Review of the National Access Regime Submission in response to the Productivity Commission’s position paper National Competition Council July 2001 Contents 1. 2. Introduction and overview ................................................... 4 Threshold matters ............................................................. 7 2.1 The objective of Part IIIA ................................................. 7 2.2 Pricing principles ............................................................. 9 2.3 Services ......................................................................... 9 The importance of service delineation .................................. 10 The current exclusions ...................................................... 13 2.4 Competition and markets ............................................... 13 2.5 Provider ...................................................................... 14 Meaning of ‘provider’......................................................... 14 Vertically integrated and separated providers ....................... 15 2.6 ‘Exemptions’ from Part IIIA ............................................. 15 Access holidays ................................................................ 15 Binding rulings ................................................................. 17 Framework undertaking..................................................... 19 2.7 Overlap with other parts of TPA ....................................... 19 3. Declaration and arbitration................................................. 21 3.1 Proposed amendments to the declaration criteria ............... 21 The ‘Tier 1’ proposals ........................................................ 22 The ‘Tier 2’ proposals ........................................................ 28 3.2 Prices monitoring as an alternative to declaration ........... 34 3.3 Negotiation and arbitration............................................. 36 Information provision ........................................................ 37 Arbitration to require extension/expansion ........................... 38 4. Certification and undertakings ............................................ 40 2 4.1 Specific proposals for certification .................................... 42 Including the certification principles in Part IIIA .................... 42 Modifications to the clause 6 principles ................................ 43 Price Cap arrangements .................................................... 43 4.2 .................................................................................... 44 Specific proposals for voluntary undertakings .......................... 44 Undertakings by ‘non-providers’ ......................................... 44 5. Administrative and procedural matters ................................ 47 5.1 The role of Ministers....................................................... 47 5.2 Institutional arrangements ............................................. 48 6. Reviews and Appeals........................................................ 54 References ............................................................................ 57 3 1. Introduction and overview The National Competition Council’s (the Council’s) first submission to this inquiry contended that the national access regime, though still new, has fostered substantial gains in utility reform. The Council argued that, whilst there is clearly scope to address flaws in the framework, the structural underpinnings are sound. The scope of declaration has been confined within a narrow band of natural monopoly infrastructure services and the negotiate/arbitrate framework is one of the least interventionist to be found in any of the various access regimes introduced in recent years. Fine tuning of Part IIIA is desirable The Council broadly supports many of the Productivity Commission’s specific proposals to complement and improve the existing regime. Among these are the proposals to: • • • introduce an efficiency-based objects clause for Part IIIA of the Trade Practices Act 1974 (TPA)(Proposal 5.1); include pricing principles within the Part IIIA framework (5.3); streamline the negotiation/arbitration framework for declared services, including measures to: streamline information provision to access seekers (6.3), ensure that arbitration outcomes are consistent with the objectives of Part IIIA (6.5), promote the primacy of negotiated outcomes (6.6); and improve transparency of arbitration processes (6.7 and 9.9); require that Commonwealth access regimes be assessed for effectiveness against the clause 6 principles (variation on proposal 7.1); and that the same test of effectiveness be applied irrespective of public or private sector ownership (7.2); modify the principles for certification of State and Territory access regimes to make some threshold matters more explicit (7.4); streamline the access undertakings framework by allowing an access provider to lodge an undertaking for a declared service (7.5); by aligning the criteria for accepting an undertaking with those applying to arbitration of declared services and to certification (7.6); and by allowing full merits review on undertakings determinations (9.4); and introduce an option of prices monitoring as an alternative to declaration (canvassed by the Commission). • • • • The Council considers that these changes would address certain limitations in the detail of Part IIIA. The effect would be to improve the operation of the 4 access framework without imposing a wholesale restructuring of that framework. ... but wholesale change poses serious risks The Council has serious reservations about the Commission’s view that the structural framework of Part IIIA is deficient. While measures to strengthen the framework are desirable, overturning it in favour of something new seems an extravagant response to the concerns raised in the Position Paper. Indeed, the Paper lacks evidence of fundamental deficiencies in the architecture of Part IIIA. While some flaws are apparent, the evidence shows these to be implementation issues (that can be addressed through the type of proposals noted above) rather than structural flaws. The Council is especially concerned by the proposals to rewrite the declaration criteria (6.1 and 6.2). These proposals appear to stem from the Commission’s concerns that the current criteria have an inappropriate focus on ‘competition’ rather than ‘efficiency,’ making the ambit of Part IIIA too wide. But experience to date shows that the current criteria have confined access regulation to a narrow range of services provided by bottleneck facilities. Further, the current wording is now subject to significant authority through the Australian Competition Tribunal’s (the Tribunal) decisions in the Sydney Airports case1 and the Eastern Gas Pipeline (Duke) case2. The Tribunal’s interpretation seems to be consistent with the objectives underpinning the Commission’s proposed changes, raising questions as to what purpose would be served by change. Rewriting the criteria would set aside the evolving body of legal precedent that is now promoting increased levels of understanding among stakeholders as to the proper interpretation of s.44G(2). The Council is also concerned that the proposed new declaration criteria are, in part, a response to perceived issues that are, in fact, illusory. In addition, some of the proposed criteria would create tensions with legal interpretation, and may, in some instances, bring perverse results – including a shift away from ‘natural monopoly’ as the threshold test in the ‘uneconomic to develop’ criterion. For these reasons, introducing wholesale change to the wording of the declaration criteria is likely to impose considerable costs with no guarantee that the Commission’s objectives will be achieved any better than, or perhaps even as well as, the current wording. 1 Sydney International Airport; Re Review of Declaration of Freight Handling Facilities (2000) ATPR 41-754. Duke Eastern Gas Pipeline Pty Ltd (2001) ACompT 2. 2 5 Another significant concern is the proposal to combine Part IIIA policy responsibilities (such as coverage and assessment of the design of access legislation) and the regulation of specific infrastructure within a single body. The Commission’s view in this regard appears inconsistent with its acknowledgement of the importance of institutional separation in its Review of the Prices Surveillance Act 1983. The Council regards access legislation design and coverage issues to be inherently quite distinct from regulation functions under Part IIIA. It is important that the regulatory framework recognises this fact. To institutionally merge these functions would oblige a single body to set the rules for the regulatory framework and then implement its own rules. This could compromise the independence of the regulatory framework, and ultimately, weaken market perceptions of the independence of Part IIIA processes. The Council notes that every coverage matter it has considered has been vigorously contested. As such, it is important to have a decision making body that is independent and perceived to be independent. Should coverage decisions be placed in the hands of the regulator/arbitrator, there is a risk that some participants may feel constrained in their ability to participate, for fear of alienating the potential future arbiter. There is also a danger that the regulator/arbitrator may be perceived as having an inherent bias in favour of coverage. Finally, the Council is concerned that the proposed merger of responsibilities would blur the distinction between the certification and undertakings processes. The certification pathway – which has made an important contribution to good design in access legislation – could lose its viability over time. Structure of the submission This submission comments on key aspects of findings and recommendations in the Commission’s Position Paper, focusing principally on areas of concern to the Council. The structure of the submission broadly follows that of the Position Paper, and covers : • • • • • some threshold considerations, including the objectives of Part IIIA; proposals on declaration and arbitration; proposals on certification and undertakings; administrative and procedural matters; and reviews and appeals. 6 2. Threshold matters In this section, the Council provides comments on recommendations in the Position Paper that relate to the broad framework of Part IIIA. These include proposals for the introduction of: • • an explicit objects clause for Part IIIA; and generic pricing principles. A number of definitional issues – with implications for specific Commission proposals – also warrant close attention. These include the definitions of ‘service’ and ‘provider’ in Part IIIA; and the role of market analysis in the interpretation and application of the criteria. 2.1 The objective of Part IIIA In its first submission, the Council noted that, while Part IIIA contains no explicit objects clause, the TPA contains a general objects clause at section 2, which identifies the object as being to: enhance the welfare of Australians through the promotion of competition and fair trading and provision for consumer welfare. The Council considers that this objects clause is consistent with the efficiency objective for Part IIIA proposed by the Commission. However, section 2 of the TPA is, as the Commission has noted, somewhat ambiguous in the role and priority to be accorded to the various concepts identified and there is no explicit indication as to how the section is to be taken into account in interpreting specific provisions. The Council agrees that the specific content of Part IIIA would benefit from a more explicit objects clause. The Commission has proposed (proposal 5.1) that a two-part objects clause be included in Part IIIA. The first part of this clause would set out an efficiency objective: (a) enhance overall economic efficiency by promoting efficient use of, and investment in, essential infrastructure services; Proposal 5.1(a) reflects two aspects of the end goal of enhancing overall economic efficiency: • • promoting the efficient use of essential infrastructure services; and promoting efficient investment in essential infrastructure services. 7 The Council notes that the notion of promoting efficient investment has two sub-elements: • • encouraging efficient investment; and reducing incentives for inefficient investment in natural monopoly infrastructure by regulating the shared use of such infrastructure, where appropriate. The Council agrees that there is currently some uncertainty as to the specific objectives of Part IIIA. For example, in the Duke decision3, the Tribunal noted that if it was an objective of the National Third Party Access Code for Natural Gas Pipeline Systems (the Gas Code) to deter inefficient duplication, the Gas Code lacks a mechanism for achieving that objective. However, it is arguable that the Gas Code (as for Part IIIA) does achieve this objective – albeit indirectly – by establishing a mechanism for access to existing infrastructure that enables a person to pursue private goals without having to duplicate infrastructure. The Council believes that including the proposed objects clause would make explicit what has been implicit, and that the clarity this engenders would improve the efficacy of Part IIIA, as well as access regimes under the Part IIIA umbrella such as the Gas Code. The Council agrees with the submission made by Australian Council for Infrastructure Development (AusCID)4 that the amendments should require that (at least this part of) the objects clause be taken into account in the interpretation and application of all Part IIIA provisions, including provisions relating to declaration, arbitration, undertakings and effective access regimes. The second part of the proposed objects clause focuses on industry specific regimes. It reads: (b)provide a framework and guiding principles for industry-specific access regimes. This statement is more appropriately seen as a description of Part IIIA, rather than as an objective (and thus, more suitable for inclusion in the Second Reading Speech for any amending legislation). One of the functions of 3 The Gas Code contains a coverage mechanism that is substantially the same as the declaration process and criteria in Part IIIA. A decision by the Australian Competition Tribunal (the Tribunal) on coverage matters under the Gas Code has relevance to the interpretation of Part IIIA. AusCID 2001, p.6. 4 8 Part IIIA is indeed to provide a framework and guiding principles for industry specific regimes through the certification and undertaking processes. The most effective way of ensuring that all industry specific regimes are disciplined by Part IIIA is to bring them within the jurisdiction of Part IIIA. 2.2 Pricing principles The Council supports the introduction of pricing principles to provide guidance in arbitrations, in the assessment of undertakings and in the certification of access regimes. An issue the Council has encountered in the certification processes is the generality of the principles contained in clause 6 of the Competition Principles Agreement (the CPA). Greater certainty with respect to the pricing principles would reduce the administrative burden of access regulation. The Council therefore supports the introduction of pricing principles within: • • the clause 6 framework – for certification purposes; and Part IIIA – for undertakings and arbitrated access conditions. The Council agrees that the principles need to be general and high level, to provide guidance on the appropriate pricing methodologies that should be available to regulators. This is consistent with it’s the Council’s view of Part IIIA as an overarching framework for access regulation. While keeping the principles general and high level inevitably reduces the level of guidance provided, the Council believes it would not be appropriate to prescribe particular methodologies. The Council considers that existing regimes and arrangements already approved under Part IIIA are consistent with the pricing principles outlined in proposal 8.1. The Council’s views on the Tier 2 proposal on price caps are considered in section 4.1 of this submission. 2.3 Services The definition of ‘service’ has relevance to a number of the Commission’s specific proposals; notably, proposals to amend the Part IIIA declaration criteria. The definition of ‘service’ has been one of the most contentious aspects in the application of Part IIIA. In some respects this is inevitable given that proper identification of a service is the very trigger for the application of Part IIIA. There have been cases (Duke, for example), where the definition of the service 9 is at issue. In others, the issue has been whether a particular service falls within the ambit of Part IIIA.5 This section comments on two aspects of the Part IIIA notion of services: • the importance of ensuring that the relevant services are described and delineated accurately; and the role and significance of the current exclusions from the definition of ‘service’. • The importance of service delineation Service is defined in section 44B of the TPA to mean: "'Service' means a service provided by means of a facility and includes: (a) the use of an infrastructure facility such as a road or railway line; handling or transporting things such as goods or people; a communication service or similar service; (b) (c) but does not include: (d) (e) (f) the supply of goods; or the use of intellectual property; or the use of a production process; except to the extent that it is an integral but subsidiary part of the service". 5 See Hamersley Iron Pty Limited v National Competition Council & ors (1999) ATPR 41-705. The correct approach to the definition of service and the application of the production process exemption also arise in a case commenced by Western Power Corporation in the Federal Court of Australia. The case challenges the Council's jurisdiction to consider an access application made by companies within the Normandy group. This issue is considered elsewhere in this section. 10 Clearly, the definition of service is fundamentally conditioned by the facility that provides the service. The notion of service cannot be divorced from the characteristics and functionality of the facility that provides the service. Whatever the context of market analysis, the first question is the identification of the good or service which is bought and sold or for which there are potential transactions. It is only once the relevant good or service, i.e. the relevant product, has been identified that one can move, if required in the particular context, to consider what other products are substitutable for that product. In Duke, for example, a central issue was whether the services provided by the Eastern Gas Pipeline (EGP) were to be described by reference to the start and end points of the gas transportation service or by reference to the markets served by that transportation service. In analysing this issue in the context of the Gas Code, the question about service identification is: what is the pipeline owner selling and a gas trader or gas user purchasing from that pipeline owner (or what service could be bought and sold)? That defines the relevant service. Such an approach is consistent both with the correct statutory interpretation and with economic analysis. In other words, the correct approach is to define the product that is sold (that is, the service) and then test for substitutes for that product to define market boundaries for that service. An approach that seeks to delineate a market and then define the relevant product by reference to that market runs counter to logic. The starting point must be to define the thing that is traded – only then is it possible to test for substitutes. The idea of introducing market analysis into the very delineation of a service risks choosing the wrong market as a starting point. This may involve an inappropriate assumption about relevant substitutes and/or confuse the distinction between the market in which the service is provided and the relevant downstream market. Accordingly, the task of service delineation does not incorporate any notion of market analysis.6 Service delineation is precisely equivalent to the product identification that occurs in standard competition analysis. What is it that is being offered or is capable of being offered by the infrastructure in question? Using this approach, the Tribunal in the Sydney Airport decision identified the relevant service as: "The service provided by the use of aprons and hardstands at SIA (Sydney International Airport) to load and unload international 6 Once the service is delineated, criterion (b) does, however, require a separate consideration of substitutes, and at least implicitly of the market, for that service. 11 aircraft at SIA and the service provided by the use of an area at that Airport to store equipment and to transfer freight from the loading and unloading equipment to and from trucks7" That was the service in the sense that it was the product made available by the airport. Similarly, in Railway Access Corporation v. NSW Minerals Council Limited8 the Full Federal Court described the notion of a service in Part IIIA in the following way: The definition of 'service' in section 44B of the Act makes clear that a service is something separate and distinct from a facility. It may, however, consist merely of the use of a facility. The definition of 'service' distinguishes between the use of an infrastructure facility, such as a road or railway line, and the handling or transporting of things, such as goods or people, by the use of a road or railway line. The fact that one service provider, such as Freight Rail Corporation, is using the railway line infrastructure facility made available to it by Rail Access Corporation for the purposes of carrying coal by rail does not mean Rail Access Corporation is carrying on, or is the provider of, a service of carrying coal by rail. Service delineation should not be confused with the subsequent testing for substitutes for that service – which is necessary to determine the market for the service. As the Tribunal said in the Duke decision: The question of what constitutes the services provided by the pipeline is fundamentally a mixed question of fact and the proper construction of criterion (b), rather than a matter of economic analysis. Every haulage service will of necessity be from one point to another. That is the commercial service actually provided by the pipeline operator to its customers. That service may be of different use to the producers in the origin market or the customers in the destination market, but it is the same service. No market analysis is necessary or appropriate in the description of the services provided by the pipeline (Duke decision, paragraph 69). Under Part IIIA and the Gas Code, it is the service that is subject to coverage or declaration that then becomes the service that is the subject of an access arrangement or the determination of terms and conditions through negotiation/arbitration. Generally, it does not make sense to negotiate or arbitrate terms and conditions of access in relation to a ‘delivery to a particular market’ transportation service. Transportation services inherently involve the movement of things or people from one point to another. 7 8 (2000) ATPR 41-754. (1998) 87 FCR 517 12 Therefore, two gas haulage services delivered into a common market from quite separate gas basins are not one and the same service. The current exclusions In its first submission, the Council questioned whether the exclusions from the definition of service were necessary. The Council noted that it seemed unlikely that, properly construed, the subjects of the exclusions would be covered under the current Part IIIA criteria. The Council considered that the only decision on the ‘production process’ exclusion was flawed as it had failed to distinguish between a production process and an input into that production process. However, this did not alter the Council’s views that the exclusions, appropriately interpreted, have limited impact on coverage. The issue of the proper interpretation of ‘production process’ has again been raised in the context of an application by companies in the Normandy group for declaration of services provided by the electricity transmission and distribution networks owned and operated by Western Power. Western Power is seeking a determination from the Federal Court that the service that is subject of the application is not a service under Part IIIA on grounds which include that it is an integral part of Western Power’s production process. The outcome of these proceedings may also have an impact on the National Electricity Code, as the industry code undertakings rely on the definition of service under Part IIIA. Increasingly, it seems that service providers are raising the scope of the exclusions as an argument against consideration of a declaration application. As this involves a threshold jurisdictional issue, it could lead to a considerable lengthening of the time taken to finalise declaration applications. Service providers can bring proceedings in the Federal Court on the definition of service, including appeals on the jurisdictional issues to the Full Federal Court. In addition, it is open to the applicant or service provider to apply for review of a decision on declaration if the service is found to be within Part IIIA. Conversely, removal of the exclusions would allow consideration of whether a service should be declared to be determined in the context of the s.44G(2) criteria. 2.4 Competition and markets In Chapter 6 of its Position Paper, the Commission indicated a preference for removing the notion of market from criterion (a) and substituting it with a notion of ‘activity’: 13 Although the Commission has a preference for the wording proposed by the Hilmer Committee (1993, p.266) – that is, ‘in a downstream or upstream activity’ – it does not see a compelling need to institute such a change for the sake of simplification (PC 2001a, p.131). The Council would be concerned about the implications of such a change. The very notion of competition imports with it the notion of a market. Competition occurs within markets. Whilst market definition should not be used as an end in itself but as a tool that is used to determine the question at hand9, it is nonetheless an important part of the analytical process in assessing competitive forces. If the notion of activity were substituted for one of market, there is significant risk that declaration could potentially apply to the mere resupply of a monopoly service. The Council would consider that inappropriate and it does not seem consistent with the Commission's underlying objectives. 2.5 Provider Meaning of ‘provider’ Section 44B of the TPA defines ‘provider’ as the owner or operator of the facility that is used (or is to be used) to provide the service. The Council considers that the intention of Part IIIA is that the provider is the entity who has the legal right to provide access or who can be required under Part IIIA to provide access to the service. Generally at law it is not possible for a person to assign an interest greater than the one they possess. Consistent with this, under Part IIIA it would not be possible for an entity to provide access to a facility to an extent greater than their legal rights to use that facility. Therefore the provider must be able to negotiate an access contract subsequent to declaration, or if negotiation fails, be bound by an arbitration determination. Similarly for undertakings, the provider must be the entity who has legal power to fulfil the terms of the access undertaking. 9 Broken Hill Proprietary Company Ltd v. Queensland Wire Industries Pty Ltd (1989) 167 CLR 177; see also Norman & Williams, "The analysis of market and competition under the Trade Practices Act" (1983) 11 ABLR 396 at 400 14 Vertically integrated and separated providers The Commission has proposed that there should be explicit recognition in Part IIIA that the regime covers eligible services provided by both vertically integrated and non-integrated service providers (proposal 5.2). While the Council agrees that Part IIIA should apply to both vertically integrated and non-integrated service providers, the Council does not consider the amendment necessary. The Tribunal in the Sydney Airports decision said: …the provisions in Pt IIIA of the Act are not limited in their application to a vertically integrated organisation, and we cannot see anything in the Hilmer report to suggest that the Hilmer Committee intended its proposed new legal regime would be limited to facilities which are vertically integrated with potentially competitive activities in upstream or downstream markets (Sydney Airports decision, 40756). It is clear that Part IIIA already applies to service providers who are either vertically integrated or non-vertically integrated. 2.6 ‘Exemptions’ from Part IIIA The Council considers it fundamentally important that access regulation not deter or delay efficient investment in infrastructure. As the Commission notes the consequences of under-investment are significant and accordingly, economic efficiency is most likely to be achieved by erring on the side of over, rather than under, compensation of service providers. That said, the balancing process is a difficult one and users and infrastructure owners are likely to have strongly opposing views. The Council considers that these issues are most acute in relation to new investment, whether that be the development of entirely new projects or new investment which expands the range of services that can be provided by existing infrastructure. The current mechanisms may not be sufficient to provide appropriate certainty for infrastructure investors and the Commission has clearly identified the problems arising in this area. This section contains a discussion of some of the options that have been identified in the Position Paper and in submissions to the Commission. Access holidays The Commission has canvassed the idea of access holidays to counter the potentially ‘chilling’ effect of access regulation on new investment. Regulation 15 of greenfields investments is a contentious issue and the principle of access holidays has attracted considerable interest. If access holidays were to be introduced a number of issues would need to be addressed, including the difficulty in identifying relevant investments and the risk of gaming by infrastructure owners. If access holidays were available, their effect would be to quarantine access regulation for at least a specified period of time. The Commission’s preferred approach is a ‘null’ undertaking, which would effectively state that there be no regulated access for the duration of the holiday. It is not clear whether there would be some regulatory discretion to be exercised in determining whether to accept a null undertaking. The Commission has indicated that it does not favour blanket exemptions for all greenfields investments, or access holidays of unlimited duration. In these circumstances, it is not clear to the Council what the trigger would be to activate an access holiday. The Council would be concerned if the determination of whether an access holiday would be available were based on an ex ante assessment of profitability of any particular project. For example, if a project was likely to earn normal returns, it could indicate that market power could not be exercised in a dependant market; in which case, declaration would not be appropriate. Conversely, if high returns are doubtful because a project is not efficient, it is unclear why favoured treatment is warranted. It is important to distinguish whether the issues that are addressed in considering whether to grant an access holiday are questions of coverage or questions of appropriate regulation. If the issues relate to whether the provider of a marginal project would have market power in the downstream market, or whether the cost of regulation of a particular service might be too high and contrary to the public interest, they would appear to go to the criteria for coverage. If this were the case, arguably they would be best dealt with through a binding ruling approach (discussed below). However, if the issues regarding the grant of an access holiday were regulatory in nature, then some form of qualified (perhaps even ‘null’) undertaking would appear to be the right approach. Stephen King raises the case of major infrastructure investments where returns are subject to uncertain demand, with the possibility of blue sky returns as one possible outcome, but a material risk of failure as another (King 2000, pp.12-15). In this case, an application for declaration would be likely if returns turn out to be high. Ex ante, eliminating the possibility of high returns could make the project commercially unviable and deny the community what may have been a socially desirable investment. The Council regards questions about appropriate returns on investment in natural monopoly infrastructure as being regulatory in nature, rather than coverage questions – that is, the binding ruling approach (see below) is not 16 appropriate. Questions about appropriate returns are more appropriately dealt with via the undertaking process. A ‘null’ undertaking process may be desirable to deal with the type of extreme case noted by King. In a number of submissions to the Commission following release of the Position Paper, there has been some discussion about access holidays being invoked where the relevant investment is contestable. This raises questions as to precisely what contestability requires and what it is that must be contestable. With these issues in mind, the Council considers that if the Commission were to endorse a system of access holidays based on an initial threshold of contestability of the project, then there may be some role for prices monitoring during the period of that access holiday. An alternative to access holidays is to enable an independent regulator to factor in the unique risks associated with greenfield investments through the regulatory process. For example, the Australian Competition and Consumer Commission (the ACCC) considered greenfields issues in its decision on the Central West Pipeline. Similarly, the Council took account of greenfields issues in its approach to the NT/SA Rail certification. This option has been canvassed in submissions by Network Economics Consulting Group Pty Ltd (NECG)10, Telstra and AusCid. Their proposals have some merit and may provide a workable solution within the current legislative framework. Binding rulings A number of submissions in response to the Commission’s Position Paper propose the inclusion within Part IIIA of a mechanism for binding advance rulings on the prospects of declaration. The Council considers that such a proposal has merit. There would, however, be difficult issues as to the extent to which the Council – or any other body charged with the task – would be in a position to form an opinion on relevant matters. This would necessarily depend on the circumstances of each application. There is currently a procedure for advance advisory opinions under the Gas Code. The relevant provisions are as follows: 1.22 A Prospective Service Provider may request an opinion from the NCC as to whether a proposed Pipeline would meet the criteria for Coverage in section 1.9. 10 NECG, "Joint Industry Submission on the Productivity Commission's Review of the National Access Regime", June 2001. 17 1.23 The NCC may provide an opinion in response to a request under section 1.22 but the opinion does not bind the NCC in relation to any subsequent application for Coverage of the Pipeline. To date one application for an advance ruling has been made to the Council. In that case the Council’s advice was that, on the basis of the information supplied by the prospective service provider, it was unlikely that the pipeline would become covered. The supplied information included that: • • • the pipeline would be only 4.5 km in length; it was being built to a optimal size for the customer it was to service; and there were other pipelines that may have been able to provide substitute services to potential third party access seekers. Under the Gas Code, the advance ruling is not binding on the Council, and as a result it is more appropriately described as an advisory opinion. The Council’s capacity to give a binding ruling would be affected by the information available to the Council, including information gathered through any public process. It would be appropriate for any binding ruling process to be conducted in a similar way to an application for declaration. It might include a process for the Council to recommend revocation of the binding ruling if there was a material change in circumstance or if the service provider purposively or negligently misled the Council in the information provided. Any such revocation should be subject to a merit review to the Tribunal. One context in which a binding ruling process may have been helpful was in regard to access to the proposed Tarcoola to Darwin rail-track. In that case, competition in downstream markets from road haulage raised a very real question as to whether the project would satisfy declaration criterion (a). However the NT and SA Governments considered that the threat of potential declaration obliged them to establish an access regime and seek its certification. It would not have been appropriate for the Council to refuse certification on the grounds that clause 6(3) was not met when there was no other mechanism for providing the Governments with the certainty they sought. The Council sees the binding ruling process having particular application in situations where: • it is unlikely that the infrastructure will have natural monopoly characteristics and, as a consequence, it is unlikely that criterion (b) will be satisfied; or 18 • the market conditions are such that it is unlikely that criterion (a) will be satisfied, for example, because the infrastructure owner is not ever likely to possess market power. The fundamental advantage of a binding ruling is that it involves consideration of the relevant issues at the time the investment is made. Even if the Council were unable to reach a firm view on one of the criteria, the process and the views reached in relation to the other criteria may nonetheless provide a much greater degree of certainty to an infrastructure owner than would otherwise be available. Given the recent complaints about levels of uncertainty attendant on infrastructure investment, any mechanism that promotes certainty is likely to be efficiency enhancing. Framework undertaking Both the AusCID and Joint Industry NECG submissions propose the introduction into Part IIIA of a mechanism to allow service providers to seek ex ante approval of important elements of potential future regulation. This pre-investment undertaking would establish the parameters for future regulation and would bind both the service provider and the regulator. The Council sees considerable merit in this suggestion and supports the proposal. It is not clear to the Council that an amendment to Part IIIA is necessary to allow for framework undertakings as there appears to be sufficient flexibility in the current criteria to allow the ACCC to accept such undertakings. However, a minor amendment with explicit support in the Second Reading Speech may be required to achieve certainty. 2.7 Overlap with other parts of the TPA Proposal 10.1 recommends that the terms and conditions of certain access arrangement provisions be exempt from the operation of Part IV of the TPA. This is to reduce the risk of regulatory overlap between Part IIIA and IV. The Council considers that there is little risk that the agreements on terms and conditions reached between service providers and access seekers within the Part IIIA framework would be in breach of the provisions of Part IV of the TPA. Part IIIA agreements generally are of a vertical nature, with the most obvious risk being that the service provider agrees not to provide services to another party – in other words, exclusive dealing. However, there is little risk that the service provider would be able to execute any such exclusive dealing provision as Part IIIA limits their ability to refuse to deal, or to deal on inappropriate terms and conditions with a third party. The ACCC was required to consider authorisation of particular arrangements relevant to access regimes in the context of the National Electricity Code and 19 the Market Rules under the Victorian Gas Regime. This was because the structure of those access arrangements required a degree of co-operation between parties who might otherwise have been competitors. The ACCC was able to consider the issues raised by authorisation in conjunction with its consideration of the electricity undertakings and the GPU GasNet access arrangements respectively. The Council notes that the type of conduct considered by the ACCC through these authorisations would not have fallen within the scope of Proposal 10.1, and as such, a Part IV exemption would have had no relevance. Nor is the Council aware of these authorisation processes creating difficulties for the service providers concerned. As such, we see no practical justification for exempting this type of conduct from Part IV. More generally, the Council believes it is not appropriate to exempt conduct that would ordinarily be in breach of Part IV from the operation of Part IV simply because it arises out of or in connection with an access arrangement. Fundamentally the focuses of Part IIIA and Part IV of the TPA are different. Whilst the terms and conditions of an access arrangement are unlikely to involve a breach of Part IV of the TPA, related conduct may do so and should be not be excluded from consideration under Part IV. 20 3. Declaration and arbitration There have now been a number of declaration applications under Part IIIA. In addition, under the Gas Code there have been several applications to revoke coverage and one to activate coverage. Three of these matters have gone to the Australian Competition Tribunal for determination and two others have been the subject of Federal Court applications; one Federal Court application remains pending. These processes have given rise to a substantive body of law as to the way in which the declaration criteria (and the relevant definitions) should be interpreted. In considering amendments to the declaration criteria, it is important to consider the impact such changes would have on this body of interpretation and on levels of certainty for infrastructure owners. There have been no arbitrations under Part IIIA and no undertakings11 have been accepted, although several have been lodged. There is therefore much more limited material available to assist in evaluating how these provisions will operate. The only guidance available is by analogy with the experience with arbitrations under Part XIC. Overall, the Council's experience has been that the declaration criteria have worked well and have not resulted in regulation of other than bottleneck infrastructure with natural monopoly characteristics. 3.1 Proposed amendments to the declaration criteria The Commission expressed concern that the current criteria for declaration, as interpreted, may lead to ‘inappropriate declaration of services’. The Commission identified three particular concerns: • the declaration criteria use competition as a proxy for efficiency, with efficiency relegated to a set of residual considerations; there is scope for declaration where competition gains would be merely trivial; and weaknesses in the natural monopoly criteria could allow declaration of services without substantial and sustainable market power. • • 11 That is other than the undertakings relevant to the implementation of the NEM. The nature of these NEM undertakings are arguably different from other likely undertakings. This point is discussed further in section 4 of this submission. 21 The Commission has proposed some minor (Tier 1) and major (Tier 2) amendments to the declaration criteria to address these perceived deficiencies. The ‘Tier 1’ proposals The Tier 1 proposals envisage changes to two of the current declaration criteria by amending: • the promotion of competition test in criterion (a) to establish, instead, a requirement that access lead to a substantial increase in competition; and the monopoly test in criterion (b) to refer to a second facility rather than another facility. • The proposed new competition test The Commission argues that this amendment would screen out trivial cases, where the costs of access regulation outweigh any benefits. The proposal reflects a concern – based on an interpretation of the Sydney Airports decision – that the current criterion (a) sets too low a hurdle. The Council considers that this concern needs to be revisited in the light of the Tribunal’s decision in the Duke matter. This decision applied the same test of ‘promote competition’ as in the Sydney Airports matter and found that it would result in a non-trivial change in the competitive environment. The Tribunal in the Duke case said: The meaning of this term was discussed by the Tribunal in Sydney International Airport. The Tribunal said (at 40,775) that the notion of “promoting” competition: “involves the idea of creating the conditions or environment for improving competition from what it would be otherwise. That is to say, the opportunities and environment for competition given declaration, will be better than they would without declaration.” The Tribunal concluded that the TPA analogue of criterion (a) is concerned with the removal of barriers to entry which inhibit the opportunity for competition in the relevant downstream market. It is in this sense that the notion of promotion of competition involves a consideration that if the conditions or environment for improving competition are enhanced, then there is a likelihood of increased competition that is not trivial. We agree. (Duke decision para 75) 22 In applying this test, the Tribunal concluded that the EGP would not have market power in the South-east Australian gas sales market and that, as a consequence, regulation under the Gas Code would not promote competition in that market. The proposed change to the competition test also raises another issue. The notion of ‘lead to a substantial increase in competition’ appears to encapsulate an approach to measuring competition that was rejected by the Tribunal in the Sydney Airports decision in favour of a test which focuses on structural impediments to competition. The Council shares the concerns expressed by the Western Australian Government submission in relation to this proposal. A requirement that access would ‘lead to a substantial increase in competition’ seems to require an actual demonstration that increased competition would, in fact, occur rather than a focus on creating the conditions for increased competition. In other words, the use of the word ‘lead’ has the potential to pay insufficient attention to the pro-competitive impact of potential entry. The introduction of the term ‘substantial’ appears to introduce a Part IV type competition test, but the proposal does not mirror Part IV because Part IV couples the notion of a substantial lessening of competition with an assessment of whether conduct ‘has or is likely to’ have the effect of substantially lessening competition. Any introduction of a ‘substantial increase in competition’ requirement should be accompanied by the notion of the likelihood of such an increase. The proposed new natural monopoly test The Commission argues that this proposed change would rule out regulation of services provided by duopolies and oligopolies, which it considers are unlikely to enjoy substantial and sustainable market power. The Council considers that the proposed change seeks to ‘fix’ a problem that, whilst reflected in one submission, has not actually arisen. The change is also in conflict with economic thinking and recent legal interpretation as it effectively shifts the ‘monopoly’ test away from the concept of natural monopoly – which, in fact, is contrary to the Commission’s expressed intention. The Council has identified four issues arising from this proposed change. First, a potential problem recognised by the Commission is that the proposed test may skew the interpretation of criterion (b) away from the current technology neutral approach toward a view that the ‘second facility’ should ‘duplicate’ the first. 23 The second issue, related to the first, is that introducing the notion of a ‘second facility’ may overturn existing authority that the ‘monopoly’ criterion should be used to take account of existing infrastructure providing, or potentially providing, an effective substitute service for the service which is subject to the declaration application. The importance of considering existing infrastructure was confirmed by the Tribunal in the Duke decision: A literal construction of criterion (b) might require the decision-maker, in the application of the criterion, to ignore the existence of pipelines which have already been developed. That is not the approach adopted by NCC in its Final Recommendation. NCC said at 47: "... the Council considers the objectives of the legislative scheme are best met by also having regard to the provision of competing services by another existing pipeline for the purposes of criterion (b). The Council concludes that where an existing pipeline already provides, or could provide with minor modifications or enhancements, services which are competitive with the services of the pipeline the subject of the coverage application, criterion (b) will not be satisfied." …No one contended that the Tribunal ought to adopt a different approach from that adopted by NCC in this respect. The expert economist called by the applicants, Mr Ergas, could think of no economic reasoning that would support a finding that pipelines that are actually duplicated deserve to be covered under the Code by virtue of criterion (b), unless for some reason the duplicated pipelines are not acceptable substitutes for each other. There is no logic in excluding the existing pipelines from consideration in the determination of whether criterion (b) is satisfied. The policy underlying the Code would not be advanced if the Tribunal were to proceed in that blinkered way. We therefore think it appropriate to enquire whether the MSP or the Interconnect provide or could be developed to provide the services provided by means of the EGP. The proper characterisation of those services is itself an issue of construction which is addressed later (Duke decision, paragraphs 5557). The proposed change may overturn this authority because the new words might be interpreted as a move away from the current approach of taking 24 account of all economically relevant existing and potential infrastructure services toward a notion of new duplication of the existing facility.12 Third, the notion of a ‘second facility’ introduces a private meaning of ‘uneconomic’ and runs counter to the Tribunal’s endorsement of a social cost/benefit approach that equates with the identification of natural monopolies: The Hilmer Report suggests that criterion (b) was intended to describe a pipeline which exhibits "natural monopoly characteristics". Whilst there is disagreement between the expert economists in the present case as to what constitutes a natural monopoly, the view expressed by NCC in its Final Recommendation (at 42) is that where a single facility can meet market demand at less cost than two or more facilities, then the facility exhibits "natural monopoly" characteristics. That suggests that if a single pipeline can meet market demand at less cost than two or more pipelines, it would be "uneconomic" to develop another pipeline to provide the same services, because those services are most efficiently provided by the existing pipeline. Thus, Posner in Natural Monopoly and its Regulation (30th edition, 1999) states at 1: "If the entire demand within a relevant market can be satisfied at lowest cost by one firm rather than by two or more, the market is a natural monopoly, whatever the actual number of firms in it. If such a market contains more than one firm, either the firms will quickly shake down to one through mergers or failures, or production will continue to consume more resources than necessary. In the first case competition is shortlived and in the second it produces inefficient results. Competition is thus not a viable regulatory mechanism under conditions of natural monopoly." (Duke decision, paragraphs 60-62). A test of natural monopoly considers economies of production over a relevant range of demand, regardless of existing market structure. A natural monopoly can exist even if there are two providers within a market, although as Posner illustrates, this would mean that there has been inefficient investment in ‘competing’ infrastructure. This suggests that a ‘second facility’ test may not be satisfied even by a natural monopoly. As such, the proposed change would shift the declaration criteria away from a focus on infrastructure with natural monopoly characteristics. 12 See submission of Mr Ian Tonking dated 18 May 2001. 25 Finally, the Council notes that current wording of criterion (b), as interpreted by the Tribunal and the Council, appears to achieve the outcome desired by the Commission. In this sense, it is not clear what purpose would be served by amending the criterion. The Tribunal, in the Sydney Airports decision said: The Tribunal considers, however, that the uneconomical to develop test should be construed in terms of the associated costs and benefits of development for society as a whole. Such an interpretation is consistent with the underlying intent of the legislation, as expressed in the second reading speech of the Competition Policy Reform Bill, which is directed to securing access to "certain essential facilities of national significance". This language and these concepts are repeated in the statute. This language does not suggest that the intention is only to consider a narrow accounting view of "uneconomic" or simply issues of profitability. … If "uneconomical" is interpreted in a private sense then the practical effect would often be to frustrate the underlying intent of the Act. This is because economies of scope may allow an incumbent, seeking to deny access to a potential entrant, to develop another facility while raising an insuperable barrier to entry to new players (a defining feature of a bottleneck). The use of the calculus of social cost benefit, however, ameliorates this problem by ensuring the total costs and benefits of developing another facility are brought to account. This view is given added weight by Professor Williams's evidence of the perverse impact, in terms of efficient resource allocation, of adopting the narrow view (Sydney Airports, 40,793). In the Duke decision matter, the Tribunal endorsed the approach in the Sydney Airports decision, as well as an interpretation of criterion (b) put by the Council in its closing submission: We agree with the submissions of the NCC that the “test is whether for a likely range of reasonably foreseeable demand for the services provided by means of the pipeline, it would be more efficient, in terms of costs and benefits to the community as a whole, for one pipeline to provide these services rather than more than one.” [Duke decision, paragraph 137]. The Council proposed this test in the Duke matter by drawing on the advice of Professor Janusz Ordover, who had been retained to assist the Council in an advisory capacity. In commenting on the appropriate approach to testing for whether the EGP exhibited natural monopoly characteristics, Professor Ordover said: ...Natural monopoly, [occurs where] it is most efficient to have a single firm serve a target level of demand. This condition is satisfied if for 26 the given range of sales, the unit cost (average cost) falls with output. This is a sufficient but not necessary condition. It seems to be the case… that EGP is likely a natural monopoly (in the above sense) for transport of gas from Gippsland Basin to Sydney, at least given the anticipated level of demand for the near (and not so near) future. This does not mean that at some later date, as demand grows, there may not be a room for another pipeline… It is in the context of criterion (a) that questions of market power properly arise. Thus, if a natural monopoly is found to occur, criterion (a) proceeds by considering whether control of the natural monopoly enables the exercise of market power in a dependant market. Thus, the issue in the Duke matter, as expressed by Professor Ordover, became one focussed on the scope for EGP to exercise market power. He said: …literature on contestable markets suggests that even a monopolist can be forced to behave competitively if there are no barriers to entry into the relevant market. There are entry barriers into pipeline transport business (but they can be reduced if a group of sellers/buyers forms a joint venture and removes the entry risk). Hence, we need to look at what other forces can constrain exercise of market power…. such constraint is provided by MSP, which transports gas to Sydney, where EGP also delivers. Hence, if EGP were to overprice delivered gas … it would render Gippsland Basin gas uncompetitive with the MSP gas. This is what constrains the ability of EGP to exercise market power in transport. The full context of the test the Council proposed to the Tribunal in its closing submission in the Duke matter may also help to provide further clarification of the application of this test: The approach favoured by the Council is one that cleaves more closely to the statutory language of criterion (b). The test is whether for a likely range of reasonably foreseeable demand for the services provided by means the Pipeline, it would be more efficient, in terms of costs and benefits to the community as a whole, for one pipeline to provide those services rather than more than one. As the Tribunal correctly observed in the Sydney International Airport case at para 204-206, the costs and benefits which must be weighed for the purposes of criterion (b) are the costs and benefits to society as a whole. The identification of the range of reasonably foreseeable demand for the services provided by means of the pipeline necessarily involves a consideration of what substitutes exist for those services. The existence of substitutes and degree of substitutability will affect the demand for the services themselves. This is what was intended by the Council in its reference at page 48.2 of the Final Recommendation to the essential question for the Council being "what other pipelines provide substitutable services for the services of the Eastern Gas Pipeline". 27 In this exercise, it needs to be recognised that demand for the services of a pipeline is derived from the demand for a bundled delivered gas product. Users of gas transportation services who are consumers or wholesalers of delivered gas may want to buy gas from a particular field because of more favourable terms and conditions or more ready availability. Users of gas transportation services who are producers may want to sell to a particular region because of greater marketing opportunities. It is the combined demand of all potential users that gives the overall demand for the services provided by means of the pipeline. The current criterion (b) is therefore a test of natural monopoly or natural monopoly characteristics. This appears to conform to the interpretation desired by the Commission. There are some risks that the proposed amendment will not achieve the desired outcome, particularly since the amendment itself would suggest that something different is intended. In view of these facts, the case for amendment appears weak. The ‘Tier 2’ proposals The Commission’s Tier 2 proposed amendments to the declaration criteria would: • • • replace the first three criteria with four new criteria: (a)-(d); remove the current ‘health and safety’ requirement; and retain the existing criteria (e) – existing effective access regime – and (f) – the public interest test. This section comments on the proposed new criteria. Proposed criterion (a): Proposed criterion (a) combines a simplification of the existing test of materiality – currently set out in criterion (c) – with a reworking of the current ‘uneconomical for anyone to develop another facility’ test – currently at criterion (b). It includes a change of wording to criterion (b) similar to that proposed as part of the Tier 1 amendments, focusing on ‘the entry of a second provider of the service’ rather than a ‘second facility’, and introduces a new concept of what is ‘economically feasible’ to replace the concept of ‘uneconomical to develop’. Significantly, the proposed criterion focuses on the characteristics of the relevant service, whereas the current criteria (b) and (c) focus on the characteristics of the relevant facility. 28 Use of the wording ‘entry of a second provider of the service’ does not raise the technology bias problem of the Tier 1 proposal, but it does retain the risk of ignoring existing facilities or service providers (see discussion under Tier 1 proposals). Indeed, use of the word ‘entry’ probably exacerbates this latter risk. Thus, by referring to the ‘entry of a second provider’, the new test may inappropriately exclude consideration of existing alternative facilities and service providers. The term ‘economically feasible’ appears at CPA clause 6(3)(a)(i) as part of the coverage test for certification of State and Territory access regimes. Given that the purpose of certification is protection against declaration, the Council has interpreted 6(3)(a)(i) consistently with its approach to s.44G(2)(b). This is not to say that the legal interpretation of ‘not economically feasible to duplicate’ would necessarily be the same as the legal interpretation of ‘uneconomical to develop’. In this sense, the ramifications of adopting ‘economically feasible’ in the declaration criteria are not clear. Nor is it apparent what purpose would be served by such a change. A shift in focus of the national significance test from the relevant facility onto the relevant service could have a significant bearing on the potential scope of declaration. It is possible to envisage numerous scenarios where the facility providing a service would satisfy a national significance test but the particular service would not. This is particularly true of a facility providing a wide range of services. For example, in the Sydney Airports matter, declaration was sought and granted for services comprising the use of Sydney Airport to conduct a number of international air-freight related activities. Given the importance of Sydney Airport in international freight shipments to and from Australia, it seems likely that any national (economic) significance test would be satisfied whether applied to the facility in this case (Sydney Airport) or the relevant service (use of Sydney Airport for international air-freight related activities). However, if the two tests were applied to say, Brisbane Airport, the outcome might be different: the airport might satisfy the significance test but the particular use or uses of the airport might not. There is a question whether this different approach would mean less or greater distortions in access regulation. Leaving aside the application of the Airports Act 1996, should two airports (facilities) with similar characteristics be regulated in different ways because of the significance of particular current usage? Would Brisbane Airport be declared only when air-freight related usage reached a particular significance threshold. Since access regulation is designed to increase usage, making access regulation unavailable because current usage is insufficiently high does not appear to be sensible. In any case, an obvious solution for an access seeker might be to seek declaration of a larger number of the services provided by a facility, to increase the likelihood of the significance test being satisfied. Thus, a party seeking access to Brisbane Airport to provide international air-freight ground 29 handling services might also seek declaration of aircraft landing and take-off services and passenger handling related services. The broader scope of the declaration application could be justified by a perceived need to facilitate access by the applicant’s clients (such as aircraft operators) as well as the applicant. The question is whether it is desirable to introduce incentives for access seekers to broaden the scope of declaration applications. The Council considers that this is not desirable. It is more difficult to predict the likely impact of the proposed shift in the focus of the economically feasible test onto the relevant ‘provider of the service’ rather than the relevant facility. Nor is the purpose of such a change apparent. In the Sydney Airports decision, the Tribunal held that criterion (b) only made sense if the reference to ‘uneconomic for anyone to develop another facility to provide the service’ was interpreted as development by someone other than the incumbent owner of Sydney Airport13; in other words ‘another provider’. This would indicate that the proposed change from ‘facility’ to ‘provider’ in this instance would make little difference. At best, however, we can only speculate. Finally, the Council notes that criterion (b) – as currently interpreted – distinguishes facilities that are natural monopolies or have natural monopoly characteristics. Testing whether a service is a natural monopoly or has natural monopoly characteristics can raise a number of problems. In particular, a facility may acquire natural monopoly characteristics as a consequence of economies of scope in the provision of a range of services. These economies may not be apparent if the focus is restricted to an individual service. This may not raise problems if the test is applied to a range of (perhaps all) services provided by the facility. But what difference is thereby intended by shifting the focus of the test from the facility to most (if not all) uses (that is, services) of the facility? Proposed criteria (b) & (c): The Commission has proposed a two-part essentiality or market power test to replace the promotion of competition test in the existing criterion (a). The first part of this test – in the proposed criterion (b) – is apparently intended to provide a separate test of supply-side substitution in the provision of infrastructure services. The second part of this test – in the proposed criterion (c) – examines the conditions of the dependent market to test whether the infrastructure owner can exercise substantial market power. The lead-up discussion for this proposal, at pages 144-145 of the Position Paper, contains some discussion of markets and substitution. Markets necessarily involve both demand and supply-side considerations. The established definition of ‘market’ in Australia makes this clear: 13 Sydney Airport decision, 40,792. 30 “A market is the area of close competition between firms, or putting it a little differently, the field of rivalry between them (if there is no close competition there is of course a monopolistic market). Within the bounds of a market there is substitution – substitution between one product and another, and between one source of supply and another, in response to changing prices. So a market 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 (Re Queensland Co-operative Milling Association Ltd (1976) 25 FLR 169 at 190).” Given that conventional market analysis already includes an analysis of supply side substitution, separate treatment does not seem appropriate. It may be that the perceived issue arises because of a lack of clarity in distinguishing between services provided at different functional levels in the supply chain.14 For example, the Position Paper includes discussion of supply-side substitution between rail track and roads. Under conventional market analysis, the notion of supply-side substitution is that where a producer can readily shift production from one product to another, then this in itself is sufficient to integrate the field of rivalry between the two products. Thus, the two products are in the same market. But as it is difficult to envisage rail track being converted into a road, or visa versa, supply-side substitution possibilities are extremely limited. Similarly, substitution possibilities in consumption at this functional level in the supply chain also appear limited: for a train operator, a road is not a viable substitute, while for a truck operator, rail track is not useful. Moving downstream in the respective supply chains both truck operators and train operators supply freight transport services in the freight transport market. Thus, there are demand-side substitution possibilities in the freight market as freight forwarders (and final consumers) choose between the transport services provided by train operators and truck operators. The fact that final consumers and intermediaries (freight forwarders) can choose to buy services directly from either truck operators or train operators suggests that these transactions are in the same freight forwarding market. But the services of using a rail track and road could not be said to be integrated into the freight forwarding market in the same way; these services are only useful to businesses operating trains and trucks respectively.15 This point was expressed by Rhonda Smith and Professor Neville Norman: 14 The Council’s approach to functional market definition is outlined at section 5.1.4 in its first submission. It is at least possible to envisage supply-side substitution possibilities in the freight forwarding market: it is possible to envisage that a truck operator may switch to operating trains, at least in the long term. Supply-side substitution in the freight forwarding market would suggest demand-side substitution possibilities in the 15 31 "In practice we can identify production and distribution sequences in which, by analogy, most of A goes into B, and B uses few other materials than A. For example, most raw cotton (A) goes into spun cotton yarn making activities (B), which use few other materials, and that product in turn goes mainly into cotton cloth and clothing (C), which may go into "D", and so on. The functional market question is how large is the range of activities in the A-B-C-D-E sequences that we consider is inside the one market. Are primary activities separate from the manufacturing? Is initial manufacturing to be treated separately from further processing? Is wholesaling separate from retailing? It will immediately be seen that the substitutability tests have no real meaning or application here. What point is there in asking how "substitutable" in one stage in production or distribution for another, when each is a complementary part of the process, like a link in a chain? We can meaningfully ask if one garment is substitutable for another garment, in the minds and intentions of the provider or the buyer; but there is no point in asking whether the materials from which either garment in (sic) made are substitutable for the garment itself! We can thus dispense with (nearly) all we know about substitutability when we come to this difficult functional aspect of market delineation.16 As the QCMA decision shows, the approach of the Tribunal and the courts to market analysis is to consider both supply-side and demand-side substitution possibilities. This approach is well established in Australian law and the Council adopts the same approach; but it is important to note that in many of these matters, supply-side substitution possibilities may be limited. Therefore, the need for a separate criterion to ensure that supply-side substitution possibilities are taken into account must be open to question. Further, the separate criterion would partly replicate the current criterion (b) test of ‘uneconomic for anyone to develop another facility to provide the service’ (the same problem would presumably arise under the Commission’s proposed screening test). Replication occurs because criterion (b) currently takes account of existing substitute services as well as the potential for the development of a new facility to provide the relevant service. As shown elsewhere in this submission, where an economically viable substitute service is already available, the current criterion (b) is not satisfied. purchase of rail track services and road services. The fact that at least one transport firm (Toll Holdings) operates both trucks and trains increases the prospect of such substitution, although it may be that this company simple chooses to operate in different markets. Generally, it would be difficult to say that there is ‘strong substitution, at least in the long run’ between trucks and trains on either the supplyside of the freight forwarding market or on the demand-side of rail track and road services markets. 16 Smith & Norman, Functional Market Definition, (1996) 4CCLJ1 at 10. 32 The energy example used to demonstrate demand-side substitution in the Position Paper recognises that there may be no feasible substitutes for a natural monopoly service – such as the point-to-point gas transportation services of a pipeline – but that conditions in the dependent market may constrain the pipeline owner’s ability to exercise market power in that market. Thus, a gas pipeline operator may not be able to exercise market power in the dependent gas sales market because other sources of gas are available in the gas sales market. Essentially, this was the basis of the Tribunal’s decision in the Duke matter. It therefore seems that the second part of the proposed essentiality test is intended to achieve a similar result to the current criterion (a), as reflected in the Duke decision. Yet some issues about the proposed test would remain: • why not refer to all dependent markets rather than downstream markets; for example, does an electricity wholesale market exist upstream of the associated transmission and distribution systems or downstream? the notion of ‘insufficient competition’ appears to encapsulate an approach to measuring competition which was rejected by the Tribunal in the Sydney Airports decision in favour of a focus on structural impediments to competition. The Duke decision endorsed the focus on structural impediments to competition. the reference to ‘exercising substantial market power’ would appear to refer to some use of market power similar to the concept of ‘take advantage of substantial market power’ in section 46 of the TPA. Is this the Commission’s intention? • • It may be possible to address these issues, but if the intention is to attain the same result as the current criterion (a) – as reflected in the Duke decision – it remains unclear whether there is any benefit to be derived from a change. Proposed criterion (d) The Commission proposes a new and distinct test that access ‘is likely to improve economic efficiency significantly’ to address a concern that the current criteria (as amended by the Tier 1 recommendations) might mean that ‘a substantial increase in competition that yielded only a negligible improvement in economic efficiency could still be sufficient to secure declaration’. It is not clear why a small or negligible increase in efficiency should be avoided. Presumably any assessment of efficiency would take account of the costs and benefits arising from regulation, including the administrative and transactions costs of regulation itself. Thus, all the costs of declaration would already be taken into account before it could be concluded that there was an increase in efficiency. If, having taken all costs into account, there remains 33 an increase in efficiency (even a small increase), this is clearly a superior outcome to the status quo. This approach is consistent with clause 5 of the CPA which encapsulates the notion that any net benefit or efficiency gain is worthy of capture. In the Position Paper the Commission said: Hence, the Commission must consider whether the benefits of access regulation are sufficient to justify its costs and, if so, whether such regulation is the best instrument for pursuing underlying objectives. (PC 2001a, p.33) The Commission’s approach may be justified if another regulatory approach, or possibly some alternative policy approach, would attain the same competition benefits at lower cost, or provide a greater net benefit than declaration. But the Commission has not analysed whether any such alternative would confer a greater net benefit than declaration in situations where the net benefit available from declaration is likely to be small.17 In the absence of such alternatives, any net benefit conferred by declaration would constitute a better outcome than the status quo and satisfy the Commission’s test of the efficacy of access regulation. For these reasons, it is difficult to justify amending the declaration criteria to prevent the community from capturing small efficiency improvements. 3.2 Prices monitoring as an alternative to declaration The Position Paper expressed a leaning in favour of introducing an option of prices monitoring as an alternative to declaration where there is some doubt whether: • • an essential facility has scope to extract monopoly profits; or the benefits of applying declaration to constrain an essential facility’s market power exceed the costs of regulation. The Commission has sought views on the merits of this alternative, and on the most appropriate institutional arrangements to give effect to the proposal. 17 The Commission’s proposal for a prices monitoring alternative to declaration is discussed in section 3.2. The Commission has not, however, proposed this alternative for cases where declaration would involve merely a small efficiency benefit, on the grounds that prices monitoring would achieve the same benefits as declaration at lower cost. Instead, it proposes this alternative where the competition benefits or the net benefits of declaration are not clear. Nonetheless, it may be true that prices monitoring would involve a greater net benefit than declaration in some cases where the declaration criteria would otherwise be met. 34 The Council supports the provision of prices monitoring as an alternative to declaration. Market power problems associated with natural monopoly can vary by degree. The availability of prices monitoring as an alternative to declaration would mean that declaration would not be imposed in some marginal cases where the criteria for declaration are met but where competition may emerge in a dependent market despite the market power of a natural monopoly service provider. Prices monitoring of the service provider would facilitate the appropriate oversight. For these reasons, the Council would consider it appropriate for prices monitoring to be considered in response to an application to declare an infrastructure service. Thus, in response to a declaration application under Part IIIA, a recommendation could be made, as appropriate according to the respective criteria, to: • • declare the infrastructure service for a period of time; declare the infrastructure service for prices monitoring for a period of time; or not declare the infrastructure service. • Institutional arrangements for prices oversight In its Draft Report on the Review of the Prices Surveillance Act 1983, the Commission said that, generally, prices monitoring should only be applied as a matter of government policy following an inquiry by an independent body and recommendation to a Ministerial decision-maker: It is important that the body undertaking the inquiry is not the agency that would implement any policy recommendation. Allocating the functions of providing policy advice and undertaking regulation to the regulator is inconsistent with the principles of good policy making discussed in chapter 4. (PC 2001b, p.83) In Chapter 4 of this Review, the Commission says: In assessing the costs and benefits of the PS Act, it is important to consider the process that was followed by the minister in determining which products, services or organisations would be subject to monitoring. Implementing best practice regulation involves, inter alia, separating the tasks of assessing the likely impact on the community of monopolistic pricing and identifying the appropriate policy response (policy formulation), from the task of administering prices oversight arrangements (policy implementation) (appendix B). This separation is needed to ensure the independence of the inquiry process (PC 2001b, p.61). 35 The Council agrees that a decision to impose prices monitoring should be quite separate from the conduct of prices monitoring for a particular business. While the former is a policy matter, the latter constitutes the implementation of policy through specific regulation. The Council also agrees that the two functions (imposing prices monitoring and conducting prices monitoring) should be the responsibility of separate agencies. The consideration of whether prices monitoring is appropriate involves different questions, information and skills compared to the application of prices monitoring. Separation of these policy and regulatory functions avoids problems associated with a regulator determining its own jurisdiction and imposes few, if any, costs of inconsistency or overlap between the responsible organisations. By analogy with this reasoning, the Council considers that the current institutional separation between the functions of declaration and arbitration under Part IIIA should be maintained. While this seems fully consistent with the Commission’s approach to prices oversight, it is nonetheless contrary to the Commission’s Tier 2 proposal to merge the roles of declaration and arbitration under one body. The Council returns to this issue in section 5.2 3.3 Negotiation and arbitration The Council generally supports the Commission’s proposals for modifying the negotiation/arbitration framework for declared services. Specifically, the Council considers that the following proposals would improve the operation of Part IIIA: • • requiring service providers to supply information to access seeker (6.3); ensuring that arbitration is consistent with the overall objectives of Part IIIA (6.5); promoting the primacy of negotiated outcomes (6.6); improving transparency of arbitration processes (6.7 and 9.9) • • The Council is uncertain that proposal 6.4 is necessary. The proposal would require that arbitration commence 30 days after declaration (unless both parties inform the arbitrator that a settlement is likely). But under current arrangements (s.44S of the TPA) either the service provider or a third party may notify a dispute to the ACCC at any time. The only constraint is the discretion given to the ACCC to terminate the arbitration if the ACCC considers that negotiations have not been conducted in good faith. Below, the Council comments further on some of the Commission’s specific proposals for negotiation/arbitration. 36 Information provision The Council considers that amending Part IIIA to require service providers to supply information to prospective access seekers would greatly improve the efficacy of the negotiate/arbitrate mechanism. In dealing with State and Territory access regimes, the Council’s experience is that information asymmetry between providers and access seekers is a major issue of concern to stakeholders. In general, access seekers need information about: 1. the process through which the service provider will negotiate access to that service; 2. the information a service provider may require from the access seeker prior to negotiating access; 3. the availability of the relevant service; and 4. information on the terms and conditions of access being offered by the service provider and the basis for those terms and conditions. The first two categories of information should be made available for all declared services shortly after they are declared. The provision of this information is not likely to be an onerous requirement on the service provider. The ACCC should have some role in vetting this information. The second two categories should be made available to the access seeker after they have submitted a request for access in the form required by the access seeker. The arbitration provisions could be automatically triggered if there is a failure to provide the information. This may not, however, be necessary, as such a failure would enable the access seeker to notify a dispute in any case. The Gas Code requires service providers to supply information sufficient for access seekers to understand how the terms and conditions of access, especially the tariff, have been derived. In general, information must be provided in the following categories: • • • • • access and pricing principles; capital costs; operations and maintenance; overheads and marketing costs; system capacity and volume assumptions; and 37 • key industry performance indicators used by the service provider. While the Gas Code offers guidance on the type of information that should be made available to access seekers, the Council recognises that the Gas Code is relatively prescriptive in nature. It would be appropriate, within a general regime such as Part IIIA, to limit information requirements to broad categories. Arbitration to require extension/expansion Section 6.7 of the Position Paper discusses the ability of the arbitrator to make determinations in relation to the extension of facilities subject to declaration. Proposal 6.8 recommends that the ACCC be limited in arbitration to require that a service provider permit interconnection, while removing their current ability to require extension of the facility. Included in the discussion before this proposal, the Commission also indicates it does not support the arbitrator being able to require the service provider to expand the capacity of the facility that provides the declared service, though there is no proposal to amend the arbitration provisions to remove that possibility. Sections 44V and 44W outline the matters the ACCC may, and may not, deal with in an arbitration determination for a declared service. Section 44V(2)(d) explicitly recognises that a determination may require the provider to extend the facility. However, s.44V(2) does not limit the matters that may be dealt with, providing that any matter relating to access by the third party to the service may be the subject of a determination. This would implicitly include a requirement for the service provider to expand the capacity of the facility providing the service. Section 44W then outlines a number of restrictions on the ACCC’s ability to make determinations. These restrictions are couched in terms of prohibitions on certain effects, rather than any particular matter. For example, the ACCC cannot make a determination that would have the effect of requiring the provider to bear some or all of the costs of extending the facility (s.44W(1)(e)). The Council notes that the user’s contribution is not necessarily an up-front charge – it may, for example, take the form of an adjustment to tariffs. The Council agrees with the first part of proposal 6.8 – to allow the ACCC to require interconnection with another facility, and considers that s.44V already does this. But removing scope for the ACCC to require extension or expansion of a facility raises serious concerns. This would result in a significant alteration to the current operation of Part IIIA and, to the extent that Part IIIA is seen as a model for State and Territory regimes, to the operation of regimes such as the Gas Code. It is likely that there will be few circumstances where the most efficient approach would be to require a provider to geographically extend a facility. Rather, a more efficient outcome will often be to provide for interconnection 38 and allow the access seeker to construct an extension. acknowledges this point. The Commission However, there may be circumstances, such as extension of pipes within a pre-existing distribution network, where it would be more efficient for the incumbent provider to extend the facility. To address these circumstances, it is appropriate that the arbitrator have the flexibility to require an extension of geographical range, as currently provided in s.44V. In other contexts, the most efficient way of increasing the supply of a service may be for the existing provider to expand capacity, rather than requiring the access seeker (or someone else) to construct another facility to provide a service. This is especially true of facilities where it is efficient to expand capacity incrementally, such as gas pipelines and rail track. There appears to be no policy justification for regulating spare and developable capacity in different ways. If Part IIIA did not apply to developable capacity, infrastructure providers would have strong incentives to design facilities with minimal spare capacity but maximal opportunities to develop capacity. Such an outcome may be contrary to the Commission’s proposed objectives for Part IIIA: promotion of the efficient use of, and investment in, essential infrastructure. The Council considers that it would not be appropriate for the restrictions in s.44W that relate to requiring the infrastructure owner to pay for extensions to be applied to determinations that relate to expansions. The issues surrounding the question of ensuring the infrastructure owner is recompensed for expansions are likely to be more complex than those for extensions. These matters are best left to the discretion of the regulator to determine on a case by case basis. The current provisions in ss.44V and 44W already do this. 39 4. Certification and undertakings Through legislated access regimes, State and Territory governments can ensure the appropriate regulation of infrastructure services within their jurisdiction. They can determine the form of regulation by establishing the institutional arrangements, pricing frameworks and dispute resolution methodology they consider would meet the needs of market participants. They can apply the regimes to individual or multiple service providers. As long as the State and Territory regimes are consistent with the clause 6 principles they will operate to the exclusion of the national access regime. This is what the Council of Australian Governments (CoAG) clearly intended in agreeing to introduce Part IIIA and reflects the co-operative approach taken by the Commonwealth, State and Territory Governments. By contrast, service providers do not have the same degree of control over the form of access regulation that will apply to them through the undertaking process. An undertaking needs to be accepted by the ACCC before it provides protection from the declaration provisions of Part IIIA and it is not possible for a State or Territory Government to legislate to determine the content of an approved undertaking under Part IIIA. It is possible for Governments to legislate to require service providers to offer undertakings to the ACCC in a particular form. However, this does not limit the ACCC’s discretion in deciding whether to approve or not approve an undertaking. The undertaking process does not provide State and Territory Governments with the same control over the form of access regulation as certification. Each of the State and Territory access regimes that are currently certified provides a regulatory framework to assist negotiation and resolve disputes over terms and conditions. On the whole these regimes do not prescribe the terms and conditions of access; rather, they establish mechanisms to have those determined. While approaches vary across regimes, common features are to establish: • • a regulator who provides independently verified information to assist negotiation; and a dispute resolution process to address situations where negotiations fail. The concept of the State regime as a regulatory framework for creating institutions and mechanisms to assist access negotiations – as distinct from specifying actual terms and conditions of access – is evident in the Gas Code. The Gas Code enacted in each State and Territory establishes processes for coverage (and revocation of coverage) of pipelines as well as establishing institutions to administer those processes. Further, it creates a right of access to covered pipelines and obliges service providers to develop terms and conditions of access (access arrangements) consistent with broad principles set out in the Code. It establishes regulators to assess these access arrangements, appeals bodies to provide accountability and arbitrators to resolve disputes. 40 The Gas Code was developed through a consultative policy advisory body, and its operation is now monitored by a similar body established under the regime, the National Gas Pipelines Advisory Committee. NGPAC is not concerned with examining the particular terms and conditions of access for specific pipelines. Its role is to ensure appropriate regulatory design, which it achieves through recommending necessary modifications to the overarching rules that make up the Gas Code. Similarly, the Council and the ministerial decision-maker under the certification process are concerned with issues of appropriate regulatory design rather than with questions about whether particular access prices are appropriate. Conversely, the purpose and design of voluntary access undertakings are quite different. An access undertaking is primarily concerned with the terms and conditions of access offered by a particular service provider – though the criteria do not exclude the possibility that an undertaking could provide a broader framework for access for a particular service provider. There are, however, a number of problems with using the undertaking process to give force to access regimes for multiple service providers. The criteria for accepting an undertaking, unlike the criteria for certification, are currently not directed toward issues of regulatory design. This would be less of a problem if the criteria were rewritten to be more consistent with the clause 6 principles. This is discussed further in section 4.2 of this submission. However, providing for the regulator to accept undertakings in the form of overarching access regimes (rather than focusing on terms and conditions) would raise wider concerns. In particular, it may not be appropriate for the Part IIIA regulator to also have the role of assessing whether it or other (usually State or Territory) regulatory bodies are independent, adequately resourced and otherwise able to perform regulatory functions within an appropriate framework of rules. In addition, as an undertaking is a voluntary process, as distinct from government policy and legislation, it contains no provision for a coverage test. Amendments to the undertaking provisions to allow for industry codes were necessary to allow the National Electricity Code to be dealt with through the undertaking process. State and Territory legislation enacted the National Electricity Code and required the relevant service providers to submit undertakings consistent with the legislation. The electricity access regime can be seen as a hybrid: utilising both the legislated state regime process and the voluntary undertaking process. The undertaking process thereby did not, and could not, take account of all matters that are examined in certification proceedings. The National Electricity Code process may not provide much direction on how the undertaking process could work to replace the certification process. 41 4.1 Specific proposals for certification The Position Paper recommends a number of specific amendments to the certification process as well as amendments to decision-making arrangements under Part IIIA that affect certification. The Council’s views on the latter proposals are discussed under section 5 of this submission. The Council supports the reasoning behind the proposal 7.1. It is appropriate for Commonwealth access regimes to be tested against the same criteria as State and Territory access regimes. However, as the Commonwealth has the power to exempt the services subject to these regimes from Part IIIA, certification is not necessary to protect the services from declaration. Therefore it might be more appropriate to require the Commonwealth to submit their regimes to the Council for an assessment of their effectiveness against the clause 6 principles (without that assessment having any formal or legislative requirements). The Council could then publish its assessment. The Council also supports proposal 7.2 – that the same test of effectiveness be applied to regimes irrespective of whether the services covered are government or privately owned. The Council’s views on proposals 7.3 and 7.4 are set out below. Including the certification principles in Part IIIA The Council is not convinced that proposal 7.3 – to write the certification principles into Part IIIA – is necessary or would result in the benefits identified by the Commission. In the Council’s view, the certification principles are already included in Part IIIA by virtue of ss.44M(4) and 44N(2). These provisions effectively ‘call up’ the clause 6 principles into Part IIIA, and require the Council and the Minister to apply those principles in formulating their respective recommendations and decisions. The current arrangement is akin to Part IIIA calling up the clause 6 principles as an attached code set by a CoAG intergovernmental agreement (the CPA). As such, the certification principles can only be amended with CoAG approval. In the Council’s view, the underpinning co-operative nature of this framework has engendered confidence among States and Territories in the certification process and has contributed to its considerable use (nine regimes covering gas, rail and port services have been certified as effective, with additional activity currently underway in electricity, gas and rail). If the principles were to be written directly into Part IIIA, the Commonwealth would acquire exclusive control over them and could amend the principles unilaterally. As identified by the Position Paper, this could undermine the co42 operative approach taken by governments in developing and implementing Part IIIA. A separate but related matter is the proposed inclusion of the objects clause and pricing principles into Part IIIA. As it is intended that these provisions would apply to all processes under Part IIIA – including certification – it would be appropriate for State and Territory Governments to be consulted in respect of any such amendments. Modifications to the clause 6 principles The Council agrees with the Commission’s view that the current approach to the clause 6 principles has worked well, and expects that it will continue to work into the future. Flexibility is important as the circumstances relevant to certification of particular infrastructure access regimes can vary widely. The Council considers that all of the matters identified by proposal 7.4 (apart from the second dot point) should be reflected in an effective access regime. More or less explicitly, the proposed list reflects the current clause 6 principles and have already been incorporated in all certified State and Territory access regimes. Nonetheless, the clause 6 principles are at times vague and imprecise (for example, in regard to pricing principles) and the Council would welcome a more explicit approach in some areas. Any modification should not, however, affect regimes that have already been certified. Any rewriting of the criteria would need the approval of CoAG. Legislative amendment would not, however, be required. Price Cap arrangements Proposal 8.2 floats an additional criterion for certification: inclusion of an explicit requirement for productivity-based price cap arrangements. The Council is concerned that this type of provision is too prescriptive and is not an appropriate inclusion in the overarching criteria. The certification principles are concerned with the regulatory framework of a regime rather than prescribing its details. Part of the approach is to verify whether a range of tools is available to an independent regulator with the discretion to use them as appropriate. However, it would not be appropriate to be too prescriptive on matters such as pricing methodologies, as knowledge and understanding of these areas is constantly evolving. There may a danger in locking regulators into an approach that could, in the future, be superseded. 43 While most regulators are working toward the use of efficiency-enhancing arrangements such as those proposed by the Commission, making such arrangements mandatory is unlikely to be constructive. 4.2 Specific proposals for voluntary undertakings The Council supports proposals 7.5, 7.6 and 9.4 which recommend a number of amendments to the undertaking criteria and processes The Council recommended amendments consistent with proposals 7.5 and 7.6 in its first round submission. The Council proposed that the undertakings process could be improved by including criteria similar to those set out in the clause 6 principles. The Council considers that this would provide greater certainty to service providers on what undertakings might be acceptable. The Council proposed that the criteria could outline that an undertaking should include: • provisions that accommodate the reasonable needs of access seekers by facilitating access through timely and clear processes; provisions to ensure appropriate information disclosure, particularly if the undertaking adopts a negotiate/arbitrate model; an appropriate dispute resolution process; and an approach to pricing that reflects the efficient use of, and investment in, the infrastructure. • • • Including criteria of this nature, coupled with the inclusion of a binding objects clause and pricing principles in Part IIIA, would provide greater certainty to service providers and direction to the regulator in the development and approval of access undertakings. The Council’s views on proposal 9.4 are provided at section 6 of this submission. Undertakings by ‘non-providers’ The Commission has sought views on whether someone other than the provider of a service in terms of Part IIIA should be able to offer an undertaking to the ACCC. 44 The Australia Rail Track Corporation (submission 28, p. 11) has noted that it is currently unable to lodge an undertaking in relation to re-supply of track access for parts of the interstate rail network that it does not own or lease. According to the ARTC, this is frustrating an agreement between relevant transport ministers (except for the Northern Territory) on a ‘one-stop shop’ approach to the supply of national train-paths. The South Australian Government (submission 36, p. 9) has raised a similar issue. It suggested that an undertaking was not an option in relation to access arrangements for the Tarcoola to Darwin railway. This was because, at the time, a competitive tender was underway to select the builder/owner/operator of the track and the provider of the service could not be identified until after the completion of that tender process. However, the relevant Governments wanted the access regime finalised to provide certainty in the tender process. As the Council notes in section 2.5, the provider of a Part IIIA service is the person or body than controls the use of a facility; that is, the entity that can determine whether access is provided and the terms and conditions of access. Under some conditions, multiple independent providers are feasible, perhaps even desirable. For example, a capacity auction for, say, rail, water or gas infrastructure may provide for multiple service providers using the same infrastructure with a separate system operator. Each owner of a part of the capacity of the infrastructure would be able to deal in that capacity independently of and in competition with, the other part-owners. This model provides a means of introducing competition in the provision of what might otherwise constitute natural monopoly infrastructure services. The vesting of property rights in water to facilitate trade in these rights under the National Competition Policy Water Agreements incorporates the principles of this model. The 1997 ARTC Agreement does not accord with this model. Under the Agreement, the ARTC owns or operates (or enters a capacity agreement with the owner of) different parts of the defined interstate rail network.18 The capacity agreements would allow the ARTC to resell rights to use those parts of the interstate network it does not own or operate. Under the Agreement, the ARTC has an ‘exclusive right to sell access for interstate operations on the interstate network for the life of this Agreement’ (clause 7.2), even where access might also be available under a state access regime. In terms of the current definition in Part IIIA, the ARTC would not be a service ‘provider’ for those parts of the interstate network that it does not own or operate (but merely acts as a reseller of access). Consequently, the ARTC 18 The ARTC does not own or operate parts of the interstate rail network in NSW (east of Broken Hill), Queensland and Western Australia (west of Kalgoorlie) and capacity agreements in relation to all of this track have not been concluded. 45 cannot submit an undertaking to the ACCC on these resale arrangements.19 The efficacy of any such undertaking may be in question, since the ARTC can only offer to sell access at prices reflecting the wholesale agreement arrangements plus a margin to cover its own costs. If the wholesale agreements do not accord with good regulatory principles, the agreements may have to be renegotiated for any undertaking to be accepted. An alternative to amending Part IIIA (to allow the ARTC to lodge an undertaking in relation to the resale of track access) is for the relevant track operators to provide an undertaking on how access would be provided to the ARTC, or otherwise on how the seamless provision of access to interstate train operators might be achieved. This was discussed in the Council’s first submission at sections 6.14 and 6.14.2. 19 In light of the Federal Court decision in Hamersley Iron Pty Ltd v National Competition Council and others (1999) ATPR 41-705, there is also considerable doubt whether interstate and intrastate train operations involve different services for the purposes of Part IIIA. 46 5. Administrative and procedural matters The Position Paper proposes two amendments that would lead to significant changes in the current institutional arrangements under Part IIIA. Proposal 9.1 (Tier 1) would remove the decision-making role of Ministers in declaration and certification processes. Proposal 9.2 (Tier 2) would combine the currently separate roles of coverage and certification recommendations/decisions with regulation of specific terms and conditions. 5.1 The role of Ministers As recognised by the Position Paper, the involvement of Ministers in the coverage decision making process is consistent with the recommendations of the Hilmer Committee, though the Committee considered that only the Commonwealth Minister should be responsible for decisions – even for State or Territory owned infrastructure. Further, the Hilmer Committee did not recommend merits review of the Minister’s decisions. The Committee determined that the Minister was in the best position to balance the various interests involved in declaration and make the final decision. Presumably any decision would have been subject to the Administrative Decision Judicial Review Act (ADJR Act) and the other forms of administrative and judicial review available for Commonwealth Government decision-making processes. The current arrangements reflect a wider balance of interests, including the interests of States and Territories in participating in the regulatory decisions affecting their own infrastructure. The review mechanism ensures consistent interpretation of the coverage criteria, mitigating the potential costs of having multiple decision-makers. Time constraints on ministerial decision-making mean that their consideration does not significantly contribute to the overall time taken to finalise a declaration application. The Council considers that ministerial involvement reinforces the policy nature of the coverage process, clearly distinguishing it from the more technical regulatory role of determining or approving terms and conditions of access. The importance of this distinction is discussed below. Further, the Council considers that Ministers, as elected representatives, may have particular insights into the criteria, especially those related to national significance and public interest that could differ from the advisory body. In relation to both the Carpentaria and SCT (Western Australia) declaration applications, the public interest criterion featured in the Ministers’ reasons for decision differently from the Council’s recommendations. 47 The Council supports proposals 9.3 and 9.8. The current sixty-day time limit for the decision-maker under declaration should be maintained, with a similar limit being imposed for certification decisions. If a decision is not made within that time, the recommendation of the advisory body should be deemed to be the decision. To improve transparency, Ministers should, however, be required to publish reasons for their decisions. For similar reasons, the Council supports the proposal (9.6) for legislative provision to be made for public comment on Part IIIA processes. 5.2 Institutional arrangements The Commission proposes that the Part IIIA functions of the Council and the ACCC be combined, for the following reasons: • the perception that “the envisaged delineation of responsibilities between the ACCC and the Council has become blurred. In particular, overlaps between the certification and undertaking processes have emerged” (PC 2001a, p229); the limited degree of discretion involved in coverage decisions weakens the case for having separate bodies; separation of regulatory responsibility carries risks of inconsistent interpretation; limited expertise is spread too thinly between different bodies; and separating responsibilities is likely to lead to an increase in time taken to establish terms and conditions of access. • • • • The Position Paper, while recognising some potential costs of combining the roles, does not fully explore the implications of the proposed amendments. Possible implications include: • convergence of the two elements of the ‘two-part process’ inherent in Part IIIA, or at least reduced transparency of coverage decisions; perceived conflict between the roles of deciding whether to regulate and then the form of the subsequent regulation. This raises the risk of damaging confidence in the efficacy of the processes; and undermining the role and scope of State and Territory regimes. • • 48 Two-part process Part IIIA clearly provides for, and relies upon, a separate, rigorous and regular examination of the need to apply access regulation to specific infrastructure services on a case by case basis. The importance of this coverage process has been reinforced by most of the submissions to the Commission’s inquiry and by the findings of the Position Paper. There was criticism of some industry specific regimes for not including similar coverage mechanisms. The importance of the coverage process is further demonstrated by the fact that each declaration application considered by the Council has been vigorously contested, involving a wide range of industry participants in debate and considerable input into the Council’s processes by interested parties. Under the current arrangements, the coverage decision is clearly distinguishable from determinations of specific terms and conditions. This separation of process establishes distinct accountability mechanisms. Coverage decisions involve different questions, information and skills compared to the arbitration of access disputes or the regulation of access. Coverage decisions are concerned with broad policy issues such as identifying natural monopoly infrastructure and analysing current and prospective competitive conditions in relevant markets. Conversely, arbitration and regulation focus specifically on the regulated infrastructure. It involves analysis of specific access prices and underlying costs, asset valuations, depreciation, rates of return and prices as well as a range of requirements for the actual provision of third party access. This specialisation ensures development of expertise and removes the risk of coverage questions being ‘caught up’ in specific questions of regulatory intervention. The division that flows from this framework provides greater transparency of process and decision making. Combining the coverage and regulatory roles within a single organisation, risks losing the separateness of these processes. In doing so, it may perhaps reduce the likelihood of subsequent access outcomes being resolved through negotiation rather than arbitration. While it is possible to ring-fence the coverage role from the regulatory role within a single organisation, the necessary degree of ring-fencing may negate any possible benefits of combining the roles, and yet the inherent tensions would remain. Conflict in roles As recognised by the Commission in both the Position Paper and the draft report on the Prices Surveillance Act, separation of the coverage and regulatory roles is likely to promote independent decision making – especially 49 where considerable discretion is left to the decision-maker. With respect to Part IIIA, the Commission considers that the level of discretion in coverage decisions has been constrained by the declaration criteria – and would become more so with the proposed increase in ‘standardisation’ of the coverage criteria. As discussed in section 6 of this submission, the exercise of substantial discretion by an administrative body is inevitable in any workable access regime. There is no overlap, or potential for areas of conflict on questions of coverage within Part IIIA. Currently the Council and various Ministers are responsible for coverage, with the ACCC responsible for regulation. The ACCC role in accepting voluntary undertakings does not involve the type of considerations relevant to applying the declaration criteria. Independent and distinct processes are essential to the effective operation of the coverage test. There is a danger that if the regulator also makes the coverage decision, participants – especially service providers – may feel unable to contest the applications to the same degree for fear of alienating the potential future arbitrator. Further there is a danger that a regulator may be perceived as having a particular mind-set that inevitably leads to regulation or is disinclined to appropriate testing for continued coverage. Such perceptions are plausible because of the essential conflict between the roles of coverage and regulation of terms and conditions post coverage. A consequence of these perceptions would be decreased confidence in and acceptance of Part IIIA as an effective regime. As discussed in section 4 of this submission, the certification and undertakings processes are essentially different in nature. The acceptance of particular terms and conditions of access through the undertaking approval process is a regulatory decision while certification is a policy role, concerned with the overall design of regulatory schemes rather than specific terms and conditions of access to particular services. Those bodies responsible for the administration of the certification and undertaking processes will continue to have significant discretion in deciding to certify regimes or accept undertakings. In the case of certification, such discretion is essential given that governments need to design regimes that are appropriate for particular infrastructure services or across an industry. Similarly, a voluntary undertaking will vary with the circumstances of the infrastructure service. Given this level of discretion and variability, it is undesirable to have decisions about the overall design of an access regime, including questions of the identity and powers of the relevant regulator, to be made by a regulator. The regulator would then be required to determine not only what rules it can apply within a particular regime, but also to apply the rules. The regulator may also be required to administer a test for independence and then 50 determine whether it meets its own test. Alternatively it may be that a national regulator would be required to make these assessments of other state based or industry specific regulators. Potential for redundancy of certification process Certification and undertakings are essentially different processes. While some of the output of those processes may appear similar – thus blurring the distinctions between them, those distinctions should remain important. Indeed, the Commission recognises a continued role for both, and the majority of submissions support this. The Council considers that giving responsibility for both certification and undertakings to the same body would further blur the distinctions between the two and raise questions about the continued role of both processes. As certification is the more limited process, only available to governments, it is the most likely candidate for redundancy. The role of State and Territory access regimes is to establish rules, processes and institutional arrangements for determining access terms and conditions, rather than to determine those terms and conditions themselves. In general, the latter degree of detail is not desirable in an access regime because it may not be flexible enough to deal with the circumstances of all of service providers and access seekers covered by the regime. If the same body were to administer both certification and undertakings, there would be a risk that government access regimes would become more prescriptive – indeed, more like undertakings. Alternatively, government regimes may become one-line pieces of legislation requiring identified service providers to submit an undertaking. It is the separation from the determination of particular terms and conditions that ensures certification is concerned with overall good regulatory design and can focus on issues that relate to that concern. The Council strongly supports a continued role for certification and recognition of that its purpose is quite distinct from the purpose of the undertaking process. This distinction is achieved under the current arrangements. The proposal to combine the roles risks losing an effective, distinct certification process that has helped deliver good access regulation across a number of industries in recent years. Perceived costs and benefits of the current arrangements In responding to the Commission’s proposal to merge policy and regulatory responsibilities under Part IIIA, the Council has argued that the reasoning underlying the proposal is not well established. In doing so, the Council has 51 drawn attention to the considerable benefits in maintaining the current institutional separation. In summary, these benefits include: • • • • preserving the importance of coverage processes; greater transparency; avoiding the risk of conflict or perceptions of conflict; and the development of distinct expertise in policy processes and regulatory processes respectively. The Commission also outlined a number of perceived costs of the current institutional arrangements. The Council considers that these perceptions are ill-founded. It has been suggested that the regulatory expertise in the public sector is too thin to be spread across multiple bodies, the inference being that processes are suffering because of a lack of expertise. This suggestion fails to recognise the difference in expertise required in considering the policy questions of coverage from those required in determining terms and conditions of access. Further, it has been the Council’s experience that while coverage considerations are important, they do not use up significant resources in the overall context of the regulatory scheme. It has been possible for the Council to deal with declaration, certification and gas coverage applications with a small number of staff dedicated to these tasks, bringing in external consultants as needed. It is likely that a single body dealing with both coverage and regulation would need to replicate these arrangements using similar resources. Another perceived cost is the potential delay in determining terms and conditions of access after declaration, on the grounds that an arbitration body would need to come to terms with the same information that the coverage body has already considered. The Council does not regard this as a cost for two reasons. First, the information and skills required for each process are considerably different, with little scope for overlap. Second, the Part IIIA framework recognises that parties should be given the opportunity to negotiate after a declaration decision has been made. There is no guarantee that arbitration will be required – and certainly not immediately. Indeed, arbitration has not been required at all to date. The risk of inconsistent interpretation of criteria has been identified as a potential cost. The greatest potential for inconsistency arises between interpretation of the certification and undertakings criteria. The Council has argued in this submission that the purpose of these processes is fundamentally different. This difference of purposes mitigates the risk of 52 significant inconsistency and means the consequence of any inconsistency is not material. As outlined in the Council’s first submission, Part IIIA provides three routes for a potential access seeker: declaration, access through an effective regime or access through an approved undertaking. It is not proposed by the Position Paper that any of these routes be removed. It is important for the criteria and processes for each route to be considered in a way that consistently applies the objectives of Part IIIA, but it is not necessary for the criteria and processes themselves to be identical for this goal to be achieved. Conclusion The current delineation of responsibilities between the Council/Minister and the ACCC is premised on the separation of responsibility for ‘policy decisions’ from ‘regulatory decisions’. The Hilmer Committee considered that deciding which infrastructure services should be regulated is essentially a policy decision for a Government Minister, subject to specific criteria and on the advice of an independent body. The Committee also considered that the ongoing regulatory and administrative work should be undertaken by the general competition regulator to concentrate regulatory expertise, co-ordinate regulatory activity across industries and reduce costs. To date, the current institutional arrangements have worked well. The Position Paper does not identify any specific failures, but suggests there could be benefits from amending the current arrangements. The Council considers those possible benefits to be overstated, and almost certainly outweighed by the risks the Council has identified. 53 6. Reviews and Appeals It is a well-established principle of Federal Government administration that the exercise of discretion by an administrative body (such as the Council or ACCC) should be subject to review or appeal by a judicial or quasi-judicial process. This principle is reflected in the establishment: • • • of a generic administrative review tribunal: the Administrative Appeals Tribunal, the Administrative Review Council to oversee and monitor the Australian system of administrative review; and in specific review bodies such as the Australian Competition Tribunal. Further, the legal exercise of jurisdiction and administrative discretion by bodies such as the Council and ACCC is subject to scrutiny by the courts under general administrative law. An administrative body is conferred jurisdiction by legislation. Thus, the Council and ACCC derive jurisdiction on access matters under Part IIIA. Within the bounds of the jurisdiction conferred, an administrative body exercises administrative discretion. This is a key characteristic of administrative bodies. It provides scope for flexibility and efficiency as compared with the determination of issues in judicial processes. At the bounds of its jurisdiction, an administrative body faces the risk of jurisdictional challenges in the courts.20 Any attempt to restrict the discretion of an administrative body by tightly constraining jurisdiction would substantially increase the risk of jurisdictional challenge in the courts and undermine the benefits of administrative processes. Thus, the exercise of substantial discretion by an administrative body is inevitable in any workable regime and checks on the abuse of administrative discretion fall to review and appeal processes. Generally, a review of an administrative decision can involve either a full reconsideration of the matter by the review body (full merit review) or a consideration as to whether appropriate principles of administrative decision making have been applied21. 20 See, for example, the two jurisdictional challenges against the Council in the Hamersley matter and the current Western Power matter. See the grounds for review contained in the Administrative Decisions (Judicial Review) Act. 21 54 Currently under Part IIIA, full merit review by the Tribunal is available for decisions by relevant Ministers on declaration and revocation of declaration, decisions by the Commonwealth Treasurer on whether or not to certify a State or Territory access regime as effective, and on arbitration decisions by the ACCC in the resolution of disputes in relation to declared services. The Council considers these review provisions to be important, given the significant implications for property rights that can arise under Part IIIA. As such, the Council supports the Commission’s Proposal 9.4 for the inclusion in Part IIIA of full merit review by the Tribunal of ACCC decisions on undertaking applications. For similar reasons, the Council does not support the Commission’s Proposal 9.5 for the abolition of ‘appeals’ against decisions to declare services under Part IIIA. The Commission justifies the latter proposal on the grounds that it would address concerns about the ‘cumbersome and time consuming nature of the appeals process’, noting that ‘it is hard to see how significant time could be saved without reducing current appeal rights’ (PC 2001a, p240). The Council considers the reasoning behind this proposal to be flawed in several respects. First, few infrastructure owners would regard the ready imposition of declaration without the availability of review as ‘the least risky area for removing appeal rights’. Declaration is a serious step that (quite deliberately) changes the nature of access negotiations between the parties by providing access seekers with a ‘legal right to negotiate access’.22 Other Commission proposals, in particular those aiming to increase certainty in negotiation and arbitration processes (such as proposals 6.3, 8.1 and 9.9) are likely to increase the ramifications of this legal right and would therefore intensify the impact of declaration on an infrastructure owner. Second, there is inevitably a trade-off between due processes to protect the interests of parties, especially infrastructure owners, and timely resolution of issues. Generally, the Council has supported an emphasis on achieving sound decision-making processes, rather that speedy (and possibly flawed) decisionmaking. Third, the Tribunal’s decision in the Duke matter has demonstrated that time limits on review and appeal processes can be effective. The Duke matter was before the Tribunal for approximately six months. Considering that the Duke matter is likely to be one of the more complex and difficult coverage proceedings to come before the Tribunal, six months would appear to be very 22 See s.44S of the TPA. 55 timely progress. Indeed, there is a case for imposing time limits on all decision-making bodies under Part IIIA in the same manner as exists under the Gas Code. Fourth, the proposal appears to be at odds with general Commission thinking on review and appeals processes, as reflected in proposal 9.4. Proposal 9.5 certainly runs counter to general thinking on administrative law and the role of administrative review in Australia. Finally, proposal 9.5 overlooks the risk that removal of review rights in relation to declaration would increase incentives for jurisdictional challenges (such as in the Hamersley and Western Power Federal Court proceedings). The proposal also increases the likelihood of applications to review declaration decisions on questions of law under the ADJR Act.23 Such applications would be more likely if other forms of appeal are removed. The proposed ‘tightening’ of the declaration criteria to deal with perceived problems associated with concentration in Part IIIA decision-making would be likely to increase opportunities for such challenges. These proceedings, would have greater potential to delay the implementation of declaration than a review by the Tribunal, and would not provide the comprehensive check on abuse of administrative discretion that is available under a full merit review. 23 Reviewable questions of law would include whether the decision-maker failed to take account of a relevant consideration, took account of an irrelevant consideration or acted unreasonably in the consideration of the declaration application. 56 References AusCid 2001 (Australian Council for Infrastructure Development), Submission to the Productivity Commission in Response to the Position Paper on the Review of the National Access Regime, June. Hilmer Review (Independent Committee of Inquiry into National Competition Policy) 1993, National Competition Policy, AGPS, Canberra, August. King, Stephen 2000 "Access: What, Where and How", in Productivity Commission and Australian National University 2000, Achieving Better Regulation of Services, Conference Proceedings, AusInfo, Canberra, November. NECG, Joint Industry Submission to the Productivity Commission’s Review of the National Access Regime, June 2001. NCC 2001 (National Competition Council) Legislation Review of Clause 6 of the Competition Principles Agreement and Part IIIA of the Trade Practices Act 1974: Submission to the Productivity Commission, January 2001. Norman, N. and Williams, P., 1983 “The analysis of market and competition under the Trade Practices Act,” 11 ABLR 396. PC (Productivity Commission) 2001a, Review of the National Access Regime: Position Paper, AGPS, March. — 2001b, Review of the Prices Surveillance Act 1983, Draft Report, Melbourne, March. Posner, Richard 1999, Natural Monopoly and its Regulation, 30th edition. Cases Duke Eastern Gas Pipelines Pty Ltd (2001) ACompT 2. Hamersley Iron Pty Ltd v National Competition Council and others (1999) ATPR 41-705. Queensland Wire Industries Pty Ltd v Broken Hill Pty Co Ltd (1989) 167 CLR 177. RAC v NSW Minerals Council (1998) 158 ALR 323. Re Queensland Co-operative Milling Association Ltd; Re Defiance Holdings Ltd (1976) 25 FLR 169. Sydney International Airport; Re Review of Declaration of Freight Handling Facilities (2000) ATPR 41-754. 57