AustLII Home | Databases | WorldLII | Search | Feedback

Melbourne University Law Review

Melbourne Law School
You are here:  AustLII >> Databases >> Melbourne University Law Review >> 2001 >> [2001] MelbULawRw 16

Database Search | Name Search | Recent Articles | Noteup | LawCite | Author Info | Download | Help

Goss, Adrian --- "The Trade Practices Act 1974 (CTH) And The Treatment of Cluster Markets In The Australian Telecommunications Industry (2001) 25" [2001] MelbULawRw 16; (2001) 25(2) Melbourne University Law Review 481


CRITIQUE AND COMMENT

The Trade Practices Act 1974 (CTH) And The Treatment Of Cluster Markets In The Australian Telecommunications Industry

ADRIAN GOSS[*]

[The nature of the Australian telecommunications industry and the consumer environment in which it operates are such that the proper definition of the relevant market in respect of specific products will often be difficult to ascertain. In many cases, it may be appropriate to treat telecommunications markets as being constituted by multiple, rather than single, products. Thus, the concept of ‘cluster markets’ will often be relevant in the assessment of anti-competitive conduct within a telecommunications market.

This article argues that the Australian Competition and Consumer Commission’s failure and Australian courts’ lack of opportunity to articulate a set of criteria for determining when a market is properly defined as being constituted by a group of non-substitutable products calls into question the Australian Competition and Consumer Commission’s ability to coherently and consistently exercise its significant powers under Part XIB of the Trade Practices Act 1974 (Cth).]

CONTENTS

INTRODUCTION

The concept of ‘the market’ is central to identifying anti-competitive conduct within the Australian telecommunications industry for the purposes of Part XIB of the Trade Practices Act 1974 (Cth) (‘the Act’).[1] The nature of that industry and the consumer environment within which it operates are such that the proper definition of the relevant market in respect of specific products will often be difficult to ascertain. In particular, consumer preference for bundled services and technological convergence suggests that the boundaries of the relevant market for individual services will often be unclear. In many cases, it may be appropriate to treat telecommunications markets as being constituted by multiple, rather than single, products. Thus, the concept of ‘cluster markets’ will often be relevant in the assessment of anti-competitive conduct within a telecommunications market.

This article will outline the role of the market in the application of Part XIB of the Act. It will then consider the circumstances in which Australian and foreign courts have applied the concept of cluster markets. In particular, it will highlight the inadequate consideration given to cluster markets in Australia. Finally, it will outline the factors that make cluster markets especially relevant to the analysis of competition in the Australian telecommunications industry.

It is submitted that neither the Australian Competition and Consumer Commission (‘ACCC’) nor Australian courts have given adequate consideration to the ‘circumstances in which nonsubstitutes should be included in the same market.’[2] Those circumstances will, it is argued, be particularly relevant in determining the proper boundaries of telecommunications markets for the purpose of assessing anti-competitive conduct. The ACCC’s failure and the courts’ lack of opportunity to articulate a set of criteria for determining when a market is properly defined as being constituted by a group of non-substitutable products calls into question the ACCC’s ability to coherently and consistently exercise its significant powers under Part XIB of the Act.

II THE MARKET IN AUSTRALIAN TRADE PRACTICES LEGISLATION

The purpose of the competition provisions in the Act is to regulate competition and market power. Part XIB of the Act prohibits carriers and carriage service providers[3] from engaging in anti-competitive conduct in a telecommunications market. Accordingly, Corones has characterised the process of defining the market as ‘a preliminary first step to identify entry barriers, calculate market shares and provide a context in which competition is to be analysed.’[4] Identifying the proper boundaries of the market within which relevant conduct occurs is, therefore, a tool for assessing anti-competitive conduct rather than an end in itself.

The importance of properly characterising the market cannot be over-emphasised. Ascertaining the boundaries within which a product competes necessarily entails understanding the competitive pressures and relative power of the participants within the market that is sought to be defined. Mason CJ and Wilson J, in their joint judgment in Queensland Wire Industries Pty Ltd v Broken Hill Pty Co Ltd,[5] identified this. They warned that, while ‘it must be borne in mind that the object is to discover the degree of the defendant’s market power’, it remains the case that ‘[d]efining the market and evaluating the degree of power in that market are part of the same process’.[6] Thus, ascertaining the proper scope of the market is a necessary precondition to answering the ultimate question of whether the relevant conduct has an anti-competitive effect. In the ACCC’s words, defining the market is ‘fundamental to the assessment of alleged anti-competitive conduct.’[7]

The market can be said to underlie, to some extent, all aspects of regulation of anti-competitive conduct in the telecommunications industry. It is relevant to investigation by the ACCC and the Australian Competition Tribunal (‘Tribunal’) of whether a claimed public benefit outweighs the anti-competitive detriment entailed in specific, proscribed conduct;[8] to regulation of access to services under Part XIC of the Act;[9] and to judicial consideration of whether a contravention of one of the substantive provisions of Part XIB of the Act has occurred.[10] In addition, the ACCC will be required to identify the boundaries of the relevant market when making tariff-filing directions[11] and issuing competition notices in response to contraventions of the ‘competition rule’.[12]

The Act defines a ‘telecommunications market’ as a market in which carriage services, goods or services for use in connection with carriage services or access to facilities are supplied or acquired.[13] The term ‘market’ is defined to include ‘goods or services that are substitutable for, or otherwise competitive with, the first-mentioned goods or services’.[14] The basic, statutorily prescribed principle upon which market definition is based is, therefore, substitutability, or in Maureen Brunt’s words ‘the extent to which a firm can “give less and charge more” without its market being undermined by rivals’ incursions’.[15]

In accordance with this definition, the ACCC, in determining the proper boundaries of the market within which a carrier or carriage service provider operates, ‘is concerned to establish the actual and potential sources of close substitutes for the good or service in question.’[16] To do so, the ACCC tests the effect on both demand and supply of a notionally small but significant and non-transitory increase in price by a hypothetical monopolist (the ‘price incentive test’).[17] Thus, the ACCC seeks to identify cross-elasticities of demand and supply. That is, it seeks to identify the degree to which consumers or suppliers are likely to switch their consumption or supply of the product in question in response to a relative price change.

The price incentive test has been the subject of significant judicial consideration and has received strong judicial approval. In Australia, it was first articulated in Re QCMA by the Tribunal, which classified a market as ‘the field of actual and potential transactions between buyers and sellers amongst whom there can be strong substitution ... if given a sufficient price incentive.’[18] The Tribunal further developed this analytical approach in Re Howard Smith Industries Pty Ltd & Adelaide Steamship Industries Pty Ltd where it observed that ‘[i]f there is a variation in the relative prices of the goods and services offered for sale in a given market, then buyers can be expected to readily switch their custom from one seller to another or from one product to another.’[19] The Tribunal’s approach was subsequently approved by both the Federal Court, in Trade Practices Commission v Ansett Transport Industries (Operations) Pty Ltd,[20] and the High Court, in the Queensland Wire Industries Case.[21]

The ACCC’s approach to market definition need not necessarily reflect that taken by the courts. Maureen Brunt observes that the Act establishes ‘a rather sharp division of labour between the courts and the two administrative bodies’ of the ACCC and the Tribunal.[22] Indeed, as between the ACCC and the Tribunal, Corones notes a division in the application of the price incentive test. The Tribunal, he observes, tends to construe markets broadly while the ACCC construes them narrowly.[23] However, any divergence in the approach of the courts, the ACCC and the Tribunal goes only to the application, and not the appropriateness of applying, the price incentive test. In Australia, there appears to be a general acceptance of this test.

III THE CONCEPT OF CLUSTER MARKETS

In broad terms, a ‘cluster market’ may be described as a market in which the transaction costs involved in providing specific goods and services separately would be so great as to necessitate their provision together. Products which are, at least in a functional sense, non-substitutable would, in those circumstances, fall within the same market. Thus, the concept of cluster markets sits somewhat incongruously beside the definition of ‘market’ in s 4E of the Act, which emphasises but does not require substitutability of goods or services within a market, and the price incentive test.

Commenting on the nature of multiple-product markets, Brunt has observed that it ‘requires no great leap to think not just in terms of “the product” but in terms of a related group of products if the problem requires it.’[24] In Australia, it appears that the problem of market definition has not often required it. The paucity of Australian authority on the application of the cluster market concept to the analysis of anti-competitive conduct suggests that the best starting point for a consideration of cluster markets is foreign authority.

A United States Authority

In the United States, which has a much lengthier experience of regulation of anti-competitive behaviour, cluster markets have arisen for consideration on a number of occasions. Two leading United States cases have established that, in certain circumstances, it may be appropriate to treat a market as consisting of multiple rather than single products. In United States v Philadelphia National Bank,[25] the Supreme Court considered the appropriate market in which various banking goods and services, such as credit facilities, checking accounts and trust administration services, competed. The Court found that, rather than forming distinct markets, those goods and services comprised one unified market for ‘commercial banking’. Clustering those products together was justified, the Court argued, on the basis that the individual products were not subject to competition because of a ‘settled consumer preference’ for acquiring them together.[26]

In United States v Grinnell Corporation,[27] the Supreme Court was asked to consider the market in which fire and burglar alarm response services competed. Again, the Court found that those services were properly clustered together in a multiple-product market. A cluster market definition was appropriate, the Court reasoned, because ‘commercial realities’ necessitated a supplier supplying those services together.[28]

Philadelphia National Bank and Grinnell opened the way for cluster markets to be analysed in a number of different contexts. United States courts have subsequently identified cluster markets as existing in areas such as acute care hospital services,[29] the manufacture and sale of marine engines[30] and grocery stores.[31] In respect of telecommunications, however, there is little specific authority. Shiff, Ergas and Landrigan characterise United States v US West Inc[32] as one case in which a cluster market analysis may have been appropriate but was not used.[33]

In US West, the District Court of Columbia considered a complaint brought by the Department of Justice in which it was alleged that the proposed acquisition of Continental Cablevision by US West would substantially lessen competition in the market for dedicated services within a particular geographic area. The Department alleged that ‘the provision of dedicated services [is] a relevant product market.’[34] Applying the price incentive test, the Department found that there were ‘no other economically comparable alternatives available to a dedicated services customer.’[35]

However, the Department also acknowledged that the market for dedicated services comprised both ‘special access’ services and ‘local private line’ services.[36] The judgment gives no indication as to whether the validity of this clustering of two different services into the same market was questioned by the Court. Shiff, Ergas and Landrigan suggest that the Department included the services in the same market because a notional monopolist of both services would face higher transaction costs in supplying the services unbundled.[37] Accordingly, the notional monopolist would only supply the services as a bundle and it was appropriate to treat those services as being part of the same market.[38] The judgment in US West, therefore, implicitly accepts grouping of non-substitutable telecommunications products within a cluster market without articulating the rationale for doing so.

In the absence of an express, telecommunications-specific precedent for the application of cluster market analysis, one may have hoped for the establishment of broadly applicable, general criteria against which to analyse potential cluster markets. However, as Ayres observes, ‘while some cluster markets have been defined correctly, the lack of a sound justifying theory has led courts to adopt conflicting and ad hoc standards.’[39] Neither Philadelphia National Bank nor Grinnell, Ayres argues, provided acceptable rationales for clustering the products in question and, as a consequence, courts have found clusters based on the existence of ‘trade associations; census classifications; functional complementarity; common technology; distribution or marketing; a unique product group and other market characteristics.’[40]

Ayres proposes rationalising the approach to cluster markets by establishing a criterion of ‘transactional complementarity’.[41] Under that criterion, courts would only cluster those goods which, when purchased from a single supplier, are cheaper because consumers’ transaction costs are significantly reduced.[42] Although United States courts have, on some occasions, utilised the notion of transactional complementarity as a rationale for identifying the existence of a cluster market,[43] that criterion cannot yet be said to have received full judicial approval.

B New Zealand Authority

New Zealand courts have considered the concept of cluster markets on two occasions. In view of the similarities between Australian and New Zealand trade practices legislation, the New Zealand experience is particularly relevant to Australia.[44] In Commerce Commission v Port Nelson Ltd,[45] it was alleged that Port Nelson Ltd used its market power in relation to the provision of services in Port Nelson to force consumers to acquire those services. The High Court identified three services: pilotage, tug services and port services and facilities. Port Nelson Ltd argued that there was only one market for vessel movement services which comprised each of the three services identified by the Court. Asserting that the three services were not substitutes for each other, the Court rejected Port Nelson Ltd’s argument and found that each service represented a separate market.

Port Nelson is significant because the High Court appears to reject the very possibility of non-substitutes being included in the same market. ‘To a considerable extent,’ the Court reasoned, ‘the services are complementary ... However, they are not interdependent. Nor are they substitutable.’[46] Shiff, Ergas and Landrigan correctly observe that ‘[t]o the extent that the judgment suggests that services must be substitutes to be included in a market, the decision is clearly erroneous.’[47] Moreover, the High Court appears to suggest that the proper definition of the market may have been different if ‘proved commercial perception’ had treated the three services as part of the same market.[48] Although the Court questions whether such perceptions assist,[49] its approach does appear to reflect the ‘commercial reality’ approach which was adopted by the United States Supreme Court in Grinnell[50] and which is criticised by Ayres as being too imprecise and subjective.[51] Port Nelson does not, therefore, take us very far towards clarifying the circumstances in which it will be appropriate to define a market as consisting of a cluster of non-substitutable products.

In TruTone Ltd v Festival Records Retail Marketing Ltd,[52] the New Zealand Court of Appeal considered the relevant market for ‘top 50’ chart albums. The appellants were record album retailers in Auckland. The respondent, Festival Records, was a record distributor which had a six per cent share of total retail sales in the album market. The appellants alleged that, by refusing to supply them, Festival Records was using its dominant position in the market to deter competition. They argued that consumers would not substitute one top 50 album for another and that, therefore, each album constituted its own market. Festival Records argued, and the Court agreed, that the relevant market was the wider New Zealand album market.

In reaching its view, the Court pointed to the short time that an album remains popular and criticised the definition of the market proposed by the appellants as presenting a ‘snapshot rather than a moving picture of continuing commercial activity.’[53] The Court held that the rapidly shifting position of albums on the charts pointed to their substitutability, and that the appellant’s argument ignored commercial reality. Although the Court properly identified the relevance of the time dimension of the market,[54] its reasoning is not entirely satisfactory because the Court was forced to admit that ‘when an album is charting well a sizeable percentage of popular album purchasers will want that product and will not be prepared to substitute it.’[55] Thus, but for the criterion of time, to which the Court gave fleeting consideration, it would have found that the products were non-substitutes. It is submitted that a more consistent, and effectively identical outcome, may have been achieved if the Court had treated top 50 albums as non-substitutes which fell within one cluster market.

C Australian Authority

In Australia, courts have given little consideration to cluster markets. Despite this, the ACCC expressly acknowledges the possibility of markets comprised of non-substitutable products.[56] Moreover, the relevance and utility of cluster markets in Australian trade practices jurisprudence has been debated for over two decades. Walker observed in 1980 that Australian courts had not yet considered ‘this problem’ of cluster market definition but, he noted approvingly, the Trade Practices Commission had demonstrated a ‘tendency’ to treat non-substitutes as inhabiting distinct markets.[57] Accordingly, while it is established that cluster markets have a place in Australian competition law, that place remains poorly defined.

The issue of cluster markets has arisen in the context of assessing competition within the Australian banking industry. United States authority, particularly the decision in Philadelphia National Bank,[58] has been adopted to justify treating banking markets as consisting of a cluster of banking services. For example, the then Trade Practices Commission used the cluster market approach when considering whether the acquisition of Challenge Bank by Westpac Banking Corporation breached the merger provisions in s 50 of the Act.[59] Albrechtsen also notes that the ACCC may have used the cluster market approach when considering the merger between Westpac Banking Corporation and the Bank of Melbourne.[60]

Albrechtsen has criticised the use of the cluster market approach in respect of both mergers. She argues that ‘the rationale of the cluster approach grows increasingly questionable.’[61] Her criticisms, which are based on specific characteristics of the Australian banking industry and the divergent approaches to merger policy between the United States and Australia, have little relevance to consideration of cluster markets in the telecommunications industry. However, those criticisms underscore, in respect of the banking cases, the lack of a coherent rationale for clustering products. The ACCC’s position in those cases, she argues, appears to have been ‘erroneously guided by concerns of diversity and customer convenience.’[62] Those concerns, and their use in defining the market, subvert the proper inquiry which is the effect of conduct on efficiency within an industry.

The issue of cluster markets also arose in the Tribunal’s consideration of the provision of freight-handling services at Sydney airport.[63] There, the Tribunal considered, among other things, whether freight-handling services and cargo terminal operator services ‘constitute distinct functional markets, with their own economic and commercial characteristics, within a broad (or cluster) market for airport services as a whole in the Sydney region.’[64] The Tribunal found that the transaction costs involved in providing those services separately were such that those services could only be provided as a package and, accordingly, that they formed a cluster market.[65]

Despite the minimal use of the cluster market concept by Australian courts, the ACCC clearly contemplates circumstances in which it will be appropriate to define markets as being constituted by a cluster of products. In its information paper entitled Anti-Competitive Conduct in Telecommunications Markets, the ACCC observed that:

In some instances the only practical substitute for the goods or services purchased from one carrier or carriage service provider would be a grouping of several goods or services purchased from another (called ‘cluster markets’). This is likely to occur where the relevant product is in fact an aggregation of several distinct goods or services.[66]

That the ACCC’s understanding of when it will be proper to define a cluster market has not been more fully articulated is cause for concern. In the absence of guidance from either the courts or the chief regulatory body, academic opinion provides the most helpful framework in which to assess the utility of the cluster market concept in Australia.

Australian academic opinion initially appeared to have rejected the use of cluster market approaches to market definition. Considering the circumstances in which it would be sensible to adopt a cluster market definition, Walker observed in 1980 that, except where multiple items are sold as a single unit, such as sections of a newspaper, ‘clustering would seem to introduce yet another opportunity for subjective judgment into a field of law which suffers from quite enough unpredictability already.’[67] He noted approvingly the Trade Practices Commission’s approach of refusing to cluster products,[68] but criticised the Trade Practices Tribunal’s willingness to cluster products,[69] as demonstrated in Re Ford Motor Co of Australia Ltd[70] and Re Rural Traders Co-operative (WA) Ltd.[71] The basis of Walker’s criticism is his belief that cluster market definitions ‘complicate the analysis of competitive effect.’[72]

Walker’s criticism was briefly addressed three years later by Norman and Williams who agreed that clustering complicates the process of market definition, particularly in relation to determining market share, but noted that complexity alone was not a significant criticism.[73] Norman and Williams accepted that, given that business strategies take many different forms, clustering may be necessary and appropriate in some circumstances.[74] In their view, ‘many of the apparent problems associated with market definition involving multi-product firms can be resolved by considering that consumers assess “bundles” of characteristics when they are deciding whether to purchase a commodity.’[75] Under this approach, products are treated as a medium through which consumers obtain certain benefits. Thus, from an analytical perspective, the nature of a particular product becomes less important than the capacity of that product to satisfy certain consumer desires.[76]

The approach is useful in that it draws analysis away from the ‘tempting, but misleading’ functional equivalents approach whereby functionally complementary goods are clustered together.[77] Similarly, it draws analysis away from focusing on consumer preference in terms of convenience as, Albrechtsen suggests, the ACCC has done in the merger banking cases.[78] However, it is submitted that the characteristics test, while it may be a useful indicator of potential cluster markets, fails to comprehensively analyse substitution possibilities because it focuses solely on demand-side substitution. It does not explain, for example, why it may be inappropriate to place metal shoehorns and plastic shoehorns in the same market because, although they may be substitutable from a consumer’s perspective, they may not be substitutable from a manufacturer’s perspective. The High Court made it clear in the Queensland Wire Industries Case that ‘[t]he basic test involves the ascertainment of the cross-elasticities of both supply and demand’.[79]

The most useful and thorough exposition of the application of cluster market analysis is that of Shiff, Ergas and Landrigan. The importance of their work to the development of a coherent and consistent approach to cluster markets is evidenced by the fact that it is the only authority cited by the ACCC in its brief consideration of cluster markets in its information paper, Anti-Competitive Conduct in Telecommunications Markets.[80] Shiff, Ergas and Landrigan, like Ayres, focus on economies of scope in demand[81] which, they assert, are poorly understood in relation to economies of scope in supply. Ayres expresses the relationship of the two concepts best when he says ‘[t]ransactional complementarity is thus a demand-side analog to economies of scope, in which the economies accrue to the consumer instead of the producer.’[82]

Shiff, Ergas and Landrigan correlate the existence of a cluster market with high unbundling costs. They note, however, that ‘while unbundling costs are a necessary condition for a demandside cluster, they are not sufficient to ensure that such a cluster exists.’[83] The proper test, therefore, should not focus on switching costs which, in any event, will often be difficult to ascertain. Instead, it should consider whether, if the price of one product is held constant relative to competitors’ prices, and all other determinants of demand are also held constant, a reduction in the supplier’s relative price of another product would result in an increase in the supplier’s share of sales in the first product. If so, Shiff, Ergas and Landrigan assert, the two non-substitutable products should be considered to be in the same market.[84] Those products are properly treated as a cluster because they meet ‘the “ideal collusive group” test that underpins modern approaches to market definition.’[85] It is no surprise that Shiff, Ergas and Landrigan’s significant contribution to the debate over cluster market definition comes in the context of a consideration of the telecommunications industry.

IV CLUSTER MARKETS IN THE TELECOMMUNICATIONS INDUSTRY

Cluster market analysis has particular relevance to telecommunications industries for two separate but connected reasons. First, there is high customer demand and provider preference for telecommunications products and services to be sold in bundles. It will often be more commercially viable, therefore, for service providers to supply groups of non-substitutable services as a package. Second, technological convergence means that the boundaries between individual services are increasingly unclear. As delivery platforms converge, so too will markets of non-substitutable products. As has been argued, neither consumer preference nor functional complementarity are sufficient reasons for treating non-substitutes as occupying the same market. However, the effects of bundling and convergence are such that it will be increasingly appropriate to treat telecommunications services as forming a cluster and, accordingly, the courts and the ACCC must articulate and adopt a consistent rationale for doing so.

A Bundling

Bundling refers to the practice of selling a group of services together as a package rather than allowing consumers to purchase them separately. From a competition perspective, bundling is a complex issue because, while bundled delivery of services may be efficient, it will often be anti-competitive. The ACCC recognises the contradictory nature of bundling and has stated that the ‘tying’ of one service, in a market where the provider has substantial market power, to another service may be anti-competitive.[86] Its position essentially represents a policy choice to favour, in Lindsay and Williams’ words, dynamic over static efficiency[87] — that is, a form of competition that promotes innovation over a form that promotes cost-minimising. In considering competitive effects in relation to a bundle of services, there will, therefore, be a threshold question of whether the relevant bundle is, in itself, anti-competitive. This will, of course, involve identifying the relevant market for the services comprised in the bundle. Assuming that unbundling is not required, the question will then arise as to whether the bundle of services should be identified as a market comprised of non-substitutes.

Bundling of services is widely prevalent within the Australian telecommunications industry. Clark-Dickson asserts that service providers are attracted to bundling because it allows them to maximise revenue from existing customers, it increases customer dependence on them, thereby reducing churn, and it is attractive to potential customers.[88] Providers may be able to take advantage of those benefits to offer cheaper prices to consumers. From the consumer’s perspective, in addition to price, ‘[t]he most important elements of the service bundle ... is [sic] a single bill or a service level agreement’.[89] In the context of considering the ‘long term interests of end-users test’,[90] the ACCC has acknowledged the importance to consumers of ‘this requirement for integrated service delivery.’[91] Clark-Dickson observes that ‘[m]ost individual subscribers have bundles from a carrier or service provider’[92] and a recent survey reveals that 86 per cent of Australia’s top 100 companies have considered bundling services with one telecommunications supplier, and 22 per cent do bundle services, using one provider for all their telecommunications needs.[93]

It is submitted that the nature of bundled services is such that a bundle of services is more likely to be treated as forming a cluster market than a series of unbundled telecommunications services supplied to a consumer. This is the case because the transactional costs for both suppliers and consumers of, respectively, supplying and receiving the services indicate that they are more efficiently supplied as a bundle. Accordingly, a firm selling those services separately could not compete with one selling them together. The prevalence of bundled telecommunications services suggests that there will be a high incidence of situations in which cluster market analysis of competitive effects will be appropriate.

B Convergence

The term ‘convergence’ describes, broadly, the process whereby technological developments are causing media markets, platforms, corporate structures and regulatory policies to become increasingly integrated.[94] In Blackman’s words:

[N]ow digital technologies permit the manipulation of all forms of information — voice, data and video — across all types of network. Consequently, the distinctions between different types of networks is disappearing — a pipe is pipe [sic] is a pipe.[95]

The disappearance of those distinctions has enormous consequences for market definition. The likely effect will be that the distinction between markets will become increasingly difficult to ascertain.

Ultimately, one can imagine that the process of convergence may lead to full substitutability of all information delivery services. The Productivity Commission has identified the potential for this to occur and has observed that ‘in a hypothetical fully converged media system, all media services would be in competition.’[96] In this scenario, the outer boundaries of any relevant market for telecommunications services may become so broad as to be obsolete in the context of Australia’s present regulatory regime.[97] A more realistic scenario, at least in the short term, is that, in the ACCC’s words, ‘the increased substitutability of various carriage services will correspondingly weaken the boundaries between markets for those products.’[98]

To the extent that convergence increases compatibility of delivery and assimilation of non-substitutes, defining markets as comprising clusters of non-substitutable products will become increasingly appropriate. Corones observes that the consequence of convergence will be that ‘[m]arket boundaries may blur, and separate “product markets” may be difficult, if not impossible, to define.’[99] Where separate markets are difficult to identify because technological change has rendered possible the delivery of those services more efficiently in combination with other services, and there is, therefore, a consumer preference to purchase those services as a group, it may be more appropriate to treat them as occupying a cluster market.

V CONCLUSION

The nature of the provision and consumption of telecommunication products is such that it will often be appropriate, when considering the effect of conduct on competition under the Act, to define the relevant market as comprising a cluster of non-substitutable products. The prevalence of bundled telecommunication services is indicative of the existence of economies of scale under which it may be more efficient to supply tied, rather than separate, services. Convergence of media technologies suggests that it will become increasingly difficult to identify separate product markets, and that in some circumstances it may be more appropriate to define a group of products as forming a cluster market.

Despite the relevance of cluster markets to the analysis of competitive effects within telecommunications markets, the circumstances in which it will be appropriate to identify the existence of cluster markets has never been adequately articulated by Australian courts or the ACCC. This failure calls into question the capacity of the ACCC to properly exercise its significant powers under Part XIB of the Act.


[*] BA (Hons) (ANU), LLB (Hons) (Monash); Barrister and Solicitor of the Supreme Court of Victoria, practising in the Technology, Entertainment and Communications Group in the Melbourne office of Holding Redlich, Lawyers and Consultants. The views expressed in this article are those of the author and do not necessarily reflect the views of Holding Redlich.

[1] Although the telecommunications-specific regime in pt XIB of the Act applies in addition to the general regime in pt IV (see the first dot point of the simplified outline in s 151AA), for the sake of simplicity, this article will refer only to pt XIB of the Act.

[2] Ian Ayres, ‘Rationalizing Antitrust Cluster Markets’ (1985) 95 Yale Law Journal 109, 110.

[3] Defined in ss 7 and 87, respectively, of the Telecommunications Act 1997 (Cth).

[4] S G Corones, Competition Law in Australia (2nd ed, 1999) 86.

[5] [1989] HCA 6; (1989) 167 CLR 177 (‘Queensland Wire Industries Case’).

[6] Ibid 187.

[7] ACCC, Anti-Competitive Conduct in Telecommunications Markets (1999) 28.

[8] Section 151BC of the Act.

[9] See especially s 152AB(2) of the Act.

[10] See especially div 2 of pt XIB of the Act.

[11] Section 151BK of the Act.

[12] Sections 151AKA and 151AL of the Act. The competition rule, which is contained in ss 151AJ and 151AK of the Act, will be breached where a carrier or carriage service provider either has a substantial degree of power within a telecommunications market and takes advantage of that power with the effect, or likely effect, of substantially lessening competition in that or any other telecommunications market, or engages in certain proscribed conduct in relation to a telecommunications market.

[13] Section 151AF of the Act.

[14] Section 4E of the Act.

[15] Maureen Brunt, ‘“Market Definition” Issues in Australian and New Zealand Trade Practices Litigation’ (1990) 18 Australian Business Law Review 86, 93, alluding to the question posed by the Tribunal in Re Queensland Co-operative Milling Association Ltd (1976) 25 FLR 169 (‘Re QCMA’).

[16] ACCC, above n 7, 29.

[17] Ibid 29–31.

[18] (1976) 25 FLR 169, 190.

[19] (1977) 28 FLR 385, 394–5.

[20] [1978] FCA 21; (1978) 32 FLR 305, 311–12 (Northrop J).

[21] [1989] HCA 6; (1989) 167 CLR 177.

[22] Brunt, above n 15, 91.

[23] Corones, above n 4, 116.

[24] Brunt, above n 15, 106.

[25] [1963] USSC 166; 374 US 321 (1963) (‘Philadelphia National Bank’).

[26] Ibid 356–7.

[27] [1966] USSC 122; 384 US 563 (1966) (‘Grinnell’).

[28] Ibid 572.

[29] United States v Rockford Memorial Corporation, 898 F 2d 1278 (7th Cir, 1990); cert denied 498 US 920 (1990).

[30] Re Brunswick Corporation, 94 FTC 1174, 1259 (1979).

[31] United States v Von’s Grocery Corporation[1966] USSC 105; , 384 US 270 (1966).

[32] (Unreported, United States District Court, District of Columbia, Jackson DJ, 28 February 1997) (‘US West’).

[33] Deena Shiff, Henry Ergas and Mitchell Landrigan, ‘Telecommunications Issues in Market Definition’ (1998) 6 Competition and Consumer Law Journal 32, 38.

[34] US West (Unreported, United States District Court, District of Columbia, Jackson DJ, 28 February 1997) [9].

[35] Ibid.

[36] Ibid.

[37] Shiff, Ergas and Landrigan, above n 33, 39.

[38] Ibid.

[39] Ayres, above n 2, 113.

[40] Ibid 112–13 (citations omitted).

[41] Ibid 110.

[42] Ibid.

[43] See, eg, Manufacturing Research Corporation v Greenlee Tool Corporation, [1982] USCA11 986; 693 F 2d 1037, 1043 (11th Cir, 1982).

[44] For an excellent comparison of Australian and New Zealand trade practices legislation, see Brunt, above n 15, 89–93.

[45] (Unreported, New Zealand High Court, McGechan J and Professor R G J Lattmore, 2 June 1995) (‘Port Nelson’), upheld on appeal to the Court of Appeal in Port Nelson v Commerce Commission [1996] NZCA 230; [1996] 3 NZLR 554.

[46] (Unreported, New Zealand High Court, McGechan J and Professor R G J Lattmore, 2 June 1995) 168.

[47] Shiff, Ergas and Landrigan, above n 33, 41.

[48] (Unreported, New Zealand High Court, McGechan J and Professor R G J Lattmore, 2 June 1995) 168.

[49] Ibid.

[50] [1966] USSC 122; 384 US 563 (1966).

[51] Ayres, above n 2, 110.

[52] [1988] NZCA 179; [1988] 2 NZLR 352.

[53] Ibid 360.

[54] See generally Corones, above n 4, 109–11.

[55] TruTone Ltd v Festival Records Retail Marketing Ltd [1988] NZCA 179; [1988] 2 NZLR 352, 360.

[56] ACCC, above n 7, 33.

[57] Geoffrey Walker, ‘Product Market Definition in Competition Law’ (1980) 11 Federal Law Review 386, 398–9.

[58] [1963] USSC 166; 374 US 321 (1963).

[59] Trade Practices Commission, No Opposition to Westpac Acquisition of Challenge: TPC, Press Release (21 September 1995) 2.

[60] Janet Albrechtsen, ‘Searching for a Merger Policy for the Australian Banking Industry’ (1997) 5 Trade Practices Law Journal 180, 197, fn 115.

[61] Ibid 197.

[62] Ibid.

[63] Re Review of Declaration of Freight Handling Services at Sydney International Airport (2000) 22 ATPR 41-754.

[64] Ibid 40,772.

[65] See generally Linda Evans, ‘Access under the Trade Practices Act(2000) 8 Competition and Consumer Law Journal 45, 49–52.

[66] ACCC, above n 7, 33 (citations omitted).

[67] Walker, above n 57, 409–10.

[68] Ibid 410.

[69] Ibid 411.

[70] (1977) 32 FLR 65 (small, medium, large and luxury sedans, station wagons, light commercial vehicles and heavy trucks).

[71] (1979) 37 FLR 244 (all artificial fertilisers).

[72] Walker, above n 57, 410.

[73] Neville Norman and Philip Williams, ‘The Analysis of Market and Competition under the Trade Practices Act: Towards the Resolution of Some Hitherto Unresolved Issues’ (1983) 11 Australian Business Law Review 396, 402.

[74] Ibid.

[75] Ibid (emphasis in original).

[76] See, eg, Robert Dorfman and Neville Norman, Prices and Markets in Australia: Theory and Applications (1980) 124–8.

[77] See generally Shiff, Ergas and Landrigan, above n 33, 35.

[78] Albrechtsen, above n 60, 197.

[79] [1989] HCA 6; (1988) 167 CLR 177, 199 (Dawson J); see also at 188–9 (Mason CJ and Wilson J), 210 (Toohey J).

[80] ACCC, above n 7, 33, fn 80.

[81] Which Ayres, above n 2, 110 refers to as ‘transactional complementarity’.

[82] Ibid 116.

[83] Shiff, Ergas and Landrigan, above n 33, 36.

[84] Ibid 38.

[85] Ibid 35.

[86] ACCC, above n 7, 48.

[87] David Lindsay and Philip Williams, ‘The Trade-Off between Competition and Efficiency in Telecommunications: The Australian Experience’ in Megan Richardson and Philip Williams (eds), The Law and the Market (1995) 100, 105–15.

[88] Pamela Clark-Dickson, ‘Service Bundling: Wrapping up the Customer’, Australian Communications (Sydney) March 2000, 61.

[89] Ibid.

[90] Which is relevant to the application of pt XIC of the Act.

[91] ACCC, Telecommunication Services — Declaration Provisions — A Guide to the Declaration Provision of Part XIC of the Trade Practices Act (1999) attachment 2, 2.

[92] Clark-Dickson, above n 88, 61.

[93] Ibid 62, citing Deloitte Touche Tohmatsu, Deloitte Telecommunications Survey: Consumer Impact of Deregulation (1999).

[94] For a discussion of the meaning of the term ‘convergence’, see Productivity Commission, Broadcasting Draft Report (1999) 49–58; Tim Dwyer, ‘“Convergence”: Reforms for New Media Technologies or, Just Another Plug-In?’ (1998) 17(2) Communications Law Bulletin 21.

[95] Colin Blackman, ‘Convergence between Telecommunications and Other Media: How Should Regulation Adapt?’ (1998) 22 Telecommunications Policy 163, 164.

[96] Productivity Commission, above n 94, 54.

[97] For a discussion of the consequences of convergence on competition regulation, see Rod Shogren, ‘Convergence of General Competition Law with Telecommunications Specific Regulation — The Australian Experience’ (1988) 1 Telemedia 153.

[98] ACCC, above n 7, 33.

[99] Corones, above n 4, 408.


AustLII: Copyright Policy | Disclaimers | Privacy Policy | Feedback
URL: http://www.austlii.edu.au/au/journals/MelbULawRw/2001/16.html