IP optimism and vice versa’.54 Scholars have called for more cross-pollination of ideas and approaches. Tim Wu, for instance, has argued that IP should learn from competition law’s ‘pragmatic and experimentalist’ culture, as opposed to just being ‘influenced by natural law and vested rights’.55 Taking inspiration from a competition law perspective of markets could provide a less protectionist and more objective way to analyse copyright infringement issues.56

B Product Market Definition Not Directly Applicable

In line with the objectives of competition law, attempts to define the market are undertaken in order to identify competitive constraints on a firm’s power to act within a market, and ultimately, the degree of market power held by the firm. While certain conduct such as cartel arrangements are per se prohibited without enquiry into the effect of the conduct on competition,57 other conduct requires a ‘rule of reason’ analysis into the competitive harms caused, or likely to be caused, by the

52 United States Department of Justice and the Federal Trade Commission, ‘Antitrust Guidelines for the Licensing of Intellectual Property’ (Federal Guidelines, 6 April 1995) 2, citing Atari Games Corp v Nintendo of America Inc, [1990] USCAFED 174; 897 F 2d 1572, 1576 (Norgle J) (Fed Cir, 1990).

53 ‘[I]ntellectual property and antitrust laws work in tandem to bring new and better technologies, products, and services to consumers at lower prices’: United States Department of Justice and the Federal Trade Commission, ‘Antitrust Enforcement and Intellectual Property Rights: Promoting Innovation and Competition’ (Report, April 2007) 1.

54 Stacey Dogan, ‘The Role of Design Choice in Intellectual Property and Antitrust Law’ (2016) 15(1)

55 Tim Wu, ‘Intellectual Property Experimentalism by Way of Competition Law’ (2013) 9(2) Competition Policy International 30.

56 This point is made by Anna Kingsbury, but in the context of the fair use exception: Kingsbury (n 11) 88.

57 These ‘are practices which Parliament has seen as so generally offensive to the competitive goals underlying the Act that they are to be condemned without consideration of any purpose or effect of substantially lessening competition in a market’: Rural Press Ltd v Australian Competition and Consumer Commission [2003] HCA 75; (2003) 216 CLR 53, [82] (Gummow, Hayne and Heydon JJ, with Gleeson CJ and Callinan J agreeing at [3]).

conduct in question.58 Defining a relevant market is a preliminary step; one then moves on to assess a firm’s market power or changes in the level of competition (if any) in that market due to the conduct of particular firm(s).59

In the Australian competition law context, markets have been described as ‘the field of rivalry between [firms]’, or stated less simply, it ‘is the field of actual and potential transactions between buyers and sellers amongst whom there can be strong substitution, at least in the long run, if given a sufficient price incentive’.60 Therefore, the identification of close substitutes, either in demand or in supply, is a crucial step in defining the relevant market.61 Courts and administrative agencies utilise the Hypothetical Monopolist Test (also known as the SSNIP test) to define the market.62 The test starts with the smallest plausible market and hypothesises that a monopolist controls that market.63 The question is whether implementation of a Small but Significant (5–10%), Non-transitory (12 month) Increase in Price (‘SSNIP’) by the monopolist would cause new entry and render the price increase unsustainable.64 Where the SSNIP is not profitable, this means that the conduct of firms within the market are conditioned by other products or services not defined in the market. The test is then repeated with incrementally broader market boundaries until one finds the narrowest market that a hypothetical monopolist could sustainably institute a SSNIP.65 It may be described as a ‘thought experiment’, particularly where there is insufficient information to enable a formal econometric test.66

58 Under a rule of reason analysis, the conduct is prohibited only if it substantially lessens competition in the market as defined (known as the SLC test or competition test in Australia): see Arlen Duke, Corones’ Competition Law in Australia (Thomson Reuters, 7^th ed, 2018) 177–8.

59 It should be recognised that ‘[d]efining the market and evaluating the degree of power in that market are part of the same process, and it is for the sake of simplicity of analysis that the two are separated’: Queensland Wire Industries Pty Ltd v Broken Hill Proprietary Co Ltd [1989] HCA 6; (1989) 167 CLR 177, 187–8 (Mason CJ and Wilson J).

60 Re Queensland Co-operative Milling Association Ltd (1976) 8 ALR 481, 517 (Woodward J) (‘Re QCMA’). This statement of the Trade Practices Tribunal has since been accepted by Australian courts: see Australian Competition and Consumer Commission v Boral [1999] FCA 1318, [121] (Heerey J). See also Competition and Consumer Act 2010 (Cth) section 4E: ‘[A] market in Australia and, when used in relation to any goods or services, includes a market for those goods or services and any other goods or services that are substitutable for, or otherwise competitive with, the first-mentioned goods or services’.

61 Australian Competition and Consumer Commission v Metcash Trading Ltd [2011] FCA 967; (2011) 282 ALR 464, 496 (‘Metcash’). Emmett J explained (emphasis in original):

Substitution, either in demand or in supply, defines that area of competition. ... It is necessary to identify close substitutes, since close substitutes will impose competitive discipline on a firm and its prices.

62 Duke (n 58) 70–1, citing Seven Network Ltd v News Ltd [2007] FCA 1062, [1777] (‘Seven Network’); Re QCMA (1976) 25 FLR 169, 190.

64 Ibid. Note that a particular problem that arises with the use of this test is the ‘cellophane fallacy’, where it is impossible to determine a competitive price apart from the current price: see United States v EI

65 Metcash [2011] FCA 967; (2011) 282 ALR 464, 496 (Emmett J). As Cento Veljanovski explains, the defendant will generally seek a wider definition to escape liability, but ‘the purpose of the test is expressly to identify the narrowest defensible market definition, not the widest plausible one’; Cento G Veljanovski, ‘Markets in Professional Sports: Hendry v WPBSA and the Importance of Functional Markets’ (2002) 23(6) European Competition Law Review 273, 280.

66 Duke (n 58) 71, citing Seven Network [2007] FCA 1062, 610 [1786] (Sackville J).

While our understanding of copyright markets could benefit from a competition law perspective, it is not suitable to apply a competition law product market definition approach directly to copyright. As Herbert Hovenkamp has explained, the ‘process is inherently binary in the sense that a product is either inside or outside of the market’ and ‘in a product differentiated market, both conclusions are commonly “wrong”’.67 This is because IP exists precisely to encourage product differentiation.68 Indeed, differentiated copyright works are subject to competition from partial substitutes. For example, JK Rowling’s Harry Potter books may compete with other magical fantasy books for readership. However, the degree of competition experienced by the rightsholder depends on the strength of the IP right and the extent to which that differentiation is protected.69

C The Australian Purposive Approach to Functional Market Definition

Market definition has been explained by Australian courts as ‘a focusing process and the Court must select what emerges as the clearest picture of the relevant competitive process in the light of commercial reality and the purposes of the law’.70 The identification of a single relevant product market, even under competition law, is arguably a fiction as in reality few markets are homogenous.71 Therefore, the market should not be taken as a universal or fixed boundary of competitive activity.72 Although approaches seeking to define a single product market are unlikely to support clearer analysis of copyright markets, other aspects of competition law’s analysis of markets are worth exploring.

Under a ‘purposive approach’ to market definition applied in Australia, the market, as defined, should support the purpose of analysing competition as impacted

67 Hovenkamp, ‘Response: Markets in IP and Antritrust’ (n 16) 2146. See also Lemley and McKenna (n 17) 2097.

68 As Lemley and McKenna explain, the antitrust approach to market definition ‘derives from a classical conception of competition involving undifferentiated goods where producers compete solely on price and quality’ but ‘goods encumbered by IP rights are ... by definition, differentiated’: Lemley and McKenna (n 17) 2101.

69 Ibid 2103–4. Note that ‘product differentiation theory’ as applied to expressive markets is discussed in Part V(A) of this article.

News Ltd v Australian Rugby Football League Ltd (1996) 58 FCR 447, 479 (Burchett J).

71 Louis Kaplow has argued that the exercise of identifying a market in competition law is fundamentally flawed as it discards useful information in the process of making course determinations: Louis Kaplow, ‘Why (Ever) Define Markets?’ (2010) 124(2) Harvard Law Review 438; Louis Kaplow, ‘Market Definition: Impossible and Counterproductive’ (2013) 79(1) Antitrust Law Journal 361. Cf recognition that it is a flawed but nevertheless useful step: Malcolm B Coate and Joseph J Simons, ‘In Defense of Market Definition’ (2012) 57(4) Antitrust Bulletin 667; Alex Sundakov, ‘What Is the Magic of Market Definition?’ (2014) 22(2) Australian Journal of Competition and Consumer Law 162.

72 See Eastern Express Pty Ltd v General Newspapers Pty Ltd (1992) 35 FCR 43, 59 (Lockhart and Gummow JJ); Australian Competition and Consumer Commission v Liquorland (Australia) Pty Ltd [2006] FCA 826, [429] (Allsop J) (Market definition is not an ‘enquiry after the nature of some form of essential existence’).

by problematic conduct.73 In other words, ‘one should begin with the problem at hand and ask what identification of market best assists in analysing the processes of competition, or lack of competition, with which the case is concerned’.74 Market definition is ‘not an end in itself’, but has the purpose of aiding ‘the identification of competitive constraints on market power’.75

Furthermore, while much of the existing discussion of market definition tends to focus on the product market, it has been recognised in the competition law context that markets have several dimensions, namely product, geographic, functional, and temporal dimensions.76 The underappreciated functional dimension is particularly useful for advancing our analysis of copyright markets. The functional dimension of the market refers to the ‘distinct vertical stages of production and/or distribution (for example, manufacturers, wholesalers, retailers)’.77 The ultimate product reaching consumers may be the same, but the firms participating in the supply chain may not be competing in the same functional market.

The functional dimension of markets is usually tangentially mentioned and there are few resources providing detailed treatment to the issue, even in the competition law sphere.78 Nevertheless, in Australian Competition and Consumer Commission v Metcash Trading Ltd, Emmett J of the Federal Court explained that:

The identification of a relevant market by functional dimension requires consideration of the differences between horizontal economic relationships and vertical economic relationships. Horizontal economic relationships generally involve rivalry, whereas vertical economic relationships are generally cooperative. Generally, horizontal interactions are characterised by rivalry among entities selling the same or similar products to the same or similar customers. ... In contrast, vertical interactions are generally characterised by cooperation among entities engaged in a supply and acquisition relationship.79

73 See Seven Network [2007] FCA 1062, 601 [1763] (Sackville J): ‘[M]arkets are not defined in the abstract, but for the purpose of analysing the processes of competition relevant to the allegations of anti-competitive conduct made in the particular case.’ Note that this point has also been recognised in

US scholarship: Steven C Salop, ‘The First Principles Approach to Antitrust, Kodak, and Antitrust at the Millennium’ (2000) 68(1) Antitrust Law Journal 187, 191 (‘Market definition and market power should be evaluated in the context of the alleged anti-competitive conduct and effect, not as a flawed filter carried out in a vacuum divorced from these factors’).

74 Queensland Wire Industries Pty Ltd v Broken Hill Proprietary Co Ltd (1987) 17 FCR 211, 218–19 (Bowen CJ, Morling and Gummow JJ). See also Queensland Wire Industries Pty Ltd v Broken Hill Proprietary Co Ltd [1989] HCA 6; (1989) 167 CLR 177, 195 (Deane J); Boral Besser Masonry Ltd v Australian Competition and Consumer Commission [2003] HCA 10; (2003) 215 CLR 374, 495 (Kirby J).

75 Stephen Corones, ‘Competition Law and Market Regulation: Market Definition in the Age of the Internet’ (2010) 38 Australian Business Law Review 309, 311. As Corones has cautioned, ‘[t]he processes of competition are not affected by how one defines the market’: at 311.

78 Rhonda L Smith and Neville R Norman, ‘Functional Market Definition’ (1996) 4(1) Competition and Consumer Law Journal 1, 9. The authors observe that there is no prescribed test for determining the relevant functional level and that the SSNIP test (as applied to product and geographic market definition) does not assist with the identification of functional markets: at 10.

Emmett J further emphasised that ‘care must be taken to ensure that different functional levels are not combined into a single product market’, as this ‘may violate the principle of identifying the smallest market’.80

The purposive approach is crucial to the identification of the relevant functional dimension of a market. This is illustrated by the trial judge’s error in the Australian case of News Ltd v Australian Rugby League Ltd (‘Super League Trial Case’),81 in which focus remained predominantly on the product dimension. In this case, News Ltd sought to establish a new rugby league competition called the Super League for pay TV, in competition with the existing Australian Rugby League Ltd (‘ARL’) and New South Wales Rugby League Ltd. The ARL implemented Commitment and Loyalty Agreements with 20 clubs requiring that the clubs refrain from participating in rival rugby league competitions for five years (prior to that, clubs only entered into yearly contracts). News Ltd alleged that the agreements contravened provisions in the Competition and Consumer Act 2010 (Cth) against exclusionary and anti-competitive agreements, and that this conduct was a misuse of market power. In defining the market, Burchett J of the Federal Court focused on the downstream functional market, that is the selling of services to spectators, broadcasters and sponsors, and considered demand substitution for spectator sports.82 Focusing on the product market from a demand perspective, different types of sports such as basketball and cricket appeared substitutable with rugby within a broader market for spectator sports.83 With a broader market encompassing a wider range of spectator sports as substitutes, the anti-competitive effect of the conduct became less apparent.

However, the problematic conduct here was the implementation of provisions restricting clubs and players from participating in other rugby leagues. Therefore, taking a purposive approach, the relevant perspective should have been that of the clubs and players. Specifically, if ARL lowered its price for rugby league player services, could the players switch to another sport?84 Under a purposive approach to market definition, one should instead consider the supply of player services as an input into league sport (each constituting two separate functional

81 News Ltd v Australian Rugby Football League Ltd (1996) 58 FCR 447 (‘Super League Trial Case’).

It should be noted that Burchett J’s trial decision was overturned on grounds that did not require a reconsideration of market definition: News Ltd v Australian Rugby Football League Ltd [1996] FCAFC 870; (1996) 64 FCR 410, 425 (The Court) (‘Super League Appeal Case’).

83 Super League Appeal Case [1996] FCAFC 870; (1996) 64 FCR 410, 490–6 (The Court). Note that Burchett J recognised at Super League Trial Case (1996) 58 FCR 447 that players of hockey, for example, do not compete with players of basketball or football for places in hockey clubs, but considered this finding to be confined to the US case referred to, ie, Philadelphia World Hockey Club Inc v Philadelphia Hockey Club Inc, 351 F Supp 462 (ED Pa, 1972). David Brewster has explained that ‘[t]his failure [to address supply side substitutability] can be seen to be a major omission in the analysis of the market involved’: David

dimensions), and utilise the perspective of those impacted by the problematic anti- competitive conduct. The functional dimension most affected by the arrangement was competition among players and clubs for league player services. A simplified diagram indicating the relationships between players, clubs, leagues, and spectators (focusing on rugby with other sports in the periphery) is in Figure 1 below.

The Super League cases can be contrasted with the United Kingdom case of Hendry v World Professional Billiards & Snooker Association Ltd,85 brought under the Competition Act 1998 (UK) and European Community Treaty provisions against abuse of market dominance.86 The World Professional Billiards and Snooker Association was an organiser of professional snooker tournaments. Similar to the ARL, it had sought to prevent a prospective rival from setting up their own tournament series through the implementation of association rules. These included restrictions on the freedom of players to decide which tournaments to play. Importantly, the Court considered supply substitution, ie, the fact that commercial tournament organisers competed for players’ services as the critical input to a tournament,87 and concluded that snooker players’ services was the relevant product market.88 The case highlights the importance of considering supply-side substitution, and not just demand-side substitution, when identifying and analysing markets.89 The question posed from a supply perspective is, whether

85 Hendry v World Professional Billiards & Snooker Association Ltd [2002] ECC 8 (‘Hendry’).

86 Competition Act 1998 (UK) s 2; Consolidated Versions of the Treaty on European Union and of the Treaty Establishing the European Community [2002] OJ C 325/01, arts 81–2.

89 That is ‘competition may proceed not just through the substitution of one product for another in use (substitution in demand) but also through the substitution of one source of supply for another in

‘the same production and distribution facilities can sometimes be used to produce and sell two or more products even though buyers [the demand-side] do not regard them as good substitutes’.90 Therefore, consideration of the supply-side in defining the market takes into account the ability of firms to shift from the production of one product to another product.91

Drawing from this discussion of competition law, a key lesson for copyright is that courts should not focus solely on identifying the product that competes for consumer demand. Each functional level could raise different considerations, depending on the anti-competitive conduct in question and the party or parties who are primarily impacted by that conduct. Therefore, it is important that courts apply a purposive approach to identifying the relevant functional market dimension, and in so doing, recognise the source of the anti-competitive conduct and to whom that conduct is directed.

V IDENTIFYING COPYRIGHT’S FUNCTIONAL MARKET DIMENSIONS

Taking inspiration from analytical tools developed under Australian competition law, we could adopt a purposive approach that differentiates the relevant functional dimensions of copyright markets. With the functional levels of copyright markets in mind, one may visualise a supply chain for the transformation of ideas into expressive intellectual capital that is then disseminated to the public through different means. More akin to a web of relationships than a chain, this is diagrammatically represented in a simplified manner in Figure 2 below.

production or distribution (substitution in supply). ... In an economist’s language, both cross-elasticity of demand and cross-elasticity of supply are relevant’: Re Tooth & Co Ltd [1979] ATPR 18,174, 18,196 (Keely J, Deputy President, Shipton and Brunt, Members), cited in Super League Trial Case (1996) 58 FCR 447, 477–8 (Burchett J). On the importance of supply substitution, see Queensland Wire Industries Pty Ltd v Broken Hill Proprietary Co Ltd [1989] HCA 6; (1989) 167 CLR 177, 199 (Dawson J). See also Toohey J at 210: ‘[D]emand substitutability has often been emphasised at the expense of supply substitutability’. The Australian approach to market definition is said to recognise supply-side substitutability more explicitly

Conventionally, use of the term ‘market’ in copyright infringement discussions entails a conceptual leap that directly links consumer demand to authorship incentives, without considering the market structures needed to connect authors to copyright consumers. This leads us to concentrate on competition occurring between authors or producers of expressive content in copyright disputes and the impact on incentives to create, even if the legal issue may have a more immediate effect on competition between disseminators. Copyright law’s focus on substitution in expression is akin to the tendency of competition law analysis to centre on demand substitution in product markets. Both approaches focus on an ultimate outcome instead of the processes and steps needed to achieve that outcome (be it consumer welfare or public access to content).

Audience demand for different forms of expression may be obvious considerations when initially approaching a copyright dispute. However, expressive competition is but one functional dimension of copyright markets, and current analysis focusing on expressive competition provides a partial understanding of the issues at stake. A single-minded focus on expressive competition could obscure other important underlying policy consideration. Therefore, recognising dissemination as a separate functional dimension provides a more holistic understanding of copyright markets.

92 Authors and producers have been separated into two levels in this diagram, although they could be the same party in many instances. Broadly, the term ‘producers’ captures firms that invest in producing copyright outputs, eg, record companies, publishers, and film producers, while ‘authors’ refer to those who are usually individual creators. It should be noted that either could be substituted by an owner or exclusive licensee – a third party that has acquired copyright or licensed the exclusive right from the author or producer.

Importantly, different copyright infringement elements may have disparate impacts on competition along these dimensions. Drawing from Australian and US jurisprudence, the discussion that follows contrasts: (1) the similarity requirement when assessing infringement of the reproduction right; and (2) the elements of infringement of the right to communicate to the public.93 It shows that the similarity requirement has a greater impact on expressive competition between works of authorship, while the communication right largely impacts disseminative competition between different means of communicating copyright content.

A Expressive Competition: Substantial Similarity

The substantial similarity element under US copyright law is relevant where non-identical copying has occurred. Infringement of the reproduction right requires ‘(a) that defendant [to have] copied from the plaintiff’s copyrighted work and (b) that the copying (assuming it to be proved) went so far as to constitute improper appropriation’.94 The latter ‘improper appropriation’ element is also known as a requirement that there be ‘substantial similarity’ between the works.95 The US legal test has at times exhibited an explicit demand substitution analysis, asking whether the defendant’s work is a substitute in the market for the plaintiff’s work.96 In other words, is there a degree of similarity such that a consumer would readily choose to obtain the defendant’s work, thereby using it as a substitute for the plaintiff’s work?97 In determining the relevant perspective, ie, the hypothetical person assessing similarity, US courts have fluctuated between a layperson and a member of the

93 Select elements of primary copyright infringement will be referred to here. It should be noted that the aim is not to provide an exhaustive discussion of infringement.

95 It should also be noted that under Australian law, scope for a substitutability analysis is more limited as courts do not compare the similarity of the works as a whole but consider only the part allegedly copied by the defendant: see, eg, EMI Songs Australia Pty Ltd v Larrikin Music Publishing Pty Ltd [2011] FCAFC 47; (2011) 191 FCR 444. Therefore, infringement may be found even if the defendant’s work is not a substitute for the plaintiff’s work in the relevant market. Cf the ‘ordinary observer test’ under US law, where courts assess the overall similarity between the two works: Shyamkrishna Balganesh, ‘The Normativity of Copying in Copyright Law’ (2012) 62(2) Duke Law Journal 203, 217, citing, eg, Horgan v MacMillan Inc, 789 F

2d 157, 162 (Feinberg CJ) (2^nd Cir, 1986). Balganesh explains the uncertainty in determining whether the ‘ordinary-observer’ should be a layperson or ‘intended audience’ member: Balganesh, ‘The Normativity of Copying in Copyright Law’ (n 95) 219–20. What is clear is that US judicial approaches to ‘substantial similarity’ are far from settled: see Matthew Bender, LexisNexis, Nimmer on Copyright (online at 30 October 2019) [13.03].

96 This is arguably a historical approach that has been superseded. It has been observed that rather than asking if the later work is a duplicate of the original’s appeal, courts now tend to ask if the relevant target audience would recognise an element of the work as being too similar to the earlier work’s expression: Lunney Jr, ‘Reexamining Copyright’s Incentives-Access Paradigm’ (n 1) 545. Lunney, at 545–6, further argues: ‘As a result, what began as a system to protect authors against copying competitors has become

a system that attempts to ensure that the author can control every valuable use of her work.’ Note also criticism that consideration of substantial similarity in these terms overlaps significantly with the fair use determination: Amy B Cohen, ‘Masking Copyright Decisionmaking: The Meaninglessness of Substantial Similarity’ (1987) 20(4) University of California Davis Law Review 719, 745–6.

97 It should be noted that there is scholarly support for explicitly taking market substitution into account in assessing similarity under the reproduction right: see Jeanne C Fromer and Mark A Lemley, ‘The Audience in Intellectual Property Infringement’ (2014) 112(7) Michigan Law Review 1251.

‘intended audience’.98 The ‘intended-audience approach’ reflects the notion of copyright as a marketable right, as its underlying logic is that copying should be prevented ‘only when it results in the creation of close substitutes that are in turn likely to divert demand away from the original’.99 The test is buoyed by an assumption that market substitution analysis ought to be undertaken with the aim of protecting owners or authors, in line with the property-centric conception of markets that emphasise the impact on profits.100

As Herbert Hovenkamp has rightly observed, market analyses in IP infringement have focused on the degree of substitution between the outputs of rightsholder and infringer (or ‘range of interfirm rivalry’), not the presence or absence of market power.101 This narrow substitution analysis comparing the plaintiff’s work and the defendant’s infringing product is much more limited when compared to a broader analysis of competition within the relevant market. The consideration of expressive competition is arguably implicit in copyright, evident in the idea/ expression dichotomy or notion of thin copyright and the merger doctrine etc.102 Encapsulated within these doctrines is a concern that owners would hold too much market power,103 should copyright protection be extended to ideas or bare information. Nevertheless, market-based analysis should perhaps be made explicit in difficult cases.

A body of scholarship on ‘product differentiation theory’ considers how the scope of copyright protection affects market conditions, in particular, the diversity of expression vying for consumer demand. These analyses of copyright’s role in encouraging entry into markets for expression has resulted in conflicting proposals. Christopher Yoo has proposed that copyright law promote greater differentiation (ie,

99 Ibid, citing Michael Der Manuelian, ‘The Role of the Expert Witness in Music Copyright Infringement Cases’ (1988) 57(1) Fordham Law Review 23, 144–5; Dawson v Hinshaw Music Inc, 905 F 2d 731 (4^th Cir, 1990) 733–4 (Murnaghan J): ‘In light of copyright law’s purpose of protecting a creator’s market, we think it sensible ... that the ultimate comparison of the works at issue be oriented toward the works’ intended audience’.

100 Cf Fromer and Lemley (n 97). Fromer and Lemley have called for approaches to infringement that consider not only technical similarities from an expert perspective, but also a consumer perspective that incorporates substitution analysis: at 1302. The rationale is that this ensures infringement is only found if the rightsholder is likely to suffer market harm, but as discussed in Part III(B) of this article the market- harm rhetoric can be problematic.

101 Hovenkamp, ‘Response: Markets in IP and Antitrust’ (n 16) 2137. See also where Hovenkamp notes ‘questions about market impact may be relevant (under fair use) ... however, the issue is not power but rather the impact of the infringer’s sales on the rights holder’: at 2149.

102 See Pamela Samuelson, ‘Reconceptualizing Copyright’s Merger Doctrine’ (2016) 63(3) Journal of the Copyright Society of the USA 470, 459–60 (recognising the promotion of competition as a function of the US merger doctrine); Jamie Lund, ‘Copyright Genericide’ (2009) 42(2) Creighton Law Review 131, 154–5 (considering the relation between ‘thin copyright’ and ‘dynamic market definition’).

103 A firm is said to have market power ‘when it can behave persistently in a manner different from the behaviour that a competitive market would enforce on a firm facing otherwise similar cost and demand conditions’: Carl Kaysen and Donald F Turner, Antitrust Policy: An Economic and Legal Analysis (Harvard University Press, 1959) 75, quoted in Queensland Wire Industries Pty Ltd v Broken Hill Proprietary Co Ltd [1989] HCA 6; (1989) 167 CLR 177, 200 (Dawson J); Melway Publishing Pty Ltd v Robert Hicks Pty Ltd (2001) 205 CLR 1, 20–1 [41]–[42] (Gleeson CJ, Gummow, Hayne and Callinan JJ); Boral Besser Masonry Ltd v Australian Competition and Consumer Commission [2003] HCA 10; (2003) 215 CLR 374, [265]–[267] (McHugh J).

allowing more close resemblance between works) but reward the creators of this greater density of differentiated works with stronger exploitation rights (eg, broader interpretations of the communication right).104 Michael Abramowicz, on the other hand, has viewed broader exploitation rights as inducing wasteful entry into such markets, as up to a certain point saturating the market with a greater variety of works is duplicative and not socially beneficial.105 These discussions about differentiated competition and market entry are insightful, but the focus remains largely on calibrating expressive competition, ie, the diversity of expression that satisfies consumer demand. The scope of exploitation rights such as the right to communicate to the public are tangentially addressed, and competition in dissemination markets (as impacted by the scope of such rights) is not analysed in detail.106

Drawing from the scholarship on product differentiation theory, similar questions could be applied to disseminative competition. To what extent should copyright law encourage the development of differentiated dissemination means to advance copyright’s public access objectives? This question has not been given a consistent level of attention and is not embedded in our analysis of copyright markets.107 Well-functioning markets are instrumental to the achievement of copyright’s public access policies. Therefore, the level of competition at all levels of the copyright ecosystem (not just the initial stage of creation) should be of concern to both law and policymakers.

B Disseminative Competition: Communication to the Public

An exclusive right of copyright owners that has a palpable impact on disseminative competition is the right to communicate to the public. Following its international introduction in 1996 through the WIPO Internet Treaties,108 this communication right, which includes the right to make copyright content available

105 Michael Abramowicz, ‘An Industrial Organization Approach to Copyright Law’ (2004) 46(1) William and Mary Law Review 35, 110; Michael Abramowicz, ‘A New Uneasy Case for Copyright’ (2011) 79(6) George Washington Law Review 1644, 1647–8. See also Michael Abramowicz, ‘A Theory of Copyright’s Derivative Right and Related Doctrines’ (2005) 90(2) Minnesota Law Review 317. Abramowicz also favours an approach to the reproduction and derivative works rights which relaxes the similarity requirement, ie, a lower level of similarity would be sufficient to prove infringement.

106 The scholarship on copyright and product differentiation theory may be seen as an attempt to link expressive and disseminative competition, albeit in an implicit manner. For instance, Yoo describes product differentiation theory as addressing the ‘structural interrelationship’ between the access and incentives sides of the trade-off: Yoo (n 104) 276. Note that Abramowicz considers dissemination but refers to it as a secondary goal: ‘[A]lthough copyright law’s paramount goal may be to increase

incentives for the production of new works, this goal may not be of as much significance at the margin, and relatively more attention should be paid to ensuring dissemination’: Abramowicz, ‘An Industrial Organization Approach to Copyright Law’ (n 105) 41.

107 This question arises in copyright cases and scholarship when new technologies arise to threaten the status quo: see, eg, Michael A Carrier, ‘Copyright and Innovation: The Untold Story’ [2012] (4) Wisconsin Law Review 891 (analysing the effect of the Napster decision on innovation in the music technology space).

108 WIPO Copyright Treaty, signed 20 December 1996, 2186 UNTS 121 (entered into force 6 March 2002) art 8; WIPO Performances and Phonograms Treaty, signed 20 December 1996, 2186 UNTS 203) (entered into force 20 May 2002) arts 10, 14. See, eg, article 8 of the WIPO Copyright Treaty, which provides:

to the public, is present in somewhat different forms across various jurisdictions.109 Nevertheless, the right can be distilled into two key elements: (1) an ‘act’ of communication that is (2) to ‘the public’. Satisfaction of both elements is required to prove infringement of the communication right.

‘The public’ has been assessed with reference to the expected scope of the copyright owner’s market, including the licensing market. In Australia, the High Court in Telstra v Australasian Performing Right Association established the ‘copyright owner’s public’ test for assessing the scope of ‘the public’.110 The Court held that the ‘copyright owner’s public’ encompassed circumstances where the copyright owner can expect to receive a fee, regardless of whether the transmission is made to only one person or a private audience.111 The case involved music-on- hold communicated to individual members of the public through their telephone sets, and the Court found that such a performance would ordinarily be to the financial disadvantage of rightsholders because the rightsholder is entitled to expect payment for the work’s performance.112

A somewhat similar perspective on infringement is found in the seminal US case American Broadcasting Companies v Aereo (‘Aereo’),113 which involved a time-shifting/streaming service that utilised multiple user directories and thousands of dime-sized antennas to emulate a ‘private’ system of communications. Again focusing on demand substitution but with owners’ interests at the forefront, the Supreme Court remarked that the ‘behind-the-scenes way in which Aereo delivers television programming to its viewers’ screens ... does not significantly alter the viewing experience of Aereo’s subscribers’.114 The Court explained that the technology used ‘[does] not render Aereo’s commercial objective any different from that of cable companies’.115

Once again there is an implicit consideration of demand substitution, with an underlying objective of protecting the copyright owner’s market. Where consumers see the means of dissemination as substitutable, it is then argued that the owner can expect to receive a fee from that communication to the consumer. The US approach implicitly considers demand for the type of service, while Australian courts go a

[A] uthors of literary and artistic works shall enjoy the exclusive right of authorizing any communication to the public of their works, by wire or wireless means, including the making available to the public of their works in such a way that members of the public may access these works from a place and at a time individually chosen by them.

109 While Australia introduced a communication right explicitly encompassing ‘making available’ in its copyright legislation, the US relies on a suite of rights, including the right of public performance and public display: see Copyright Act 1968 (Cth) ss 31(1)(a), 10(1). Cf Copyright Act of 1976, 17 USC § 106. See also United States Copyright Office, ‘The Making Available Right in the United States: A Report of the Register of Copyrights’ (Report, February 2016) <https://www.copyright.gov/docs/making_available/ making-available-right.pdf>.

110 ‘Is the audience one which the owner of the copyright could fairly consider a part of his [or her] public?’: Telstra Corporation Ltd v Australasian Performing Right Association Ltd [1997] HCA 41; (1997) 191 CLR 140, 155–6 (Dawson and Gaudron JJ).

113 American Broadcasting Companies Inc v Aereo Inc, 134 Sup Ct 2498 (2014) (‘Aereo’).

step further to explicitly consider demand for a licence. Having the rightsholder’s expectation of payment drive the analysis is circular.116 Broader consideration of market structures for the dissemination of copyright content is absent from such approaches to the communication right.

C Comparing Expressive and Disseminative Competition

Key differences between the functional dimensions of competition in copyright markets are apparent, in light of the preceding analysis of substantial similarity and communication right.117 The question of similarity in expression focuses on the boundaries of the copyright subject matter that ought to be protected. In contrast, other legal elements, such as what is a communication to ‘the public’, relate to the ability to exploit the subject matter and the range of dissemination means that will be subject to the copyright owner or exclusive licensee’s control. If we look at Aereo for example, a finding on the scope of the right dictated the extent to which an incumbent disseminator was able to exploit copyright content without third party interference with ‘their market’.

In addition, a finding on the scope of the communication right more immediately affects a class of existing copyright content, with broader impact on industry structure and dealings. While a finding on substantial similarity may influence other creative practices to the extent that it establishes precedent on how similarities between works will be assessed, the comparison of expression is largely limited to the plaintiff and defendant’s works in question. Precedent in relation to the communication right on the other hand could be applied generally to a category of dissemination means that relies on similar technology. Therefore, the communication right’s coverage of certain means of dissemination could confer upon rightsholders or exclusive licensees a degree of control over dissemination markets.118

Contrasting the impacts of a finding of non-infringement in each instance highlights a further point. Where a court finds that there is no substantial similarity and therefore, no infringement, the defendant’s activity is beyond the copyright

116 The argument is that it provides no meaningful definition except to say that a communication or transmission to that public is within the owner’s exclusive right: Kimberlee Weatherall, ‘An End to Private Communications in Copyright? The Expansion of Rights to Communicate Works to the

Public: Part 2’ (1999) 21(8) European Intellectual Property Review 398, 402. See also Cheryl Foong, ‘The Making Available Right: Problems with “the Public”’ in John Gilchrist and Brian Fitzgerald (eds), Copyright, Property and the Social Contract: The Reconceptualisation of Copyright (Springer International Publishing, 2018) 265.

117 It should be noted that not all doctrines or legal elements fit neatly into the categories of impacting expressive competition or disseminative competition. The US fair use doctrine, for example, can be utilised to exempt a range of conduct: see, eg, Fox News Network LLC v TVEyes Inc, 883 F 3d 169 (2^nd Cir, 2018). In this case, the audiovisual snippets made accessible to subscribers were somewhat different from the full programming provided by television providers in an expressive sense, and the medium of an online and searchable database was also different. An in-depth discussion of fair use is beyond the scope of this article. Nevertheless, this conceptual framework and novel way of thinking about expressive and disseminative competition could spur greater understanding of a range of copyright doctrines, including fair use.

118 This could be due to exclusive rights over a substantial volume of content. Furthermore, ‘settled arrangements can work powerfully against distribution entrants’: Picker (n 3) 463.

owner’s control; this aspect is not unlike a finding of non-infringement under the communication right. However, the defendant’s creation (assuming it meets the threshold legal requirements) is still subject to copyright protection and therefore may be owned and controlled by the defendant. The expression can be exploited by the defendant in its own right for copyright rents. In comparison, non-infringement of the communication right means that the technology for dissemination is not subject to proprietary protection. It is not subject to the copyright owner or exclusive licensee’s control. The defendant may have secured a first-mover advantage, but copyright protection cannot be utilised to limit access to content via that dissemination means.

	Expressive Competition	Disseminative Competition
Protection of	Boundaries of subject matter	Ability to exploit exclusively via certain dissemination means
Breadth of Finding	Limited to the plaintiff’s and defendant’s respective works	Covers a class of copyright content
Impact of a Non- infringement Finding	Defendant can still protect their work under copyright	Third parties can utilise the dissemination means to communicate copyright content

A means of dissemination that is not subject to proprietary protection effectively becomes a zone of non-exclusivity – it is a level playing field upon which disseminators are able to build new business models. This zone of non- exclusivity may be described as an ‘innovation spillover’.119 An innovation spillover is a form of positive externality, an economic concept which refers to benefits (or costs, if it is a negative externality) ‘realized by one person as a result of another person’s activity without payment’.120 An externality is therefore, generally ‘not fully factored into a person’s decision to engage in the activity’.121

Conventional property theory posits that such externalities must be internalised, ie, captured by property rights, so that rational actors are incentivised by the market to act in certain ways.122 However, internalisation using existing property systems

122 The assumption is that if private property owners are encumbered with third-party costs and entitled to appropriate third-party benefits, then ‘their interests will align with the interests of society, and they will make efficient (social welfare-maximizing) decisions’: ibid 265. This is described as a strong version of Harold Demsetz’s normative thesis that private property rights that internalise all externalities are social welfare maximising and therefore desirable: at 264–5, citing Harold Demsetz, ‘Towards a Theory of Property Rights’ (1967) 57(2) American Economic Review 347.

and institutions means awarding rights to the incumbent, not broadening the range of actors that can generate social benefits from that externality. As Brett Frischmann and Mark Lemley explain, innovators tend to capture only a small portion of the social value of innovations and ‘[i]ndustries with significant spillovers generally experience more and faster innovation than industries with fewer spillovers’.123 Efforts to internalise such externalities could inadvertently limit the occurrences of such spillovers.124

Innovation spillovers come in a variety of forms and are not easily measured.125 As a result, innovation spillovers tend to be undervalued.126 This has implications for disseminative competition in copyright markets. The illusiveness of these spillovers makes it easier to ignore competition in dissemination markets and the innovation benefits that such competition generates. However, relegating disseminative competition to the ‘too hard’ basket is no longer a viable strategy for copyright law. We need to actively engage with the notion of disseminative competition and understand copyright’s role in fostering such competition.

VI COMPETITION IN DISSEMINATION MARKETS

The preceding discussion has highlighted how copyright doctrine plays a part in calibrating competition across different functional market dimensions. Honing in on dissemination markets, the discussion that follows explicitly considers the dynamic and technology-reliant nature of dissemination markets, emphasising the need for copyright law to recognise the value of disruptive innovations in such markets.

In addition, copyright law has tended to focus on demand substitution in order to protect incentives to create, rather than analyse competition in the market as a whole. Contemporary criticisms of competition law’s consumer welfare standard are instructive, as they show how an outcome-focused yardstick could obscure close

123 Frischmann and Lemley (n 119) 268–9, citing Dietmar Harhoff, ‘R&D Spillovers, Technological Proximity, and Productivity Growth: Evidence from German Panel Data’ (2000) 52(3) Schmalenbach Business Review 238, 258. Frischmann and Lemley go on to explain that spillovers create opportunities to be exploited by entrepreneurs, activities that are part of a virtuous cycle by ‘in turn creating new knowledge spillovers that support still more entrepreneurial activity’: at 269, citing Zoltan J Acs et al, ‘The Knowledge Spillover Theory of Entrepreneurship’ (Discussion Paper No 5326, Centre for Economic Policy Research, November 2005).

125 Ibid 259, citing, eg, William J Baumol, The Free-Market Innovation Machine: Analyzing the Growth Miracle of Capitalism (Princeton University Press, 2002) 5. Examples given by Frischmann and Lemley (n 119) are the temporal benefits once inventions are no longer under patents, geographic benefits such as contribution to the local economy, or interindustry benefits to unrelated fields: at 260–1. In essence, innovation spillovers are indeterminate public goods, unlike the market system for personal property rights (the nature and scope of which are generally quite clear), and therefore not easily subject to internalisation: at 272.

126 ‘The market mechanism exhibits a bias for outputs that generate observable and appropriable benefits at the expense of outputs that generate positive externalities. ... [P]otential positive externalities may remain unrealized if they cannot be easily valued and appropriated by those that produce them, even though society as a whole may be better off ...’: Brett M Frischmann, ‘An Economic Theory of Infrastructure and Commons Management’ (2005) 89(4) Minnesota Law Review 917, 988–9.

consideration of market structures. Rather than focus on immediately perceivable, short-term outcomes, the key task should be protecting market structures that allow innovation to thrive in dissemination markets, such that consumer welfare is advanced in the long run.

A Fostering Schumpeterian Competition

The way in which copyright content arrives in the hands of consumers is significantly driven by technologies that change and are superseded over time. This technological change forces dissemination markets to evolve, and in this sense competition in dissemination markets differs markedly from competition in expressive markets. In the face of disruptive technological change, the type of competition that takes place is Schumpeterian competition, whereby competition is occurring for the market (not just in the market).127 Joseph Schumpeter, in 1942, coined the concept of innovation as ‘creative destruction’ describing the ‘process of industrial mutation ... that incessantly revolutionizes the economic structure from within, incessantly destroying the old one, incessantly creating a new one’.128 Rather than tinkering at the margins of existing dissemination means, the level of differentiation brought about by such innovation upends the foundation of existing dissemination means.129

Copyright dissemination markets may go through phases, periods of stability that are undermined as new disruptive and more efficient means of dissemination arise. This accords with William Abernathy and Kim Clark’s observation, building on the work of Schumpeter, that ‘innovation is not a unified phenomenon: some innovations disrupt, destroy and make obsolete established competence; others refine and improve’. 130 The former is characteristic of Schumpeter’s ‘creative destruction’, but firms may transition between modes of innovation when ‘the focus of innovation shifts from meeting emerging needs with new concepts, to refining, improving and strengthening the dominant design and its appeal in the market’.131 The tendency for products to follow a life cycle that culminates in a ‘dominant design’ has been documented by business management and economics scholars.132 This product lifecycle theory posits that at the early stages, uncertainty

127 It should be noted that Ariel Katz has considered the magnitude of substitutive competition and Schumpeterian competition at different stages of a copyright work’s life cycle, primarily focusing on expressive competition: Ariel Katz, ‘Substitutions and Schumpeterian Effects over the Life Cycle of Copyrighted Works’ (2009) 49(2) Jurimetrics Journal of Law, Science and Technology 113.

128 Joseph A Schumpeter, Capitalism, Socialism and Democracy (Routledge, 1996) 83 (emphasis omitted).

129 This accords with Schumpeter’s argument that for creative destruction, the competition which counts is ‘competition which commands a decisive cost or quality advantage and which strikes not at the margins of the profits and the outputs of the existing firms but at their foundations and their very lives’: ibid 84.

130 William J Abernathy and Kim B Clark, ‘Innovation: Mapping the Winds of Creative Destruction’ (1985) 14(1) Research Policy 3, 4.

132 Product lifecycle theory is said to originate from William Abernathy and James Utterback’s 1978 business article: William J Abernathy and James M Utterback, ‘Patterns of Industrial Innovation’ (1978) 80(7) Technology Review 41. It should be noted that the theory has been subject to criticism: see, eg, Nariman K Dhalla and Sonia Yuspeh, ‘Forget the Product Life Cycle Concept!’ (1976) 54(1) Harvard Business Review 102.

in the market leads to a variety of new and novel product designs (reflecting firms’ different bets on the future).133 As a dominant design emerges, production processes are standardised and the market becomes more homogenous.134 Industries tend to become concentrated at this later stage of stability.135

Similarly, in the copyright context, we tend to see an increasing centralisation and homogenisation of content dissemination models following periods of disruption.136 In other words, copyright industries are not immune to observed product lifecycle stages, and for dissemination markets it is the transition between the stability of the existing model of dissemination to the emergence of a new model which deserves closer attention. In the face of technological change, leading companies often strive to satisfy mainstream customers and fail to recognise the value of innovations that appeal to smaller or emerging markets.137 This explains why disruptions must often originate from an outsider, as incumbent firms will strive to protect the market from change.138 Furthermore, consumer preferences as to how they access copyright content will only be taken into account by incumbent disseminators if their conduct is conditioned by competition from rival disseminators.139 Without disseminative competition, experimentation by rivals and an incentive to meet consumer demand, incumbents are able to focus more on bolstering existing dissemination models and less on satisfying emerging consumer preferences.

B Protecting Competitive Market Structures

Copyright law ought to explicitly value and encourage experimentation by disseminators that respond to consumer demand. Such experimentation could lead to a range of innovation spillovers that the law cannot anticipate. Copyright law need not drive this experimentation but should support market structures that provides space for experimentation. Laws directed at market structures may be contrasted with laws aimed directly at achieving broad outcomes.

133 Paul Windrum and Chris Birchenhall, ‘Is Product Life Cycle Theory a Special Case? Dominant Designs and the Emergence of Market Niches through Coevolutionary-Learning’ (1998) 9(1) Structural Change and Economic Dynamics 109, 110.

134 Franco Malerba et al, ‘Demand, Innovation, and the Dynamics of Market Structure: The Role of Experimental Users and Diverse Preferences’ (2007) 17(4) Journal of Evolutionary Economics; Heidelberg 371, 372.

136 At the initial stages of disruption, communications may be decentralised, but these communications will ultimately tend towards centralisation: Deven R Desai, ‘The New Steam: On Digitization, Decentralization, and Disruption’ (2014) 65(6) Hastings Law Journal 1469.

137 Joseph L Bower and Clayton M Christensen, ‘Disruptive Technologies: Catching the Wave’ Harvard Business Review (Web Page, 1 January 1995) <https://hbr.org/1995/01/disruptive-technologies-catching- the-wave>. For more in-depth discussion of disruptive innovation, see Clayton M Christensen, Innovator’s Dilemma: When New Technologies Cause Great Firms to Fail (Harvard Business Review Press, 1997).

138 For a more detailed discussion of the value of decentralised decision making in IP and innovation markets, see Tim Wu, ‘Intellectual Property, Innovation, and Decentralized Decisions’ (2006) 92(1) Virginia Law Review 123.

139 It has been observed that interactions between consumer and firm enable ‘coevolutionary learning’, which influences the development of a dominant design or leads to the maintenance of several designs within specific market niches: Windrum and Birchenhall (n 133) 115.

Competition law perspectives are again instructive in this regard. In the competition context, the focus on outcomes is embodied in the consumer welfare standard, which posits that anti-competitive practices should be prohibited or regulated in order to protect consumer welfare. US antitrust scholars such as Lina Khan and Tim Wu have underscored the weakness of a consumer welfare standard that is measured in terms of short-term price effects on consumers only.140 Khan explains that ‘[f]oundational to this view is a faith in the efficiency of markets, propelled by profit-maximizing actors’.141 Wu argues that under a narrow view emphasising efficiencies and consumer welfare, there is limited ‘consideration of the “dynamic” costs of monopoly, like stagnation or stalled innovation’, and the virtues of avoiding central planning are lost.142 Harm is equated with how a firm chooses to exercise its market power through price based levels, and whether a firm has developed this power and distorted the competitive process in some other way is not given due consideration.143

A central point that permeates both Wu and Khan’s scholarship is that the law should encourage market structures that support the competitive process, and not equate competition with envisioned outcomes or value.144 The legal system is better equipped to eliminate subversions and abuses of a process, as opposed to ‘ascertaining values that can be exceedingly difficult, if not impossible, to measure’.145 An analogous point can be made about copyright law: the access- incentive paradigm calls upon judges and lawyers to measure the public benefits flowing from highly complex and diverse transactions.146 Uncritical reliance on the consumer welfare standard in competition law and over-emphasis on protecting incentives in copyright encourage us to bypass valuable consideration of market structures, as illustrated in Figure 3 below. Analysis of market structures helps to bridge this divide between envisioned outcomes and the legal doctrine. Importantly,

140 Lina M Khan, ‘Amazon’s Antitrust Paradox’ (2017) 126(3) Yale Law Journal 710, 716. See also Tim Wu, The Curse of Bigness: Antitrust in the New Gilded Age (Columbia Global Reports, 2018) (‘The Curse of Bigness’).

141 Khan (n 140) 719, citing Richard A Posner, ‘The Chicago School of Antitrust Analysis’ (1979) 127 University of Pennsylvania Law Review 925, 932. Both Khan and Wu credit the Chicago School for inaccurately viewing antitrust problems through a lens of price theory. See Khan at 744: ‘the Chicago School shifted the analytical emphasis away from the process – the conditions necessary for competition

– and toward an outcome – namely, consumer welfare’ (emphasis in original). Wu’s criticism is at Wu,

144 Khan urges that we should not peg competition ‘to a narrow set of outcomes’, but ‘examine the competitive process itself’ by ‘analyzing the underlying structure and dynamics of markets’: ibid 717. Central to this argument is that ‘a company’s power and the potential anti-competitive nature of that power cannot be fully understood without looking to the structure of a business and the structural role it plays in markets’: at 717. Likewise, Wu’s core argument is that we should focus on protecting the

‘process [of competition], as opposed to the maximization of a value’: Wu, The Curse of Bigness (n 140) 136 (emphasis in original).

146 As Wu notes in relation to antitrust, ‘[i]n practice, the consumer welfare standard asks judges and lawyers to do something nearly impossible: to measure the welfare effects of highly complex transactions or conduct’: ibid 135. See also Khan (n 140) 739: ‘Even if we accept consumer welfare as the touchstone of antitrust, ensuring a competitive process – by looking, in part, to how a market is structured – ought to be key’.

an emphasis on protecting competitive market structures prevents the problem of abuse of market power, or in the copyright context, addresses strategic use of copyright exclusive rights to grow or bolster market power, as opposed to seeking to regulate it after the fact.147

In short, a preoccupation on achieving certain outcomes, without regard to market structures that facilitate such outcomes, may be counterproductive. Insufficient attention is paid to market structures that ultimately generate beneficial outcomes for consumers, and both copyright law and competition law are plagued by this problem, in different ways and to different extents.148

VII PROPOSALS FOR COPYRIGHT DISSEMINATION MARKETS

An important first step towards closer analysis of copyright markets is to recognise that markets are social and economic constructs; they are neither natural nor do they have inherent boundaries.149 Therefore, we should seek to understand

147 As Khan explains, ‘[f]ocusing primarily on price and output undermines effective antitrust enforcement by delaying intervention until market power is being actively exercised, and largely ignoring whether and how it is being acquired’: Khan (n 140) 738.

148 In Australia, a structuralist approach to determining the competitive impact of a firm’s conduct has prevailed, adopted by the Trade Practices Tribunal in Re QCMA (1976) 25 FLR 169, and accepted by the High Court of Australia in Queensland Wire Industries Pty Ltd v Broken Hill Proprietary Co Ltd [1989] HCA 6; (1989) 167 CLR 177, 188 (Mason CJ and Wilson J), 200 (Dawson J). The Australian Competition Tribunal

has further called for recognition of strategic behaviour, presciently stating in Re Application by Chime Communications Pty Ltd [No 3] [2009] ACompT 4, [12]:

It is often necessary to examine the forces that explain industry structure rather than to regard structure primarily as the determinant of conduct and performance. It is now widely recognised that firms behave strategically in order to shape the market environment to their own competitive advantage. ... This shift in thinking recognises the realities of modern markets and the observed strategic behaviour of firms.

However, this explicit recognition of strategic behaviour does not appear to have been widely adopted by Australian courts: see Duke (n 58) 50.

149 Bruce G Carruthers and Sarah L Babb, Economy/Society: Markets, Meanings, and Social Structure (Pine Forge Press, 2000) 2.

copyright markets and appreciate the competitive tensions calibrated by copyright law, rather than approaching market definition in the abstract. When interpreting the scope of an exclusive right, courts should scrutinise the extent to which proposed interpretations affect market structure.

In difficult cases involving new dissemination technologies, the disruption of existing market positions is often the key driver of the dispute. This takes place at a crucial transition period, a dynamic process that renders a static analysis of substitution insufficient. Exclusive rights may be used by copyright owners or exclusive licensees in a strategic manner to hinder disruptive innovations from developing and to protect their existing business models. Unfortunately, courts are not currently equipped to explicitly address this in copyright infringement cases.

In order to adequately analyse disseminative markets in copyright, two steps ought to be taken: (1) deepening our conceptual understanding of copyright dissemination markets within the existing doctrinal framework; and (2) introducing a rule of reason analysis to copyright infringement analysis that encompasses closer consideration of market structures.

A Recognising a Conceptual Blind Spot

The first step is to recognise copyright’s conceptual blind spot – a doctrinal insistence on protecting markets for content in the abstract. The fallacy is that the law need only set the entitlement and that markets will take care of the rest. Such rhetoric blindly accepts that broader rights will generate greater profits for creators, when such rights may primarily serve to strengthen and secure incumbent disseminators’ positions of market power.

In essence, we must be critical of our faith in copyright markets. For instance, when interpreting the meaning of ‘public’ in the communication right, market harms in the abstract do not justify a broader interpretation of the right. We should critically analyse the impact certain interpretations will have on markets for communication of content to the public. The argument that competitive dissemination market structures need to be preserved operates as a counterpoint to assertions that broader rights will secure incentives to create.

This conceptual shift may be instrumental to the development of legal elements or provisions that are not well-defined and open to interpretation. The points made here may be understood as mid-level principles to guide the development and interpretation of copyright infringement rules. However, a conceptual shift alone may not effect closer consideration of market structures. A methodology for objectively analysing competition in copyright dissemination markets would be needed.

B A Copyright Rule of Reason Analysis

A more substantive change would entail taking a page out of competition law’s approach and applying a ‘rule of reason’ analysis to copyright infringement.150

150 It should be noted that Christopher Sprigman has argued for a rule of reason approach to copyright analysis, but with a focus on expressive competition: Christopher J Sprigman, ‘Copyright and the Rule of Reason’ (2009) 7(2) Journal on Telecommunications and High Technology Law 317. Sprigman

Different liability standards are applied in competition law: one either applies a per se standard where the negative effects on competition are assumed to be clear, or a rule of reason analysis where the harms to competition are dependent on the context and not self-evident.151 In order to determine whether proposed conduct has the purpose, effect, or likely effect of substantially lessening competition in the relevant market under Australian competition law, the court has to:

1. consider the likely state of future competition in the market ‘with and without’ the impugned conduct; and

2. on the basis of such consideration, conclude whether the conduct has the proscribed anti-competitive purpose or effect.152

The court is looking into the future and asking whether the conduct will ‘reduce competitive rivalry that would otherwise have existed in the market but for the conduct in question’.153

A strict standard of liability, analogous to the competition law per se standard, is applied to primary infringement of copyright. One need only prove that the exclusive right has been exercised by the defendant without permission; the broader market as impacted by the decision is largely inconsequential. In other words, market harm is assumed to flow from infringement. A rule of reason analysis of copyright infringement undermines this assumption. It can bring to light the additional tensions arising, particularly in transition periods between existing and new means of dissemination. At the transition stage, the question is how copyright could have an impact on Schumpeterian competition. In these instances, courts should be equipped to go beyond the infringement tests to investigate the state of disseminative competition.154

Under a rule of reason analysis modified for copyright law, courts could identify the functional dimension that the plaintiff and defendant operate in and consider the impacts on disseminative competition in that copyright market. The question is this: what is the likely state of disseminative competition with the allegedly infringing conduct, and what is the likely state of disseminative competition without the allegedly infringing conduct? In explicitly analysing market structures

argues that we should apply default per se liability rules to consumptive uses of copyright, but other non-consumptive uses require a closer analysis of the harms to creator incentives. Consumptive uses are deemed as the use of exact copies, which emphasises promotion of expressive competition (harms to creator’s incentives) and ignores disseminative competition.

151 See Duke (n 58) 175–7. See above n 57 in regards to the High Court’s description of the ‘per se’ prohibitions.

152 Stirling Harbour Services Pty Ltd v Bunbury Port Authority [2000] FCA 1381, [12] (Burchett and Hely JJ).

153 Duke (n 58) 107 (emphasis in original); Stirling Harbour Services Pty Ltd v Bunbury Port Authority

154 One way of doing so is to legislatively implement this rule of reason test. However, where there is a significant amount of uncertainty under existing legislative provisions, considerations from this rule of reason test could feed into the pre-existing legal elements of infringement. An example would again be the right to communicate to the public, as courts in various jurisdictions have yet to develop clear

and transparent approaches to interpreting the right: see Cheryl Foong, ‘Volition and the “New Public”: A Convergence of US and EU Judicial Approaches to Communications to the Public’ (2020) 42(4) European Intellectual Property Review 230; Foong, The Making Available Right: Realizing the Potential of Copyright’s Dissemination Function in the Digital Age (n 2).

that support disseminative competition, and potential strategic measures taken to inhibit disseminative competition, the following questions are relevant:

In asking these questions, courts are not just relying on a static analysis of demand substitution but prompted to consider the market more broadly, including strategic conduct aimed at substantially lessening disseminative competition. This rule of reason analysis would require an investigation into the relevant market structures and add complexity to infringement disputes.155 Added complexity may not be welcome, but as disseminative markets for copyright content become increasingly centralised and dominated by a few firms, maintaining a blind eye to disseminative competition is not a sustainable way forward.

VIII CONCLUSION

What do ‘markets’ mean in the copyright context? With this question as a starting point, this article has exposed the limitations of the dominant conception of copyright markets. The tendency has been to prioritise copyright’s ‘property’ status, as a precondition to the existence of copyright markets. This property-centric view ignores the resulting condition of markets once copyright protection is established. This dominant perspective is woefully inadequate. In critically analysing the notion of markets in copyright, it is apparent that competition in copyright markets occurs across different functional dimensions. Expressive competition is but one functional dimension. Closer consideration of disseminative competition and expressive competition highlights the distinct policy tensions arising along these functional dimensions.

Current legal doctrine myopically focuses on demand substitution and substitutive harms faced by copyright owners. In dissemination markets, Schumpeterian competition, a form of competition that characterises transitions from old to new technologies, should be at the forefront of copyright analysis. One way to refocus our analysis of copyright law in dissemination markets is to develop and apply a copyright rule of reason analysis. Through this analysis, the negative impacts on competitive dissemination market structures and costs to

155 In addition, it has been argued that successful implementation of a rule of reason test in copyright law will require modification as to who bears the burden of proof: Stefan Bechtold, ‘Deconstructing Copyright’ in P Bernt Hugenholtz (ed), Copyright Reconstructed: Rethinking Copyright’s Economic Rights in a Time of Highly Dynamic Technological and Economic Change (Wolters Kluwer, 2018) 59, 80.

ongoing innovation should be explicitly recognised. During periods of disruption, third parties may reap certain benefits that are not directly accrued to copyright owners or exclusive licensees. However, these third parties bring spillover effects, generating positive innovation benefits for society.

To conclude, the term ‘markets’ has been used in copyright law to support assertions of market harm without explicit analysis of the impacts on dissemination. This article has provided a critical perspective on markets that distinguishes disseminative competition from expressive competition. Inspired by competition law’s more objective perspective on markets, the proposed conceptual and doctrinal reforms are important steps toward refining copyright’s role in an increasingly concentrated online environment.

University of New South Wales Law Journal

Foong, Cheryl --- "Reconceptualising Copyright Markets: Disseminative Competition as a Key Functional Dimension" [2021] UNSWLawJl 36; (2021) 44(3) UNSW Law Journal 1014

RECONCEPTUALISING COPYRIGHT MARKETS: DISSEMINATIVE COMPETITION AS A KEY FUNCTIONAL DIMENSION

I INTRODUCTION

II ‘MARKETS’ IN COPYRIGHT RHETORIC AND THEORY

III MARKET-BASED CALLS FOR REFORM

A Exclusive Right of Commercial Exploitation

B Infringement Test: Reasonable Expectations of Market Harm

IV LESSONS FROM COMPETITION LAW

A Conceptual and Doctrinal Differences

B Product Market Definition Not Directly Applicable

C The Australian Purposive Approach to Functional Market Definition

V IDENTIFYING COPYRIGHT’S FUNCTIONAL MARKET DIMENSIONS

A Expressive Competition: Substantial Similarity

B Disseminative Competition: Communication to the Public

C Comparing Expressive and Disseminative Competition

VI COMPETITION IN DISSEMINATION MARKETS

A Fostering Schumpeterian Competition

B Protecting Competitive Market Structures

VII PROPOSALS FOR COPYRIGHT DISSEMINATION MARKETS

A Recognising a Conceptual Blind Spot

B A Copyright Rule of Reason Analysis

VIII CONCLUSION