COMMERCE COMMISSION V BAY OF PLENTY ELECTRICITY LIMITED HC WN CIV-2001-485-917

COMMERCE COMMISSION V BAY OF PLENTY ELECTRICITY LIMITED HC WN CIV-2001-485-917 13 December 2007IN THE HIGH COURT OF NEW ZEALAND WELLINGTON REGISTRY CIV-2001-485-917BETWEEN COMMERCE COMMISSION Plaintiff AND BAY OF PLENTY ELECTRICITY LIMITED Defendant Hearing: 12-16, 19-23, 26-28 February, 1, 2, 5-8 March 2007 Coram: Clifford J Professor Martin Richardson (Lay Member) Appearances: B Brown QC, D Laurenson and P Rainsford for Plaintiff J Farmer QC, H McIntosh and T Smith for Defendant Judgment: 13 December 2007JUDGMENT OF THE COURTSolicitors: Commerce Commission for the plaintiff Russell McVeagh, Wellington for the defendantContents Introduction [1]Glossary [10]The pleadings• The Commission's central assertion [11]• The s 36 case [15]• Section 36 – particulars [18]• Section 36 – relief [20]• Section 36 – BOPE's position [23]• The s 27 case [28]• Section 27 – BOPE's position [32]Overview of the evidence [34]Industry background [52]• Electricity industry reform [53]• Meters and switching [71]BOPE acquires an energy business [80]1999-2000 – the market reality [83]BOPE's approach to competitors' requests for access to its meters [97]• Individual retailers- Mighty River [98]- Genesis [124]- TrustPower [151]- Meridian [172]• Other retailers [193]- EOL [194]- Empower [202]- Contact [209]• Evidence from consumers [215]• Other evidence of BOPE's dealings with competing retailers [223]• BOPE's standard letter [230]Our conclusions on BOPE's approach to competitors' requests for access to its meters [235]The legal framework [275]• Section 36 [276]- Market [280]- Dominance/substantial market power [283]- Use/taking advantage [305]- Purpose [319]• Section 27 [326]- Purpose [330]- Effect/likely effect [342]- Substantially lessening competition [345]- Market [352]Discussion• Overview of analysis [353]• Section 36- The unusual feature of this case [365]- Refusal to supply [371]- Raising rivals' costs: The creation of barriers to entry [399]- First alternative analysis – cost imposing obstructive conduct [441]- Second alternative analysis – market power independent of cost-imposing conduct [483]- BOPE's purposes [493]• Section 27- Arrangement or understanding [531]- In a market [539]- Effect/likely effect – local market [545]- Effect/likely effect – national market [546]- Purpose – local market [547]- Purpose – national market [550]- Conclusion [551]Judgment [553]Introduction[1] The Electricity Industry Reform Act 1998 ("EIRA") required local energy companies to separate their electricity lines and electricity supply businesses by 1 April 1999. Those companies had, therefore, to sell one or other of those businesses by that date. [2] In the eastern Bay of Plenty ("the Eastern Bay") the local energy company – at that time called Bay of Plenty Electricity Limited – decided, as did many others, to sell its electricity supply business. [3] A 50:50 joint venture comprising Todd Petroleum Mining Company Limited – acting through its wholly owned subsidiary Todd Energy Limited – ("Todd") and Pacific Hydro Limited ("Pacific") agreed to purchase that business. [4] The joint venture partners incorporated a new company, also using the name Bay of Plenty Electricity Limited ("BOPE"), and BOPE became the agent of the joint venture for the purpose of running the newly acquired electricity supply business. The former Bay of Plenty Electricity Limited changed its name to Horizon Energy Distribution Limited ("Horizon"). [5] The transaction was structured as the sale and purchase of assets. Included in the assets acquired by BOPE were approximately 22,000 electricity meters located at customers' premises. Horizon retained ownership of, and continued to operate, the local electricity lines network in the Eastern Bay ("the Horizon Network"). [6] Most, but not all, energy companies which sold their electricity supply businesses also sold their meters. In this way, most meters came to be owned by the firms, like BOPE, which acquired those electricity supply businesses. [7] In these proceedings the Commerce Commission alleges that BOPE breached ss 36 and 27 of the Commerce Act 1986 ("the Act") in the way in which, from on or about 1 April 1999 onwards, it dealt with requests from potential competitors for access to its meters.[8] The Commission's central allegation is that BOPE, as the owner of the meters located at its retail customers' premises, prevented and deterred potential competitors from supplying electricity to those customers by denying those competitors access, on economically viable and practicable terms, to those meters. Competitors were forced to install their own meters, or buy meters from BOPE. At the same time BOPE engaged in conduct which made it more difficult for BOPE's customers to deal with those competitors. The direct and indirect additional costs thus incurred by those competitors were economically significant. In terms of s 36, as it is now expressed, BOPE had taken advantage of a substantial degree of market power for a prohibited purpose. In terms of s 27, it and its shareholders had entered into an arrangement that had as its purpose, effect or likely effect, a substantial lessening of competition. Moreover, BOPE had given effect to that arrangement. [9] BOPE's response is that the Commission's market analysis was wrong. BOPE did not have substantial market power to use. BOPE's decisions were commercially rational, and made in its own business interests. They were not made with the purpose of hindering or preventing, and did not in any material way hinder or prevent, other firms from competing with BOPE.Glossary[10] Various terms are defined where they first appear in the text of this judgment. We list those terms and definitions here:Act means the Commerce Act 1986;AnchorMart Scheme means the contract entered into by Mighty River with AnchorMart Limited – New Zealand Dairy Group's merchandising arm – to supply electricity to NZDG suppliers and sharemilkers;BOPE means Bay of Plenty Electricity Limited;Caygill Report means the June 2000 report of an inquiry into the electricity industry commissioned by the Government in February 2000 and chaired by Dr D Caygill;Consumers means small commercial and domestic electricity consumers who consume less than approximately 100,000 kW hours per annum;Contact means Contact Energy Limited;Eastbay means Eastbay Metering Limited;Eastern Bay means the Eastern Bay of Plenty;Eastern Bay Electricity Metering Services Market means the market identified by the Commission for non-time of use electricity metering services in respect of Horizon Consumers;Eastern Bay Retail Electricity Market means the market identified by the Commission for the retail supply of electricity to Horizon Consumers;ECNZ means Electricity Corporation of New Zealand Limited;EGRs means the Electricity Governance Rules enacted as regulations pursuant to s 172H of the Electricity Act 1992;EIRA means the Electricity Industry Reform Act 1998;EOL means Energy Online Limited;Empower means Empower Limited;ESA means an Electricity Supply Authority;Genesis means Genesis Energy Limited;Horizon means Horizon Energy Distribution Limited;Horizon Area means the geographic dimensions of the electricity markets as defined by the configuration of the Horizon Network in the Eastern Bay;Horizon Consumers means Consumers who are connected to the Horizon Network;Horizon Network refers to the local electricity lines network in the Eastern Bay that Horizon retained ownership of and continued to operate after its initial transaction with BOPE;ICP means installation control point;KCE means King Country Energy Limited;The Lines Co means The Lines Company Limited;MARIA refers to the Metering and Reconciliation Information Agreement that was established by the electricity industry as a metering standard and information sharing protocol;Meridian means Meridian Energy Limited;Mighty River means Mighty River Power Limited;NGC means the Natural Gas Corporation Limited;NZED means the New Zealand Electricity Department;NZEM means the electricity industry established by the New Zealand Electricity Market on a contractual basis;Pacific means Pacific Hydro Limited;Todd means Todd Energy Limited;TrustPower means TrustPower Limited; andVector means Vector Limited.The pleadingsThe Commission's central assertion[11] In its statement of claim, the Commission bases its case on an assertion regarding the economic significance for potential competitors of BOPE's unwillingness, in effect, to lease its meters to those competitors. [12] In making that assertion (the Commission's "central assertion"), the Commission first identifies what it defines as electricity metering services. Such services, in the Commission's view: (a) Comprise the provision of non time of use electricity metering equipment situated at the premises or residences of consumers and associated services needed to maintain that equipment in a fully functional state; and (b) In addition, may comprise the service of the reading of meters and provision to an electricity retailer of the information obtained from the reading of meters. [13] The Commission then asserts that a person wishing to supply retail electricity in competition with BOPE to small commercial and domestic electricity consumers(those consuming less than approximately 100,000 kW hours per annum) ("Consumers") connected to the Horizon Network ("Horizon Consumers") must either: (i) Enter into an agreement with BOPE for the provision by BOPE to it of electricity metering services in respect of Horizon Consumers; or (ii) Purchase from BOPE the meters and ripple relays installed at the premises or residences of Horizon Consumers; or (iii) Install separate meters and ripple relays at the premises or residences of Horizon Consumers. [14] The Commission further asserts that, in the absence of an agreement with BOPE for the provision by BOPE of electricity metering services, it would not be economically viable or practicable, at prevailing retail prices, for a person other than BOPE to supply or seek to supply retail electricity to Horizon Consumers. The second and third alternatives – purchase from BOPE or installation – are therefore, and by necessary inference, in the Commission's view not economically viable or practicable at prevailing retail prices. There is, the Commission says, no close substitute for entering into an agreement with BOPE for the provision by BOPE to such a person of electricity metering services.The s 36 case[15] The Commission's s 36 case, as pleaded, can be summarised as follows: (a) There are two relevant markets: (i) A market ("the Eastern Bay Retail Electricity Market") for the supply of retail electricity to Horizon Consumers; and (ii) A market ("the Eastern Bay Electricity Metering Services Market") for non-time of use electricity metering services in respect of Horizon Consumers.(b) Alternatively, if there is no separate Eastern Bay Electricity Metering Services Market, then the functional dimensions of the Eastern Bay Retail Electricity Market include the provision of non-time of use electricity metering services to Consumers. (c) The geographic dimensions of these markets is the area defined by the configuration of the Horizon Network in the Eastern Bay ("the Horizon Area"). (d) BOPE was dominant/has substantial market power in the Eastern Bay Electricity Metering Services Market, or, in the alternative, the Eastern Bay Retail Electricity Market. (e) BOPE used that dominance/took advantage of that substantial market power by failing or refusing to enter into electricity metering services agreements with a number of specifically identified firms who had entered or wished to enter the Eastern Bay Retail Electricity Market in competition with BOPE. (f) BOPE did so for the substantial purpose of restricting the entry by those persons into, or preventing or deterring competitive conduct by those persons in, the Eastern Bay Retail Electricity Market. [16] As primarily pleaded this is, therefore, an assertion of the use of dominance/taking advantage of substantial market power in one market (the Eastern Bay Electricity Metering Services Market) for the purpose of affecting competition in another market (the Eastern Bay Retail Electricity Market). The Commission only pleads use of dominance/taking advantage of substantial market power as regards the Eastern Bay Retail Electricity Market in the alternative, and then as regards the functional dimensions of that market relating to non-time of use meters. Therefore: (i) There is no broader assertion of BOPE being dominant or having substantial market power in the Eastern Bay Retail Electricity Market generally, for example as regards the supply of electricity itself; and(ii) The particulars of dominance/substantial market power in both markets relate as pleaded to meters. [17] However, if we agree that the relevant market for use of dominance/taking advantage of substantial market power purposes is the Eastern Bay Electricity Metering Services Market, we must still uphold the existence of the pleaded Eastern Bay Retail Electricity Market. This is because it is in that market that BOPE's behaviour is alleged to have been directed, and in respect of which relief is sought.Section 36 – particulars[18] BOPE is alleged to have been dominant, and equally to have a substantial degree of market power, in the Eastern Bay Electricity Metering Services Market or, alternatively, the Eastern Bay Retail Electricity Market, by reason of the following factors: (a) BOPE's position as the incumbent electricity retailer in the Eastern Bay, owning almost 100% of the non-time of use electricity metering equipment and as a consequence having nearly a 100% share of the Eastern Bay Electricity Metering Services Market. (b) The absence of any competitor or potential competitor in the Eastern Bay Electricity Metering Services Market. (c) The adoption and following by BOPE of a general practice of not entering into meter leasing agreements for time of use meters. We note that: (i) The reference to time of use meters should clearly be a reference to non-time of use meters, and we have understood it as such. (ii) The phrase "meter leasing agreements" is not separately defined. Further, the Commission does not, at this point, use its defined phrase "agreements for electricity metering services". The Commission defines electricity metering services as including theprovision of non-time of use meters at Consumers' premises. A practice of not entering into meter leasing agreements for non- time of use meters, in other words not providing those meters to competitors for them to use, can therefore be seen as a subset of a practice of not entering into agreements for electricity metering services. (d) The existence of significant barriers to entry to the Eastern Bay Electricity Metering Services Market in the form of: (i) BOPE's obstructive behaviour as regards new entrants' installation of non-time of use meters; (ii) Consumer resistance to installation of new meters, including as influenced by BOPE's obstructive behaviour; (iii) The economies of scale arising from "the number and geographical propinquity" of BOPE's meters and ripple relays; (iv) The non-recoverable market specific costs associated with the installation of new meters; (v) The requirement that when non-time of use meters are removed and replaced that the replacement meters be MARIA compliant; and (vi) The actual or perceived likelihood of higher meter reading costs and maintenance costs for new entrants. [19] The Commission then provides extensive particulars of use/taking advantage. These focus on BOPE's failure or refusal to enter into electricity metering services agreements with a number of retailers.Section 36 - relief[20] By way of relief the Commission first seeks a declaration that "the conduct of BOPE in refusing access to its meters as pleaded" was a contravention of s 36. In our view, the reference to "the conduct of BOPE in refusing access to its meters as pleaded" is a reference to BOPE's conduct in failing or refusing to enter into electricity metering services agreements with the named retailers, and more particularly its general practice of not entering into meter leasing agreements. It was that conduct which was explicitly identified by the Commission, in terms of the elements of s 36, as constituting the use of BOPE's dominant position and the taking advantage of its substantial degree of market power. [21] The Commission also identified certain other behaviour by BOPE as giving rise to barriers to entry, including in particular BOPE's behaviour in obstructing steps taken by intending new entrants to install non-time of use meters, the impact of that obstructive behaviour in increasing customer resistance to having new meters installed and BOPE's policy that it would only sell MARIA-compliant meters. Such behaviour was not pleaded as constituting use, or taking advantage, and is therefore not the target of the relief sought. [22] The Commission also seeks, if a declaration is granted and BOPE continues to act in breach of s 36, mandatory or restraining orders to "facilitate the provision of electricity metering services", and a pecuniary penalty and costs.Section 36 - BOPE's position[23] The core of BOPE's defence to the Commission's s 36 claims comprises denials of: (a) the Commission's characterisation and definition of electricity metering services; (b) the Commission's tripartite classification of the metering choices available to persons wishing to compete with BOPE for the sale of electricity to Consumers; and(c) the Commission's central assertion. [24] Rather, BOPE asserts that a person wishing to earn Revenue (as defined by BOPE) in competition with BOPE from the supply of retail electricity to Consumers could obtain meters and associated metering services (for example reading and maintenance), or the Data traditionally provided by meters, in a number of ways including by procuring services, procuring data or procuring meters. [25] BOPE further asserts that a range of firms have in the past competed successfully with BOPE in the Horizon Area and supplied Horizon Consumers with electricity, and are currently doing so. It refers in this context to all of the firms listed by the Commission in its pleading. [26] On the basis of those core pleadings BOPE: (a) denies the existence of any local or regional markets for the supply of electricity; (b) denies the existence of a market for electricity metering services, but says if it exists it is a national market; (c) denies BOPE's possession of dominance, or a substantial degree of market power, pointing to the historic and current competitive activity already identified. [27] In responding to the Commission's allegation as to use/taking advantage, BOPE acknowledges that, from 1999 to date, it generally did not enter into meter leasing agreements, and had a policy of not doing so. It asserts, however, that it had a number of business reasons for deciding on and implementing that policy. These relate particularly to the fact that it did not have, and saw no reason for or benefit to it in establishing, a meter leasing business. It says competitors had a number of economically viable and practicable alternatives open to them as regards meters, including as offered by BOPE. As to the Commission's pleaded particulars, it generally acknowledges core factual assertions, but asserts valid business reasons for those actions and an absence of any anti-competitive purpose or effect.The s 27 case[28] The Commission's s 27 causes of action first rely on the pleading as to: (a) the necessity for firms seeking to compete with BOPE for the supply of electricity to Consumers to enter into electricity metering services agreements with BOPE; and (b) the particulars of the ways in which retailers sought to enter into electricity metering services agreements with BOPE, and BOPE failed or refused to do so. [29] The Commerce Commission then pleads two causes of action under s 27(2) (the third and fourth causes of action). The third cause of action alleges that on a date unknown to the Commission a number of parties (including BOPE, Pacific Hydro BOPE Ltd, Todd BOPE Ltd, Todd Energy Ltd, and various officers of those companies) entered into an arrangement or arrived at an understanding, a provision of which was that BOPE would refuse to lease its meters to potential entrants into its incumbency. It is alleged that that provision had as one substantial purpose the purpose of substantially lessening competition in the Eastern Bay Retail Electricity Market by preventing or hindering the entry of potential competitors. The Commission also alleges that, irrespective of whether BOPE was a party to that arrangement or understanding, it gave effect to it by its conduct. [30] The fourth cause of action repeats the allegation that BOPE entered into an arrangement or arrived at an understanding, a provision of which was that it would refuse to lease its meters to potential competitors. The fourth cause of action alleges that that provision had the effect, or was likely to have the effect, of substantially lessening competition in the Eastern Bay Retail Electricity Market by preventing or hindering competitive entry. Again, the Commission alleges that at various times BOPE gave effect to that arrangement. [31] Under the third cause of action, the Commission seeks a declaration that BOPE breached s 27(2) by giving effect to the arrangement or understanding for the purpose of substantially lessening competition in the Eastern Bay Retail ElectricityMarket. Under the fourth cause of action, the Commission seeks a declaration that BOPE breached s 27(2) by giving effect to the arrangement or understanding, which had the effect or the likely effect of substantially lessening competition in the Eastern Bay Retail Electricity Market.Section 27 – BOPE's position[32] In reply to the third and fourth causes of action, BOPE denies the allegation that it entered into an arrangement or understanding the purpose of which was to substantially lessen competition. BOPE admits that from 1999 to the present date it generally did not enter into meter leasing agreements and had a policy of not doing so, but it states that was because it had not acquired a meter leasing business along with the electricity retail business, it did not wish to incur the cost of establishing such a business, there was no need for it to do so and there was no benefit in doing so. [33] BOPE further points out, as it has done in relation to the s 36 causes of action, that there were a number of ways for potential competitors to enter the market other than through meter leasing. The viability of these options is demonstrated, says BOPE, by the fact that a number of electricity retailers did enter its incumbency during the period complained of. Accordingly, BOPE denies that the policy of not leasing meters had the effect of substantially lessening competition or was likely to have such an effect.Overview of the evidence[34] The Commission's evidence can be seen as falling into four broad categories: (a) Background information; (b) Evidence from competitors and others as to their dealings with BOPE; (c) Evidence from consumers who had wanted to deal with BOPE's competitors; and(d) Expert evidence. [35] Mr Thorn, the Commission's general manager, and Mr Heaps, an independent consultant with extensive industry experience, provided background information. [36] Mr Thorn gave evidence as to: (a) The Commission's experience over time with the electricity industry and how, in the context of the industry reform process that has been underway since the late 1980s, the Commission and the Courts have considered a number of transactions involving the consolidation of ownership within the industry; (b) The background to these proceedings; and (c) The source of the Commission's understanding of industry practices as regards agreements between industry participants involving metering arrangements at the level of small business and retail consumers, and hence its information in the statement of claim relating to electricity metering services. [37] Mr Heaps provided a very helpful introduction to the electricity industry, and to the reform process that has been underway since the late 1980s. He focussed, in particular, on issues relating to the introduction of competition to the retail market. [38] Documentary evidence was provided pursuant to Rule 441 as regards seven retailers overall. These were Mighty River Power Limited ("Mighty River"), Genesis Energy Limited ("Genesis"), TrustPower Limited ("TrustPower"), Meridian Energy Limited ("Meridian"), Energy Online Limited ("EOL"), Empower Limited ("Empower") and Contact Energy Limited ("Contact"). [39] Evidence was provided by four witnesses who, as executives of electricity retailers, had had experience with BOPE as their firms tried to compete in BOPE's incumbency:(a) Mr Reilly, of Mighty River; (b) Mr Pickup, of Genesis; (c) Mr Harnett, of TrustPower; and (d) Mr Treadwell, of Meridian. [40] Mr Barnett, from The Lines Company ("The Lines Co"), and Messrs Congalton and Dennehy of Natural Gas Corporation ("NGC"), provided experience from the perspective of firms which supply electricity metering services. [41] The Commission also called evidence from five retail consumers as to their experience in endeavouring to switch their retail electricity supplier from BOPE to a competitor. Their evidence focused on the difficulties they had encountered, with particular reference to the inconvenience of having to have new meters installed. This evidence was provided in written form by way of witness statements, and by consent of the defendant was not required to be read. There was, therefore, no challenge to the veracity of this evidence. [42] Finally, the Commission called expert evidence. [43] Dr Pleatsikas, an American economist, gave evidence particularly as regards issues of market definition, the possession by BOPE of a substantial degree of market power and BOPE's use of that market power. His evidence was significant for the emphasis it placed on matters the Commission had identified as constituting barriers to entry, rather than BOPE's policy of not entering into meter leasing agreements in and of itself. [44] Mr Taylor, an independent financial consultant, gave evidence as regards BOPE's possession of a substantial degree of market power. He also provided a range of evidence regarding metering services, including as influenced by scale, and regarding the relative costs of providing such services by way of the lease, sale and/or purchase of meters.[45] Mr Westergaard, also an independent consultant, provided evidence on market share, "churn" rates, the cost of interruptions of electricity supply and average wholesale electricity prices both in the BOPE area and for New Zealand as a whole. [46] For BOPE, three executives, Mr Tweedie (Todd's managing director), Mr Bahirathan (Todd's Group Manager Electricity) and Mr Doidge (BOPE's former Business Developer Manager) provided evidence. Their evidence covered a wide range of issues but – consistent with BOPE's acknowledgement of its decision not to offer meter leasing arrangements to competing retailers – focussed principally on BOPE's perspective on that decision and on its interactions with retailers throughout the relevant period. The principal theme of that evidence, taken overall, was that BOPE's decisions were commercially rational ones made in its own business interests, and were not ones which in any material way hindered or prevented those retailers from competing with BOPE. Retailers had a variety of feasible alternatives for acquiring access to meters and metering data to compete with BOPE for Horizon Consumers. Furthermore, BOPE was at all times exposed to actual and potential competition in what it regarded as national markets. Even if local markets could be identified, BOPE was not dominant within them. [47] BOPE's evidence also addressed, from BOPE's perspective, the extent to which retailers had competed with it for the supply of electricity to Horizon Consumers, notwithstanding BOPE's admitted policy of not offering to enter meter leasing agreements. [48] BOPE also called evidence from the owner of a local metering services firm. Mr Crosbie, a director and the manager of Eastbay Metering Limited ("Eastbay"), gave evidence of the way in which his firm provided labour-only metering services to retailers in the BOPE area and on a range of issues relating to metering. [49] BOPE's expert evidence was provided by: (a) Professor Sweeney, an American economist.(b) Mr Murray, a New Zealand economist and electricity industry specialist. [50] Professor Sweeney's expert evidence focused on the proposition that the Commission's case against BOPE involved tilting an otherwise level playing field in favour of BOPE's competitors. He supported that argument by, in general terms, contesting the Commission's economic analysis as reflected in its statement of claim. [51] Mr Murray provided extensive material on industry context for this case. His position was that the relevant markets were national ones, and further there was no separate electricity metering services market. He also provided extensive evidence on electricity retail prices, margins and metering costs.Industry Background[52] In this section we outline our understanding, on the basis of the evidence presented to us – particularly that of Mr Heaps and Mr Murray – of the context for this case provided by the reform of the electricity industry in New Zealand. In our view this context is important, particularly for a proper understanding of market issues.Electricity industry reforms[53] Since the mid 1980's the electricity industry in New Zealand, as in a number of other countries, has been subject to an ongoing process of reform. [54] Traditionally central government, through the New Zealand Electricity Department ("NZED"), owned and operated the infrastructure whereby electricity was generated and transmitted nationally. Local government, operating through municipal electricity departments and power boards as electricity supply authorities ("ESAs"), owned the local distribution networks. ESAs purchased electricity from NZED and sold and distributed that electricity to customers connected to their localdistribution networks. Effectively, NZED had a monopoly at the national level whilst each ESA had a monopoly in its particular region. [55] Until the 1980s the monopoly status of NZED and the ESAs as regards the supply of electricity was thought to be a structural necessity. Very simply, this was because electricity generators inject their electricity into the national grid and all generated electrical energy flows through this into distribution networks and on to customers. As a result it is impossible to identify which electricity comes from which power station and therefore which power station supplies which customers. [56] The reform process was aimed at – amongst other things – the creation of competitive markets for electricity. Key steps in this process included: (a) The establishment, in 1987, of the Electricity Corporation of New Zealand Limited ("ECNZ") as an SOE in place of NZED. (b) Further restructuring, in 1993/94, whereby: (i) Transmission (the national grid) was separated from ECNZ with the formation of TransPower. (ii) ESAs were replaced by energy companies. (iii) All customers were, at least in theory, made contestable. (c) The establishment, in 1996, of the New Zealand Electricity Market ("NZEM") governance structure and rules which enabled the wholesale market to operate and generators to compete at the wholesale level. (d) The enactment, in 1998, of EIRA, and the consequent split by 1 April 1999 of energy companies into lines and energy businesses. (e) The split, by 1999, of ECNZ into four generators (Contact (1996) and Meridian, Mighty River and Genesis (1999)).[57] During the same period developments in electronic data management systems and techniques facilitated the allocation of quantities of electricity to buyers and sellers, making competition between sellers for buyers possible. In New Zealand, these matters were generally the responsibility of industry participants, acting in response to expectations set out in Government policy statements which provided for regulation if Government's expectations were not met. Voluntary, contractual, arrangements between industry participants were relied on in many areas. Key developments included: (a) Competition for medium and large customers was made possible by late 1994. This involved: (i) The establishment, by the industry, of a metering standard and information sharing protocol known as the Metering and Reconciliation Information Agreement ("MARIA"). (ii) The installation of time of use meters 1 which enabled more accurate recording of electricity consumption by those customers. (b) Competition for domestic and smaller commercial customers, for whom the installation of time of use meters was not economic, required a number of further developments. These were: (i) The establishment, in April 1999 under MARIA, of a central registry of Consumers enabling domestic and smaller commercial customers to be tracked and linked to retailers and metering data1 There are two types of meters: time of use meters and non-time of use meters. Time of use meters are more sophisticated and expensive than non-time of use meters, and record electricity consumed by reference to successive half hourly periods. The investment in such meters, at least as relevant for this case, can only be justified for larger customers. As electricity use is, nationally, measured by reference to half hour periods, time of use meters enable the electricity used by larger customers to be recorded and reconciled with electricity generated and purchased in the wholesale market in a reasonably straight-forward way. The meters at issue in this case are not time of use meters. Non- time of use meters are the meters that we are familiar with from our own residential experience. They record use, without any reference to the point of time at which, or the period of time during which, particular use occurs. These meters are read from time to time and aggregate use figures – for the period from one meter reading to the next – are thereby obtained. This case concerns non-time of use meters.over time, thus enabling a customer to "switch" from one retailer to another. (ii) The introduction, also in April 1999, of profiling. Profiling allowed monthly quantities metered at the premises of domestic and smaller commercial customers to be broken down into half hour quantities, and in that way factored into the reconciliation process. [58] Metering standards were specified in MARIA. Classes of metering were introduced based on the meters' limits of accuracy. The higher the quantities to be measured, the higher the accuracy level required. The MARIA rules required that all meters would eventually become more accurate and compliant with the standards set in the rules. [59] Compliance with those metering standards was, in effect, required from the outset as regards time of use meters. [60] In the case of non-time of use meters it was decided that, rather than require their costly replacement immediately, a long period would be allowed to fit in with reasonable meter replacement cycles. The MARIA rules had a requirement for all non-time of use meters to be compliant by 2015. Until then, installed meters would be deemed compliant. Any new non-time of use meters, however, had to be certified as compliant. [61] As noted, the rules to enable domestic and small commercial customers to choose between competing retailers, that is to switch, involved the creation of a central registry of retail customers. That registry, by reference to discrete installation control points (ICPs) each of which was accorded a unique number, recorded which customers were supplied by which retailers. At the same time, procedures were introduced whereby the registry would be amended to reflect the fact that one retailer had "won" a customer and another had "lost" a customer. These included providing an "effective date" for a customer switch, and for a meter reading as at that date to determine liability for electricity supplied by the "old" supplier up to that date, andby the "new" supplier on and from that date. Data exchange procedures were established for those purposes. [62] The rules introduced on 1 April 1999 were not as effective as had been hoped. Further modifications were brought into force in April 2000. [63] The Commerce Commission has, in a series of decisions, reflected on the significance of these developments for its approach to market definition issues. [64] By late 1994 the combined effect of: (a) the formation of energy companies in place of the electricity supply authorities, and subsequent consolidation amongst those companies; (b) the restructuring of ECNZ to create TransPower and Contact Energy; and (c) the decision to allow for contestability of all consumers, had created – in the eyes of the Commission – a national market for retail supply to medium to large customers, based on time of use metering. Local retail markets, co- extensive with the various network boundaries of the energy companies, remained for domestic and smaller commercial customers where time of use metering was not economic. The Courts upheld this view in litigation: Power NZ Ltd v Mercury Energy Ltd [1996] 1 NZLR 686 (HC) and [1997] 2 NZLR 669 (CA). [65] The Commission confirmed those views in two subsequent decisions, Number 302 in July 1997 – giving clearance for Powerco Ltd to acquire Egmont Electricity Ltd – and Number 317 in February 1998. [66] In late 1998, the Commission was called upon to determine a clearance application by Contact to acquire the assets which comprised the gas retailing business of Enerco NZ Limited. Contact argued that the then recent reforms, or the reform processes then underway, provided reason for the Commission to reassess relevant market definitions. The Commission summarised that argument as follows:Contact argues that recent Government electricity industry reform provides reasons to reassess relevant market definitions in respect of small electricity consumers, principally:• the separation of power companies' lines businesses and supply businesses (electricity retailing and generation) through the Electricity Industry Reform Act 1998; and• the Government's expectation that electricity industry participants will establish, by April 1999, deemed profiling to enable small electricity consumers to switch retail suppliers, or else face the Government introducing a mandatory 'default' system by regulation. (at [37])[67] The Commission agreed, reaching the following conclusion at para [40] of Decision 333 on 10 December 1998:The Commission has considered the changes which have occurred, and are continuing to occur, in respect of electricity retailing, and their implications for defining the relevant markets. The Commission is satisfied that there is clear evidence of electricity suppliers being able to switch supplies between different categories of consumers, including small consumers, depending on market opportunities. Suppliers do not appear constrained to supplying limited geographical areas or to supplying to consumers on particular networks only. Small consumers now have, or will have in the near future, a choice of suppliers. This situation increasingly matches that of larger consumers. Therefore, the Commission concludes that it is no longer appropriate to define discrete markets for the supply of delivered electricityto small consumers and to medium and large consumers. Accordingly, in assessing the Application, the Commission concludes that the relevant market in respect of electricity retailing is the nationalelectricity retail market. (supra, para [41]) .[68] Therefore, in 1999 when the businesses of the energy companies were split, and new owners became responsible for running the electricity supply businesses, competition for domestic and smaller commercial customers had – at least in theory – been enabled. [69] The businesses which acquired the regional electricity supply businesses were typically the large generators, thereby providing those generators with direct access to retail customer bases. In the Commission's view, those generator/retailers would be able to compete with each other for Consumers in a national market.[70] In 2004, the Government moved from the self-regulating industry structure and introduced legislation that brought in an industry regulator, the Electricity Commission. Both MARIA and the NZEM rules were combined, with other components such as transmission issues, to form the Electricity Governance Rules ("EGRs"). The EGRs were enacted as regulations, giving their requirements the force of law.Meters and switching[71] Central to this case is the question of the approach taken by BOPE to requests by retail competitors for access to its meters. [72] BOPE's decision, as the incumbent retailer 2 in the Horizon Network, and in contra-distinction as we understand it to all other incumbent retailers, was not to enter into reciprocal meter leasing agreements and (generally) not to lease its meters to actual or potential competitors. It is therefore necessary to understand the practical significance in the retail electricity market of the issue of access to meters. [73] It is not necessary to go in great detail into the procedures provided by MARIA for measuring electricity use and for switching customers. Put very simply, and as they affect issues in this case, those rules relied, both as to switching and for measuring and reconciling electricity use on an ongoing basis, on retailers having access to data provided by non-time of use meters. At the same time, the MARIA rules themselves did not regulate that issue of access to meters. That is, obtaining access to the data recorded by non-time of use meters raised issues of the ownership of, and access to, those meters that were not addressed by MARIA. [74] Therefore, retailers needed to contract with each other as regards that matter. [75] Most retailers were able, in a fairly short period of time after the introduction of retail competition, to agree with each other contracts whereby meters installed by2 The term "incumbent retailer", with reference to a particular local network owned by a lines company, refers to the firm which acquired from that lines company the business of supplying electricity within that network. On 1 April 1999 that incumbent firm would have been supplying all consumers within that network. The terms incumbent area or "incumbency", with respect to a particular retailer, are used accordingly.one retailer at a customer's premises – typically in the early days by the historic incumbent retailer – were able to be used by new entrant retailers who, in the newly competitive market, "won" customers. A fairly standard form of contract evolved, conceptually based on the "incumbent" retailer leasing its meters to its competitors. Whilst taken together meters were assets of some value for retailers, individually they were not of great value and the amounts involved to provide for an acceptable commercial return were such that the contracts did not require – no doubt after an initial settling down period – a great deal of commercial negotiation. [76] These contracts were generally arranged on a reciprocal basis between retailers. That is, retailer A would contract with retailer B so that each granted the other access to their meters on equivalent terms. [77] This behaviour can be seen as following in line with Government expectations as to the industry's role in facilitating retail competition. Those expectations were reinforced throughout the industry by the possibility of regulation. [78] The generator/retailers may also have taken the view, because these agreements were being entered into on a reciprocal basis and because it would appear – from the evidence provided to us – that most generator/retailers took a similar approach to competition outside their incumbent areas, that the risk of losing customers to each other was also "reciprocal". That is, their net exposure to loss of customers, and hence their net exposure to each other under these reciprocal arrangements, would not be overly significant. [79] New entrant retailers, who themselves were not in a position to offer access to meters, could contract bilaterally but on similar terms.BOPE acquires its energy business[80] Todd and Pacific completed the acquisition of the BOPE business, as required by EIRA, on 31 March 1999 so that on and from 1 April 1999 BOPE owned and operated the BOPE business on their behalf. Todd and Pacific decided to contract out responsibility for undertaking most of BOPE's operational functions to King Country Energy Limited ("KCE"), a company in which Todd had been aninvestor for some time. KCE was to provide all BOPE's billing, call centre, general administration, general management and payroll functions. KCE's general manager, Mr Peter Till, became BOPE's general manager – albeit based in Taumaranui. BOPE's commercial manager, Mr David Bulley, who had been the commercial manager at the local energy company, was to report to Mr Till. [81] It would appear that when Todd and Pacific first acquired the BOPE business, BOPE – at least to some extent and for some time – may have acted as their agent and not itself directly owned that business. The precise position was never clarified. The case was pleaded by the Commission, and responded to by BOPE, on the basis that BOPE itself acquired and operated that business. It was not suggested by BOPE that the actual position, to the extent different, was material and we have therefore proceeded on the basis that BOPE, initially owned by Todd and Pacific as 50:50 joint venturers, through wholly owned subsidiaries, owned and operated the BOPE business at all material times. [82] There have been a number of changes to the structure whereby Todd subsidiaries hold shares in BOPE, including one amalgamation. Nothing in this case turns on those details. On 15 August 2001 Pacific sold its interest in BOPE to Todd, and since then BOPE has been wholly owned by Todd.1999-2000 – the market reality[83] In the period leading up to 1 April 1999 each of the three new SOEs – Meridian, Mighty River and Genesis – together with Contact and the other large generator TrustPower, were actively seeking to buy electricity supply businesses from local energy companies. Each of those firms acquired a number of such businesses, giving them direct access to retail electricity customer bases, a natural hedge for their generation businesses. [84] Having acquired those businesses, those firms then all looked to consolidate and defend their own customer bases, and to expand and compete beyond their own areas in a variety of ways. Issues relating to procedures for customer transfer, and access to meters, were therefore matters of general concern within the electricityindustry at that time. Notwithstanding the Government's expectations as to the arrangements that were to be in place by 1 April 1999 to facilitate such retail competition, and to make it easier for retail customers to switch between suppliers, considerable problems remained. [85] BOPE placed considerable emphasis on these issues. In Mr Tweedie's view, the period 1999 – 2000 was one of huge instability and uncertainty in the industry, which caused a range of difficulties for retailers and consumers. These were reflected in poor billing practices, including unwarranted disconnections, and in generator/retailers competing for consumers without themselves having put in place the necessary support systems and arrangements. These were matters which influenced BOPE's decision as to the approach it was to take when competitors asked for access to its meters. [86] A number of the witnesses called by the Commission confirmed that, in that early period in 1999 and into 2000, the new market did not operate smoothly. [87] Mr Thorn acknowledged that there were genuine concerns in a range of areas. Quite apart from the Commission's concerns as to what was going on in the BOPE area, he agreed that there was widespread public concern with the problems involved in switching from one electricity supplier to another, that there were large numbers of complaints about delays, about the ability for customers to switch easily from one retailer to another and about interrupted billing cycles. These were all things that were occurring nationally. [88] Each of Messrs Heaps, Reilly, Pickup and Harnett also acknowledged these difficulties. Mr Reilly, in fact, spoke of the "chaos" that existed as the newly integrated generator/retailer businesses struggled to familiarise themselves with the retail businesses they had acquired, often with inexperienced staff, whilst at the same time embarking on marketing initiatives to acquire new customers. As we understood the evidence, these problems related particularly to retail customer databases and associated software. The large generator/retailers were struggling to integrate the databases they had inherited when they had acquired their various businesses. This resulted in firms undertaking marketing campaigns without having all relevant customer support procedures in place. At the same time, individual firmand industry switching arrangements were not working well. Mr Pickup commented that switching problems and delays were exacerbated by the fact that, in his view, virtually no retailer had designed or built systems for losing customers, that is enabling customers to be switched away. [89] The existence of these problems was confirmed by the June 2000 report ("the Caygill Report") 3 of an inquiry into the electricity industry which had been commissioned by the new Government in February 2000. In a passage referred to by Mr Tweedie, the Caygill Report spoke of the difficulties associated with the entry of the generator retailers into the market in 1999:With the lines/energy split, a number of new retail businesses entered the market. In particular, the retail bases were attractive to generators and a vigorous fight for market share took place. These companies appear in a number of cases to have entered the market with excessive haste. New entrants were often not adequately prepared and in many cases showed a lack of planning and foresight. As a result, consumers suffered. (Caygill Report, CB 6/1500, at para 209) Some retailers did not have, at least initially, the necessary administration and computer systems in place to provide the level of service expected. In part, this reflects the tight deadline by which retailers needed to prepare for competition (once it became practical through profiling). Some retailers may have considered they had no option but to compete, whether they were ready to do so or not. (Caygill Report, supra, at para 210)[90] Switching issues were also commented on by the Caygill Report:MARIA first implemented a switching protocol in April 1999. The protocol provided rules to be followed by retailers and distribution companies when switching customers from one company to another. This protocol was not robust enough to prevent the delays and other problems experienced and was difficult to enforce. A new switching protocol, introduced in April this year, together with the bedding down of retailers' systems, is intended to resolve the problems previously experienced in the retail sector. (Caygill Report, supra, para 212)[91] As noted in the Caygill Report, in 2000 efforts were made at an industry level to address some of these difficulties.3 The Commission did, at one point, object to the introduction of the Caygill Report in evidence, to the extent that report was relied on as to the truth of its contents, as opposed to its contents constituting a record of the views of its authors. As the case progressed, we think it is fair to say that both parties referred to the Caygill Report as a useful analysis of, and commentary on, aspects of the electricity industry relevant to this case.[92] Significant developments to the MARIA rules came into force in April. A MARIA Governance Board paper of 11 April 2000 provided the following explanation:Since the implementation of Customer Switching on the 1 April 1999, the rules that were originally incorporated within MARIA have not delivered the desired benefit that the MARIA Governance Board had originally envisaged. The proposed rule change rectifies these difficulties by removing the existing ambiguities in the current process. (Supplementary Common Bundle, 48-52, at 50).[93] Mr Heaps, in cross-examination by Mr Farmer, acknowledged that the new procedures provided a detailed procedure for the switching of a customer from an old retailer to a new retailer and that under the new rules it was not possible for the old retailer to prevent the switch to the new retailer taking place. [94] Towards the end of 2002 the Commission again reviewed the structure of the retail electricity industry. Genesis applied to acquire EOL's electricity retailing business. In decision 476 dated 10 October 2002 the Commission concluded once more that the relevant market was the national market for electricity retailing. In the course of reaching that conclusion, however, the Commission identified certain regions throughout the country which appeared to have a limited number of active electricity retailers. Those areas were typically rural areas with small numbers of customers. [95] The Commission expressly singled out difficulties with access to meters in the Bay of Plenty region:In the Bay of Plenty, Bay of Plenty Electricity is the only electricity retailer. Other retailers have previously sold electricity in this region but have withdrawn, owing to their inability to gain access to electricity meters. The Commission has also identified an area where there are specific barriers to entry that have deferred or prevented competition. (Decision 476, paras 56 and 57)[96] The Commission had, of course, by that time commenced these proceedings, having filed its original statement of claim on 21 December 2001.BOPE's approach to competitors' requests for access to its meters[97] Central to the Commission's allegations against BOPE are dealings between BOPE and a number of electricity retailers seeking access to BOPE's meters to enable them to supply electricity to Consumers in competition with BOPE. We turn now to those dealings. We review, in turn, BOPE's dealing with each of the firms who approached it on the question of access to its meters in conjunction with their wish to compete with BOPE for the sale of electricity to Consumers. We also review evidence provided by individual Consumers. Finally, we review certain evidence provided by BOPE, principally relating to its perspective on the extent to which retailers have competed with it for the supply of electricity to Consumers notwithstanding its policy of not leasing its meters.Individual retailersMighty River[98] Mighty River first contacted BOPE on the meter access issue in June 1999. [99] Mighty River had that month entered into a contract with AnchorMart limited – New Zealand Dairy Group's merchandising arm – to supply electricity to NZDG suppliers and sharemilkers ("the AnchorMart Scheme"). Mighty River's commercial strategy was to cement its position in the Waikato, rather than being nationally focussed. As a result of the contract, however, Mighty River acquired farming customers throughout the country. A small number of these were BOPE customers. [100] To give effect to the AnchorMart scheme, Mighty River needed access to meters. At the time it entered into the contract, it had not made access arrangements with BOPE, although it did have reciprocal arrangements in place with a number of other retailers. Ultimately, Mighty River was able to negotiate metering agreements with existing retailers in all relevant network areas (some 22 all told) except the Horizon Area.[101] On 25 June Mr Reilly emailed a copy of the generic metering services agreement Mighty River had used with other retailers to Mr Bulley. On 14 July Mr Bulley referred the draft agreement to Messrs Deppe and Bahirathan, noting that the charges did not look unreasonable especially "if we are interested in operating in their region" (CB2/478). Receiving no reply, Mr Reilly followed up with Mr Bulley on 16 July, noting that he knew Mr Bulley had referred the matter to Wellington proposing temporary arrangements and asking if he should deal with someone at Todd. [102] Mr Bulley did not reply to Mr Reilly until 12 August 1999, and then only to say he would be away for a couple of weeks and that he was "not sure how we move this one forward" (CB 2/549). Mr Bulley advised Mr Reilly to contact Mr Till with a formal request, which they would deal with on his return. [103] Mr Reilly emailed Mr Till on 13 August, but received no reply. Internal BOPE emails of 19 October and 17 December from Mr Bulley reflect Mr Reilly having followed up on the meter access issue by phone on several occasions, and record Mr Bulley's understanding that the issue was with Todd (Messrs Bahirathan and Deppe). Mr Bulley at one point proposed adopting the same approach as with TrustPower's request, but limited to "say 50 while the issue is being sorted" (CB 3/714). [104] By January 2000, customers in the BOPE incumbency serviced by 14 ICPs had accepted the Mighty River AnchorMart offer. Switching notices had been sent to BOPE in respect of all those customers. BOPE had not, however, provided the information required to switch the customers. No progress had been made in securing access to BOPE's meters. Mighty River considered that the need to secure access to BOPE's meters had intensified. [105] BOPE was aware of Mighty River's concerns. Papers for an interim JV meeting for December 1999 contain the following comment:Mighty River – agreement tabled. Mighty River are putting considerable pressure to put an agreement in place. They have stated that they need this agreement to supply the customers they obtained through the AnchorMart deal. This impasse is to be reviewed by Richard when all avenues to get customers back has been exhausted. (CB 4/1148)[106] Mr Reilly arranged a meeting for 21 January with Mr Fraser of BOPE to discuss matters, and on 17 January emailed him the generic arrangement. [107] That meeting was cancelled by BOPE. [108] Mr Reilly emailed BOPE on 21 January, expressing his frustration with the lack of progress, and was told he should now contact Mr Tweedie – whom he had unsuccessfully tried to contact by phone on several earlier occasions. Mr Reilly emailed Mr Tweedie on 28 January. [109] Receiving no reply, on 31 January 2000 Mr Reilly requested that the Commerce Commission investigate BOPE's behaviour. [110] Mr Tweedie described these interactions between Mighty River and BOPE as "sporadic communications". Mr Tweedie said that he could not recall phone calls from Mr Reilly, and that during this period Todd was concentrating on understanding the business. It did not have the resources to respond to Mr Reilly in the way he wanted. Mr Tweedie acknowledged that BOPE's failure to respond might have been seen by some competitors as procrastination. [111] When asked why he did not at any point reply to Mighty River and advise them they could put their own meters in, Mr Tweedie said it was not his position to advise Mighty River or any other competitor how they might run their business. They always had the option of taking the steps they ultimately did, and those were to put in their own meters or buy BOPE's. [112] By mid February 2000 Mighty River had internally reached the decision that the only way to honour its AnchorMart commitments was to install its own meters. Notwithstanding that, it continued to seek access to BOPE's meters. An email of 28 March from Mr Bulley to Todd confirms this. [113] On 29 March, Mr Deppe wrote to Mr Reilly, proposing that a short "Interim Agreement" be entered into to allow for negotiations over a meter agreement. The draft proposed Mighty River pay $5,000 to BOPE as a contribution to BOPE's legal fees. Mr Tweedie acknowledged that Todd had not, as far as he was aware, on anyother occasion asked a counter-party for an up-front contribution to its own negotiation costs. Mr Reilly responded on 7 April indicating a willingness – on a slightly revised basis not including the payment of the $5,000 – to proceed. [114] On 2 May, Mr Deppe replied, declining to enter into further negotiations. He pointed to systematic under-reporting by Mighty River of electricity off-take within the BOPE incumbency (Mighty River had a number of large time of use customers in BOPE's area), amounting to some $500,000. [115] Mr Reilly replied in a letter dated 6 June 2000. He explained that Mighty River was taking steps to ensure that the events in question - which were not denied - did not recur, and that any inconvenience to BOPE was regretted. He concluded by asking whether BOPE still intended to enter into a reciprocal metering services agreement as proposed in his letter of 7 April 2000. [116] Mighty River did not receive a reply to that letter. [117] Mighty River therefore decided to go ahead with its plan to replace the BOPE meters. At that time (June 2000) 13 farmers (35 ICPs) had accepted Mighty River's AnchorMart offer. Mighty River advised AnchorMart that it had become necessary to replace the meters. AnchorMart advised it would contact the 13 farmers to see if they were agreeable to their meters being replaced. In late June 2000 AnchorMart advised that three of the 13 farmers had decided to withdraw from the scheme. [118] Mighty River arranged for a contractor to replace the existing meters with new meters. At some stage a further two farmers pulled out of the scheme. Ultimately, eight farmers with a total of 27 ICPs had their BOPE meters replaced in July and August of 2000. [119] The total cost to Mighty River of having these meters (and associated relays) replaced was $13,083 (exclusive of GST). The average cost per ICP was $484.56. [120] Mr Reilly's evidence was that Mighty River only adopted the approach of installing its own meters to meet its contractual requirements. Mr Reilly identified a number of drawbacks with this approach:(a) It was not economically viable in the long term, due to low retail margins. (b) Mighty River faced the risk, if the customer switched back to BOPE, that BOPE might not lease the meter from Mighty River. Although the meter could be re-used, installation costs might have to be written off and retrieval and re-testing costs would be incurred. (c) Meter reading and maintenance costs would be higher, due to the lack of economies of scale. (d) Installing a new meter involved a relatively short (1/2 hour) disruption to power supply. Consumers did not appreciate that, and time was involved in managing issues arising. [121] Mr Reilly did acknowledge in cross-examination that Mighty River, because of its size, might be able to absorb such costs and losses – given the relatively small number of meters involved. [122] Mr Reilly's evidence was that following the AnchorMart experience Mighty River did not whilst he was an employee – that is up until January 2001 – target the BOPE area. He did not explicitly say why that was. In our view, it is not clear that this was solely a result of the AnchorMart experience. Mr Reilly had also said that Mighty River was very selective as to which markets it entered, focussing on its incumbent areas (north of the Harbour Bridge in the Vector network), and in and around the Waikato, where its generation was based. Those factors may also have contributed to Mighty River's decision. [123] Mr Reilly acknowledged he was aware that in mid 2004 Mighty River, under its Mercury brand, had entered the BOPE market targeting Consumers. Mighty River was at that time prepared to install its own meters to do so.Genesis[124] Genesis first contacted BOPE on the meter access issue in February 2000.[125] Genesis wanted access to BOPE's meters to service Consumers acquired pursuant to two national deals it was negotiating: (a) a contract with the Rural Alliance Collective to offer to supply farmer customers of three farming organisations; and (b) a national contract for the supply of electricity to the New Zealand Fire Service. [126] As a result of those two contracts, Genesis acquired new Consumers throughout New Zealand in areas where it was not the incumbent. These included approximately 170 Consumers in the BOPE incumbency. The acquisition of those Consumers was not the result of any dedicated marketing effort by Genesis in the BOPE incumbency area. [127] Mr Pickup dealt initially with Mr Bulley. An email of 17 February 2000 records Mr Pickup's understanding – following a telephone conversation that same day – that BOPE had agreed to Genesis having access to BOPE's meters for customers who switched from BOPE to it. [128] That agreement was subject to the conclusion of a full reciprocal agreement. Mr Pickup proposed an agreement covering meter access arrangements between Genesis, BOPE and Todd/Fresh Start. He provided Mr Bulley with a copy of the agreement that had previously been sent to Todd. [129] Prior to that time, Genesis had had dealings with Todd, whereby Genesis had granted Fresh Start – a joint venture between Todd, KCE and BOPE – access to Genesis' meters. By February 2000: (a) Genesis and Todd had entered into interim access arrangements for Fresh Start pursuant to an agreement of 15 August 1999; and (b) Genesis had on 24 December 1999 sent Todd a formal, more detailed, agreement and on 24 January had given notice terminating the interim agreement.[130] When Genesis contacted BOPE on metering issues it was, therefore, awaiting Todd's response on that agreement. [131] From this point onwards, Genesis' dealings with Todd as regards Fresh Start's access to Genesis' meters, and with BOPE, as regards reciprocal access, overlap. [132] Between March and June 2000 little was achieved. [133] At one point, Mr Bahirathan expressed an interest in a two-way meter access agreement. The next day, he indicated Fresh Start only wanted to deal with Genesis on access to Genesis' meters. Mr Pickup replied, noting that it was Mr Bulley who had suggested the one agreement approach. Genesis wanted to sort out all access issues at once. [134] Around this time, Mr Pickup became aware that BOPE was refusing to process customer switches to Genesis. Throughout March and April Mr Pickup tried to progress the issues with both Messrs Bulley and Bahirathan, but without success. On 26 April 2000, and after he returned from two weeks leave, Mr Reilly sent emails to both gentlemen asking for urgent clarification, and requesting that the issues of meter access and customer switching be finalised as soon as possible. Mr Pickup referred to the unattractiveness for both parties of the alternative, namely installing their own meters. [135] On 26 April 2000 Mr Bulley advised that, subject to a sign-off from Todd's directors, it had been agreed to move forward with the customers' switches from BOPE to Genesis. He said that Genesis should contact BOPE's agents at KCE to arrange for the transfers the following Monday (1 May 2000). By 23 May, however, KCE was still not processing the switches. [136] On 24 May 2000 Mr Bulley said his go ahead was still subject to the sign-off from Todd's directors. The next day, Mr Bulley told Mr Pickup to contact Mr Deppe. Genesis by this stage were, one can only conclude, getting more than a little frustrated. Genesis' CEO, Mr Jackson, became involved.[137] A meeting was organised between Messrs Jackson and Tweedie for 12 June 2000. At that meeting, the question of Genesis' access to BOPE meters was linked with a number of other issues on which Todd, and Mr Tweedie in particular, sought assistance from Genesis. Such assistance was not offered, and the Genesis representatives were told that Todd did not have authority to grant access to BOPE meters, and that Genesis would have to apply in writing to BOPE, which of course they had first done back in February. [138] This was contrary to what Mr Bulley had originally told Genesis, namely that the meter issue would have to be dealt with by Todd. It was also contrary to Mr Bahirathan's apparent willingness to deal with these issues. The position was explained by reference to the fact that Todd could not speak for the BOPE joint venture, but needed to get approval at joint venture meetings from Todd's partner, Pacific Hydro. [139] On 15 June 2000, Genesis did formally write to BOPE asking for access. BOPE replied on 6 July indicating that decision would be taken at the joint venture board meeting to be held on 1 August. [140] Genesis did not want to wait until 1 August and arranged a meeting with Mr Bulley on 12 July 2000. That meeting did not resolve the issue. [141] Finally, on 13 July 2000 Mr Bulley telephoned Mr Pickup, making an offer to sell BOPE meters to Genesis in respect of those customers waiting to be switched to Genesis. This was to be an interim measure, while BOPE resolved its policy issues relating to access to meters. Mr Tweedie's approval would be required. [142] In an email the next day, Mr Pickup advised Mr Bulley that Genesis would accept the purchase proposal, subject to negotiating terms. [143] At the same time, Genesis decided that it needed to confirm with the customers waiting to be switched that they still wished that to occur. [144] Genesis received a draft meter sales agreement from BOPE at some stage before 4 August 2000, and on that day replied with a number of amendments.Negotiations continued into September and October. In late August, BOPE advised all meters would have to be upgraded to be MARIA-compliant prior to sale. This added to Genesis' concerns about possible delays involved. A price in the order of $250 for a standard single phase installation of two meters and a relay was mentioned by BOPE. Genesis also sought agreement from BOPE that if Genesis lost a customer, BOPE would buy the meter back, irrespective of who the customer switched to. That proposal was accepted. At the same time, BOPE required an option to purchase meters which Genesis might otherwise sell to a third party. [145] By 10 November 2000 Genesis and BOPE had reached agreement. Genesis had also completed its inquiries of the original 170 customers. Although Mr Pickup could not be sure, he thought the final number of BOPE customers who switched to Genesis was less than 37. [146] The agreement was eventually signed on 11 December 2000. BOPE then allowed customers to be switched to Genesis. [147] Mr Pickup's evidence was that Genesis would have preferred to lease BOPE's meters, but considered it had little choice. It had to do something to get the customers switched to it. Technically, Genesis could have installed its own meters, but installation costs would – relative to the small profit margins on sales to Consumers – have been prohibitive. [148] Genesis eventually sold the vast majority of its meters to NGC. BOPE did not exercise its option, and Genesis from then on leased its meters from NGC. [149] In 2002, and after Mr Pickup had left Genesis, Genesis wrote to BOPE advising that in the future Genesis would arrange for new metering equipment to be installed by its meter service provider for customers it acquired in the BOPE area. As that letter post-dated Mr Pickup's employment, he was not in a position to comment further on its significance. [150] Mr Pickup did not make any comment as to the effect, if any, Genesis' experiences with BOPE had on its efforts to market electricity in BOPE's incumbency.TrustPower[151] TrustPower first raised the issue of meter access with BOPE in general terms in an email of 6 April, sent to a number of industry participants. That email proposed various standard procedures to supplement the then existing market rules. Mr Bulley replied, expressing interest in negotiating with TrustPower. Nothing would appear to have come of that. [152] In September 1999 TrustPower wanted access to meters at sites in the BOPE Area, to serve its national customer Telecom. Like other generator/retailers, TrustPower's commercial strategy was to consolidate business in its incumbency areas, and to expand its business in a targeted fashion throughout New Zealand, in particular by acquiring larger customers. Telecom was one such customer. When TrustPower acquired customers in areas where it was not the incumbent, it had to procure access to meters. With all incumbents other than BOPE, it achieved this pursuant to meter leasing agreements on a reciprocal basis. TrustPower did not specifically target the BOPE area. [153] After some initial discussions, Mr Harnett sent Mr Bulley TrustPower's standard metering agreement on 3 September 1999. Mr Bulley provided it to Messrs Bahirathan and Till that day, and the question was discussed on 7 September. BOPE's initial position was that it would lease the 20 meters actually involved to TrustPower, subject to TrustPower agreeing to pay a disputed transmission invoice. TrustPower's response on 9 September was that it would not proceed on that basis. If necessary it would install its own meters, or purchase BOPE's meters. [154] On 10 September Mr Bulley emailed Todd, including Messrs Tweedie and Bahirathan, in the following terms, referring to the TrustPower response:It still appears that they are only concerned about the Telecom sites (20) in our patch. To avoid having scrap meters should we consider selling the meters as suggested? The other issue for consideration is that a bi lateral arrangement as tabled allows for a notice period of 1 month and is site specific ie we can exclude any sites from the agreement – therefore leaving us (and them) in control if we were unhappy with any marketing push. As also discussed on Tuesday we need to be ready to move into TrustPowers region and without a metering arrangement for the mass market it will be difficult. (CB 3/614)[155] By 13 September, Mr Bulley had Mr Tweedie's approval for a 20 meter reciprocal arrangement to cover the Telecom sites. He advised Mr Harnett of this on 21 September. Although TrustPower proposed a 100 site agreement, TrustPower was prepared to accept the arrangement in order to honour its obligations to Telecom. TrustPower forwarded standard documentation on 28 September, already executed by it. The specific details were to be recorded in an exchange of letters. [156] That documentation was never signed by BOPE. [157] Notwithstanding this, from 1 September onwards the Telecom non-time of use sites started to be switched from BOPE to TrustPower, and TrustPower used the BOPE meters at those sites. [158] On 24 December 1999 Mr Bulley emailed Mr Harnett to discuss their metering agreement. Mr Bulley subsequently suggested, in January 2000, extending the arrangement to 100 sites. Mr Harnett recalled agreeing to that suggestion. [159] That situation prevailed until August 2000. Internal Todd/BOPE emails of 2 and 3 August 2000 then refer to an apparently imminent campaign by TrustPower in the BOPE area, and to the need for BOPE to be ready to respond. Mr Harnett could not recall whether TrustPower was proposing such a campaign. He could say that TrustPower had not in fact undertaken such a campaign. Mr Tweedie's evidence was that the BOPE management took the threat seriously. [160] It was agreed by Mr Tweedie that a "short sharp stopper" (Mr Bulley's terms) would be sent to TrustPower, terminating the existing lease arrangements Mr Bulley's email to Mr Tweedie includes the following comment on the proposed course of action:The advantage of the above is that it is consistent with our position with other retailers and can be tabled if we are challenged by the Commerce Commission. Also it may have a desired effect on the rumours we are hearing! (CB 6/1669)[161] On 4 August 2000, Mr Bulley emailed Mr Harnett advising him that BOPE was withdrawing from the meter leasing agreement, and that no further rental agreements would be entered into. A possible purchase option was mentioned, withBOPE to contact Trustpower when details were developed. TrustPower continued to lease the meters, principally non-time of use meters, that were the subject of those arrangements, prior to their termination by BOPE. [162] There were further interactions between Trustpower and BOPE on meter access in November 2000. This time, BOPE approached TrustPower about leasing some TrustPower metering assets (streetlighting relays and non-time of use meters) relating to the Rotorua District Council. The Council was a TrustPower customer that had moved to BOPE. [163] TrustPower replied, saying it would not lease to BOPE if BOPE would not lease to it. BOPE confirmed its unwillingness to enter lease arrangements. That situation prevailed for some months. During the same period, each of Trustpower and BOPE also refused to process switches to the other, unless new meters were installed. [164] After a few months TrustPower decided this was an unproductive exercise. It also resumed leasing its meters to BOPE. As at July 2006, TrustPower was leasing 1,162 non-time of use meters (together with 1,517 relays) at 762 properties to BOPE in TrustPower incumbent areas. [165] It appears that BOPE also stopped rejecting TrustPower switch requests in the BOPE area and allowed TrustPower to use BOPE meters. As at October 2006 TrustPower was using non-time of use meters owned by BOPE at 37 ICPs in the BOPE area. Of those 37 ICPs, 36 were Telecom or other "national" customers. [166] As at October 2006, TrustPower also owned 27 non-time of use meters at 27 ICPs. These meters were installed during the period when BOPE was refusing to allow customer switches to TrustPower unless TrustPower installed its own meters. They are mainly installed at national customers with a few installations at non- national customers. [167] Very late in the piece, in fact on 6 July 2006, TrustPower became aware that BOPE had never invoiced it for the use of BOPE meters in the BOPE area. TrustPower drew this to BOPE's attention. BOPE required TrustPower to sell toBOPE the Trustpower meters and relays BOPE was using, or BOPE would replace those with its own. BOPE similarly advised TrustPower that TrustPower was to replace or buy the BOPE assets it used by the end of October 2006. Those assets were to be upgraded for MARIA compliance prior to sale. [168] TrustPower decided not to buy the BOPE meters, but to replace them with its own. TrustPower wished to avoid BOPE changing existing meters to certified MARIA compliant meters, as that exercise would have given BOPE an opportunity to attempt to win back customers. Although its preference would have been to continue to lease those meters from BOPE, TrustPower was prepared to incur the higher per meter ownership cost for its national customers. [169] Mr Harnett explained TrustPower's preference to lease meters from BOPE, rather than installing its own, by reference to the following factors: (a) The disincentive of having to disconnect power to install a new meter; (b) The higher costs associated with installing a small number of meters; (c) If the customer switched away from TrustPower and back to BOPE, TrustPower might have to write-off installation costs, and would incur recovery and re-testing costs. [170] Mr Harnett said that TrustPower had never attempted to target the BOPE area, one of the reasons being the difficulties it experienced with access to BOPE's meters. At the same time, Mr Harnett's evidence confirmed that TrustPower had continued to offer at least some national deals in the BOPE incumbency, and that the stranding risks were lower in that kind of deal. Mr Harnett also accepted that TrustPower's chief executive, Mr Tempest, had in 2004 publicly expressed the view that BOPE had competitive pricing so that, even if BOPE did enter into reciprocal meter leasing arrangements, it might not be economical for TrustPower to compete with BOPE in its incumbency. [171] Whilst Mr Harnett expressed the view that a mass marketing campaign based on meter installations was not an economic proposition, he was not able to say thatbut for BOPE's metering policy TrustPower would have undertaken any such campaigns. This was, in our view, consistent with his explanation of TrustPower's commercial strategies over time of defending its incumbency areas and undertaking targeted marketing (eg for national customers) outside those areas. In our view, the more likely explanation is that BOPE's attitude to meter leasing confirmed that policy, rather than being the reason for it in the first place.Meridian[172] Meridian was another generator/retailer which acquired Consumers in the BOPE area in connection with a national offering – an offer of electricity in connection with a favourable Sky subscription price. This promotion started in July 1999 and ran through until November 1999. Again, like other generator/retailers, Meridian entered into reciprocal metering agreements with a number of incumbent retailers to support its marketing activities. [173] Mr Treadwell sent a draft metering agreement to Todd on 3 August 1999. No evidence was provided as to any response from Todd, or of any follow up by Meridian. It would appear that Todd simply did not respond to Meridian. [174] By December 1999 some 365 customers in the Horizon Area had accepted the Meridian offer. Although no progress had been made on the issue of access to meters, the customer switching protocol had been initiated. Meridian had given notice to BOPE that these Consumers wished to switch and BOPE had provided "initial" acceptances in the majority (290) of cases. No switch request had actually been effected. In early December, however, a very public dispute broke out between Meridian and BOPE. The immediate background to that dispute would appear to be a conversation Mr Bulley had with Horizon's general manager. [175] On 29 or 30 November Mr Bulley had spoken with Horizon about the position regarding the Consumers who were to be switched to Meridian. Mr Tweedie said two things then became clear to BOPE: (a) Although Meridian did have a use of system agreement with Horizon, that agreement could not operate until Meridian had a way of gettingongoing metering data. As BOPE knew no agreement had been reached on metering, it concluded that Meridian could not yet activate the Horizon agreement. (b) Under the switching process, once a switch had been instigated the Consumers would no longer be BOPE's. BOPE understood this to mean the Consumers would then be in a billing lacuna, as they would still be drawing electricity from the network but would be unbilled by either retailer because BOPE was not billing them (being no longer responsible for them) and Meridian could not bill them as the switches had not been completed and Meridian did not have access to the data. BOPE understood it would remain as the incumbent retailer responsible for the line and network charges for those Consumers. [176] BOPE estimated at that time that the cost to it for line and energy charges for the switching Consumers could run to approximately $20,000/month. [177] Mr Tweedie acknowledged that that understanding was wrong, but claimed that it lay behind its subsequent actions. [178] BOPE drafted a letter to go to the customers involved. In the letter BOPE said bluntly that the situation could not be tolerated and it therefore would finalise their accounts and arrange for their supply to be disconnected. Wednesday 8 December, or as soon as possible after that date was set, for the disconnection. [179] BOPE made a press release on 6 December, publicising its intention to disconnect its Consumers (approximately 200) who were in the process of switching to Meridian. BOPE's threat to disconnect Consumers attracted considerable local publicity. [180] The press release also produced an immediate response from Meridian, disputing the facts as asserted by BOPE, and drawing BOPE's attention to its failure to process the switch requests that it had formally acknowledged. Various letters and emails were exchanged between Meridian and BOPE over the next few days. BOPE continued to focus on Meridian's alleged failure to make appropriate arrangementswith Horizon and/or as regards metering data, and/or requiring Meridian to publicly apologise to Consumers for the inconvenience caused by its actions. Meridian asserted that it had made appropriate arrangements and, furthermore, that it would be responsible for relevant line and energy charges. Agreement on a way forward could not be reached, and BOPE's threat to disconnect remained on the table. [181] The parties met again on 13 December. That same day, BOPE advised Meridian that it was preparing a metering agreement. By letter later that day Meridian confirmed its willingness to negotiate such an agreement. Meridian also sought confirmation from BOPE that it would not disconnect Consumers without a reasonable period of notice, and said that if it did not receive that confirmation by 4.00pm it would have no choice but to seek an urgent injunction to that effect. No such undertaking was forthcoming. [182] On that basis, Meridian instructed its solicitors and injunction proceedings were commenced on 15 December. [183] BOPE, according to Mr Tweedie, now realised its mistake:As a result of that proceeding (and particularly Mr Treadwell's affidavit filed in support of it), BOPE became aware that it had been mistaken about the billing position under the switching process..BOPE therefore had no reason to disconnect the customers. (Tweedie brief, para 4.27)[184] As a result of that development, and BOPE's confirmation that it would not disconnect any Consumers until they had actually switched from BOPE to Meridian, Meridian withdrew its injunction application. This was, however, not the end of the matter. [185] At that point, Meridian decided to install its own meters and set about making arrangements with Consumers and with United Networks to achieve that procedure. Throughout January 2000 the parties continued their dispute, BOPE asserting that Meridian had not made appropriate arrangements for the removal of meters and for the finalisation of the switches. BOPE justified its actions by pointing to the fact that, in the absence of an agreement with BOPE for access to BOPE meters, Meridian would still have had no method of obtaining metering data from the Consumers involved. Meridian would not have been able to bill Consumers.Furthermore, if BOPE had provided that data, without contractual agreement, Meridian would have railroaded BOPE into a situation equivalent to having entered into such an agreement. Therefore BOPE thought it was within its rights to continue to refuse to switch Consumers. [186] From Meridian's point of view, the whole dispute had been unnecessary. If BOPE had adopted an approach similar to that taken elsewhere in the industry, interim meter access arrangements would have provided the necessary metering data, and Meridian would have met all relevant energy and line charges. [187] A protocol for switching and meter replacement was finally agreed on 24 March 2000 and the switches were able to occur. [188] Of the 315 Consumers who initially applied to switch to Meridian as part of the Sky promotion, over 200 of them ultimately did not switch. BOPE acknowledged that during the injunction hearing period, and pending final switching, BOPE had continued to make efforts to win back affected Consumers. It did so through public statements, and letters directly to Consumers involved. [189] Mr Tweedie acknowledged that, in hindsight, some of the comments in those press releases and letters were overly aggressive marketing particularly where they referred to the cost and inconvenience Consumers would face in attempting to switch to Meridian. There would appear to be little doubt that a significant number of Consumers were dissuaded from taking up the Meridian offer as a result of BOPE's "win back" efforts, including those which Mr Tweedie acknowledged exaggerated the difficulties and costs involved with transferring retailers. [190] Although Meridian did install the meters for those remaining Consumers, in Mr Treadwell's evidence doing so was "not economically justifiable". Meridian only did it to honour its nation wide agreement. [191] Mr Treadwell's evidence was that:There were a number of reasons why installing its own meters in the Horizon network area was uneconomical for Meridian. The main reason was the relatively high up front meter and installation costs as compared with therelatively low retail margins for non-time of use customers. Meter and installation costs were as follows: Installation cost per meter $40 plus GST Single phase single register meter $82 plus GST Three phase meter $262 plus GST Ripple relay $117 plus GST An average household had two meters and one ripple relay, which meant the installation and meter/ripple relay costs typically totalled around $361 plus GST per household. However, the retail margin on a domestic customer (based on the industry average consumption of 8,000 Kwhrs per annum) was approximately $80 per annum before the payment of meter leasing costs (which were on average 4.5-5cents per day per meter). It would therefore take a number of years to recoup the relatively significant up front meter and installation costs. (Treadwell Brief, paras 42 and 43).[192] Mr Treadwell also noted the stranding risk if customers subsequently switched back to BOPE and BOPE refused to lease the meters from Meridian. He also noted the lack of scale associated with owning a small number of meters in the BOPE incumbency. Those costs and risks were, Mr Treadwell said in his evidence in chief, the main reasons why after its experience with the sky promotion Meridian did not attempt to enter the Horizon Area for the remainder of his time with Meridian (he left in April 2003). When asked, however, whether he could recall any other specific instances of Meridian excluding the Horizon Area from a Meridian nationwide promotion, he only referred to the Sky promotion.Other retailers[193] Documentary evidence only was provided as regard BOPE's interactions with three other retailers. These interactions were referred to by Mr Brown when opening the case for the Commission, but did not feature in the trial to any material extent thereafter. They were, nevertheless, part of the Commission's pleadings. Moreover Mr Tweedie did, in his evidence, comment on certain aspects of these matters. Accordingly we summarise our understanding of them as follows.EOL[194] EOL first contacted BOPE on the meter access issue in May 2000. It was, at the time, targeting commercial customers in BOPE's incumbency.[195] A letter from EOL to BOPE of 31 May 2000 refers to a prior conversation in a way which indicates that BOPE's "no leasing" approach had been clearly communicated:If your intention is not to provide access to your metering and effectively provide a barrier to competition in your area could you please confirm that to me in writing. Our strong preference is for common sense to prevail and for the meters to be left in place. However, we would very reluctantly install our own meters if you leave us with no other option if we are to be able to offer customers a choice of supplier. (CB 5/1494)[196] On 1 June 2000 Mr Doidge advised EOL that, in effect, BOPE would not lease its meters to EOL, BOPE's preference being to retain its own equipment and its belief being it would be unsound for it to establish a meter leasing business. [197] BOPE responded to EOL's marketing efforts with a number of press releases. These press releases asserted that EOL was making false claims as regards meters and meter reading, and as regards the true costs of supply by EOL. An article in the local "Whakatane Beacon" of 2 June reported Mr Bulley drawing Consumers' attention to the need for meters to be changed, and that they could lose power for a while. [198] EOL is reported as stating that any changes in meters would be at its cost, and would only involve a 20 minute disruption. [199] An email of 24 November from EOL to BOPE records EOL having replaced BOPE's meters at 4 ICP's. This gave rise to concerns within BOPE, as it would appear MARIA switch procedures may not have been followed. Whatever the nature of those concerns, Mr Bulley recorded in an internal BOPE email of 11 December:I am also very concerned with the Energy on Line issue, can Graham track down the contractor and threaten something if any more of our meters are touched? (CB 7/1819)[200] A further "Whakatane Beacon" article in October 2002 reported EOL as having cancelled contracts with 58 Consumers because of difficulties encountered in the switching process.[201] Mr Tweedie noted that EOL had subsequently become a wholly owned subsidiary of Genesis, and "now trades in the Bay alongside its parent".Empower[202] In early January 2000 Empower notified BOPE that it had acquired a small number of Consumers in the BOPE incumbency as a result of national deals. On 20 January Empower sent BOPE a metering agreement to consider. This was a very simple one page document that Empower had previously used with TransAlta and Genesis. [203] Receiving no reply, Empower sent an email to BOPE on 25 February 2000 asking whether the metering agreement was acceptable. [204] BOPE replied in an email of 28 February 2000, saying that BOPE's current policy was for retailers to install their own metering but that the issue was on the agenda for a joint venture partners' meeting the next day. Empower would be updated following that meeting. [205] An internal BOPE document – papers for BOPE JV meeting of 29 February – indicates that BOPE's failure to respond to Empower may not have been a simple oversight. In a report on outstanding metering agreement issues, the following comment appears:Empower – letter received before Christmas, patience wearing thin with BOPE non-response. (CB 5/1317)[206] It is not clear what then happened, but on 8 August 2000, following the 1 August JV board meeting, BOPE confirmed that meter leasing would not be offered. It suggested two alternatives: BOPE could sell the meters to Empower, with a buy-back option for BOPE, or Empower could provide its own metering. [207] Empower ultimately purchased two MARIA certified meters (to be installed by BOPE) at two sites.[208] There was no evidence as to whether or not these interactions had materially affected Empower at the relevant time, or whether they had any subsequent impact on its marketing activities.Contact[209] Limited evidence was provided of interactions between Contact and BOPE. [210] On 17 June 2001, and apparently not for the first time, Contact raised with BOPE the question of meter leasing arrangements. Anticipating BOPE's response based on issues of administrative costs and administrative difficulties, Contact offered to self-invoice for metering services provided to it by BOPE. [211] Mr Doidge replied the next day, confirming BOPE's no lease policy but undertaking to pass the matter to the owners for consideration. Nothing further eventuated. [212] In January 2005 Contact received a copy of BOPE's standard letter, in response to a Consumer switch notification Contact had sent BOPE. Contact responded to that letter, averting to the stranding risk issue and querying whether BOPE would lease the meter from Contact if BOPE won the Consumer back. It also stated its view that installing new meters upon gaining new Consumers was not a viable option for it. [213] After an exchange of correspondence, BOPE indicated on 21 April 2005 that:the preferred solution for a customer transferring from Contact to BOPE in any of the regions we operate would be for BOPE to purchase the metering from Contact, subject to the equipment being of industry standard and its cost being acceptable. (CB 9/2451)In other words, BOPE would purchase Contact meters. [214] No evidence was provided from Contact of the effect, if any, of these interactions on Contact's marketing activities in the BOPE incumbency.Evidence from consumers[215] Of the five Consumers who provided written statements, four of these were BOPE customers who responded to offers by Meridian. [216] Mr Sedgman accepted the Sky promotional offer. He filled out an application form, and received a notice back from Meridian that he would be switched and they would make all the necessary arrangements. He subsequently received a final invoice from BOPE and, a few days after that, he was contacted by Meridian and arrangements were made to disconnect his electricity so that a new meter could be installed. Those arrangements were carried out without any apparent inconvenience to Mr Sedgman. Some time later, in August 2000, he was still receiving invoices from BOPE. On 9 August he received a letter saying he would be disconnected because he had not paid his BOPE invoice. After some confusing communication from BOPE, he was in fact disconnected at about 6.30pm on 9 August. He was not reconnected until 12.30pm the next day. BOPE subsequently apologised to him, admitting they had no right to disconnect him and that the disconnection had been their error. [217] Messrs Nunweek and Chappell were also BOPE customers who accepted the Sky offer. They got caught up in the dispute between Sky and Meridian in December 1999. Both took strong objection to BOPE's threat at that time to disconnect their power. Mr Nunweek eventually did not go ahead with the Meridian offer, because he considered it wrong that in order to change from one electricity supplier to another they were required to replace a perfectly good meter. Although he understood he would not be charged for the cost of installing the new meter he knew someone would have to pay for it and thought that it might ultimately be them at least indirectly. Mr Chappell persevered in face of BOPE's threats and, through direct communication with Meridian, in December became satisfied that his power would not be disconnected. Mr Chappell was outraged at what he considered to be the standover tactics of BOPE, and was determined to go through with his transfer to Meridian. He remains a Meridian customer. [218] Miss Spelmeyer decided also to accept an offer from Meridian. It was not clear whether it was the Sky offer or not. However, on reading publicity in the"Whakatane Beacon" that customers transferring from BOPE to Meridian might have to have their electricity disconnected she decided to remain with BOPE. She ran a bed and breakfast business and could not take the risk of being without electricity. She, however, considered the position to be ludicrous. [219] The fifth witness, Mr Gary Van der Worth, responded to an Energy Online offer. He accepted that offer with respect to two commercial businesses he was involved in on 30 May or thereabouts. He then heard nothing for a couple of months and contacted EOL to find out what was happening. He was told that the meters would have to be changed because BOPE did not let EOL use BOPE meters. That was causing the delay. [220] The meter for one business was then changed and EOL began invoicing. A disconnection for approximately one to two hours was involved. Mr Van der Worth was subsequently contacted by the landlord of the building that that business occupied. Apparently the landlord had been advised by BOPE that BOPE's record showed that the building occupied by the business was unoccupied, because the meter had been disconnected. BOPE therefore proposed cutting off power to that building. Mr Van der Worth contacted EOL, and the misunderstanding was sorted out without the business's electricity being cut off. [221] The meter for the second business had still not been changed. When Mr Van der Worth contacted EOL he was told that BOPE was preventing EOL proceeding with any more connections. EOL did not come back to them after that, and that business, in fact, remained a BOPE customer at that time. [222] Responding to local publicity in the "Whakatane Beacon" about difficulties businesses faced switching from BOPE to EOL, Mr Van der Worth contacted the paper and was the subject of an article. After that article was published in October 2002, EOL contacted Mr Van der Worth, and arranged for the meter for the second business to be changed so that the second business could in fact become an EOL customer. That was eventually arranged in December 2002, involving electricity being switched off for one to two hours whilst the meter was replaced. That second business remained a customer of EOL at the time in 2006 when Mr Van der Worth provided his written statements.Other evidence of BOPE's dealings with competing retailers[223] As can be seen from the foregoing, by far the largest part of the Commission's evidence relating to BOPE's dealings with competing retailers related to the period up to and including the end of 2000, during a considerable part of which period there was general consensus that the industry had been in a state of some turmoil, and market arrangements did not work as effectively as anticipated. [224] There was little, if any, direct evidence from retailers as to their interactions with BOPE after that time. [225] BOPE itself provided evidence as to entry by competing retailers into the local market, designed to support its contention that the basis upon which BOPE dealt with those retailers did not create any material impediment to those dealings. [226] In particular, it was BOPE's position that when retailers recognised that BOPE was not going to lease them its meters, and investigated the alternatives to leasing, they acknowledged the viability of those alternatives and actively competed for domestic customers in the Horizon Area. The result was, in BOPE's contention, that there were now five substantial electricity retailers who had priced and offered electricity to Consumers in competition with BOPE, namely Contact, EOL, Genesis, Mighty River (through its Mercury brand) and Meridian. [227] Mr Tweedie noted, in evidence not disputed by the Commission, that in calendar year 2006 460 BOPE customers in the Eastern Bay had switched to other retailers (94 to Contact, 189 to Genesis, 135 to Mighty River, 32 to EOL, 8 to Meridian and 2 to TrustPower). [228] Mr Bahirathan provided a chart of switches from BOPE to other retailers. Mr Bahirathan's position was that, particularly in the case of Meridian, Genesis, Contact and Mighty River (through the Mercury brand) this showed steady switching in conjunction with competitive retailing by those firms in the Bay Area. [229] In cross-examination of Mr Bahirathan in particular, the Commission pointed to internal Board papers reporting on the state of competition in BOPE's incumbentarea. The Commission's submission was that these were a record, and an unusually frank record, of the extent to which BOPE had succeeded in its aim of limiting competition. BOPE saw them, rather, as evidence that BOPE was at all times aware of the risk of competition, and the need therefore to behave competitively as regards both the price and quality of its service offering.BOPE's standard letter[230] In January 2002, Mr Hall was appointed as BOPE's internal general counsel. [231] Following Mr Hall's appointment, BOPE developed a standard letter to send to retailers who requested customer switches. That letter, which would appear to have come into existence in late 2002, advised retailers that BOPE did not lease meters. BOPE was, however, prepared to sell meters to retailers. Where an existing meter had not been upgraded to be MARIA compliant, BOPE would install a MARIA compliant meter, and sell that meter. [232] The standard letter also advised of other options available to retailers, such as installing their own meters or leasing meters from a third party. It included a cost comparison, which showed retailers were better off installing their own meters, relative to the costs they would have incurred to lease meters. [233] In terms of the use of, and response to, that letter, available documentation recorded the earliest date such a letter was sent to a retailer was 20 July 2003. [234] It was not clear whether any retailer had accepted BOPE's offer to sell meters.Our conclusions on BOPE's approach to competitors' requests for access to its meters[235] The Commission's case against BOPE is based on the approach taken by BOPE to competitors' requests for access to its meters. On the basis of the evidence referred to in the preceding section of this judgment, we now summarise the approach taken by BOPE to that issue.[236] Most clearly in the period from April 1999 to mid to late 2000, BOPE did not willingly engage with competitors on the question. It generally sought to avoid and/or delay dealing with them on the issue. [237] As recorded above: (i) Mighty River first contacted BOPE on the issue on 25 June 1999. It had not received any substantive reply by 31 January 2000. (ii) Genesis first contacted BOPE on the issue in February 2000. Although BOPE's initial response, as understood by Genesis, was positive, no substantive response was received until 13 July 2000 (BOPE's offer to sell). Various responses to Genesis, referring its representatives to Todd, then back to BOPE, then to the joint venture partners, are in our view significant indicators of the approach taken. (iii) TrustPower did get reasonably prompt access to meters at 20 sites in September 1999. That arrangement was, however, never formalised and was "cancelled" in August 2000. (iv) Meridian first contacted Todd on the issue on 3 August 1999. Todd simply did not respond. Following the December 1999 dispute, the parties did engage with each other. Agreement was not reached, however, until late March 2000. (v) EOL, Empower and Contact had more limited interactions with BOPE, which did not involve such obvious delays. BOPE did, however, internally acknowledge delay as regards Empower through January and February 2000. [238] When it did engage on the issue, BOPE declined to approach the matter on the basis of the contractual arrangements that were proposed by its competitors, and which were generally accepted in the industry.[239] For a period, BOPE responded in a variety of ways to requests for access. Thus: (a) Mighty River proposed leasing arrangements in June 1999, but decided in February 2000 to install its own meters. BOPE indicated a willingness to negotiate on leasing in late March 2000, but withdrew that offer in May 2000. Mighty River ultimately installed its own meters. (b) Genesis proposed reciprocal leasing arrangements in February 2000. Whilst Todd dealt with Genesis on Fresh Start's leasing of Genesis metres, BOPE did not engage significantly on the leasing issue in relation to its own meters. In July 2000 BOPE offered to sell BOPE meters to Genesis. That arrangement was finalised in November 2000. (c) TrustPower reached agreement with BOPE in September 1999 for leased access on a limited basis (20 meters). Those arrangements were purportedly cancelled in August 2000. After a period of non- cooperation, TrustPower and BOPE did provide each other lease based access to meters from later in 2000. (d) Meridian's July 1999 proposal to lease had received no reply when hostilities broke out in December 1999. Leasing was never agreed to by BOPE. Meridian installed its own meters from March 2000 onwards. Although Mr Bulley proposed leasing after that date, this never eventuated. (e) BOPE was fairly clear with EOL, Empower and Contact, advising that leasing was not BOPE's policy in June 2001, February 2000 and June 2000 respectively. [240] The general, no-leasing policy – as accepted by Todds in its pleadings – developed over a period of time.[241] As early as February 2000 BOPE's position was put by Mr Bulley to Empower as being that "our policy is for retailers to install their own metering" (CB 5/1305). Responses to the Commerce Commision, earlier in February on the Mighty River complaint, also confirm this approach. [242] It would appear a decision of some sort not to lease meters may have been taken, or at least discussed, around 1 June 2000. Internal and external BOPE correspondence around that date refers to BOPE's preference to keep its meters for itself, and not to rent them out. This position was formalised, at the latest it would appear, by the time of the joint venture meeting on 1 August. In evidence, Mr Tweedie described the decision – which was not recorded in this way in any formal record – in the following terms:3.11 By August 2000, the BoPE JV was in a position to make a firm decision on meter leasing. At the BoPE JV meeting of 1 August 2000, the BoPE JV discussed the issue, and Mr Bulley and Mr Doidge emphasised to us both the administrative problems facing BoPE as regards a meter leasing business, and the lack of any business case for one. It was resolved at that meeting that: (a) BoPE would not generally enter into meter leasing agreements with other retailers. (b) That message would be made clear to them going forward. (c) However, BoPE would also advise them that they could purchase BoPE's meters or install their own. (Tweedie Brief, para 3.11)[243] That decision was reflected very soon after that meeting in correspondence from Mr Doidge to Empower. Mr Doidge refers both to BOPE not setting up a meter rental business, and being prepared to sell, with a buy back option, meters on a site by site basis. [244] The 1 August 2000 decision was confirmed in a joint venture meeting in December 2000. [245] The July 2000 offer to Genesis appears to be the first indication of BOPE's willingness to sell its meters to a competitor. A willingness to sell meters was also referred to by Mr Doidge in his letter to Empower, but had not otherwise beenoffered at the time (for example, to Mighty River or to Meridian), each of whom considered they had in effect been forced to install their own meters. [246] The Commission accepted that, by the time of the 1 August joint venture meeting, the decision had been made that BOPE would not provide access to BOPE's meters as requested by other retailers. The Commission did not accept, however, that any resolution to the effect that BOPE would generally sell its meters to competitors was made at that time. Mr Doidge's reference, in his email of 8 August to Empower, to the agreement being on a "site by site" basis supports the Commission's contentions. Similarly, the negotiations with Genesis, and a reference by Mr Tweedie in July that the sale deal had only been agreed on the basis that a limited transfer of Consumers had been achieved by Genesis, is also consistent with the Commission's contention. On balance, we think it is more likely than not that BOPE did not develop a policy of being willing to sell meters until some time later; although given that after the initial flurry of dealings in 1999 and 2000 there would not appear to have been significant requests for access for some time, that fact in and of itself may not be particularly material. [247] During this period (1999-2000) BOPE engaged on occasions in media publicity designed to put Consumers off the idea of changing retailers. It is not clear, however, as to the extent to which this was a general practice. It certainly occurred in the Meridian dispute, in a way which Mr Tweedie acknowledged may have been excessive and imprudent. Other instances are seen in interactions between BOPE and Empower. [248] It is clear that BOPE, whilst it was dealing with competitors on the question of access to its meters, was also strenuously attempting by direct contact to win back Consumers who had accepted offers from those competitors. This was very clearly the case during the Meridian dispute. BOPE would appear to have done the same with Consumers who had accepted the Mighty River (Anchormart) and Genesis (Rural Alliance Collective) offers. Mr Tweedie's evidence was that such "win-back" campaigns were standard industry practice. [249] During the course of its negotiations with Genesis for the sale of meters, (August – November 2000) BOPE at some point advised that all meters to be soldwould be upgraded by BOPE to be MARIA compliant prior to sale. That was, however, not mentioned as part of BOPE's "policy" at the time of the August 2000 joint venture meeting. [250] BOPE's policies were further developed and refined during 2002 and into 2003. As a result of that process, BOPE had by July 2003 at the latest developed a standard letter which it sent to retailers from whom it had received a switch notification for an ICP. That letter confirmed that: (a) BOPE did not lease its meters; (b) BOPE was willing to sell the competing retailer its meters; (c) If the competing retailer was interested in purchasing meters, BOPE would send it a firm price on request; and (d) BOPE would upgrade the meters to be MARIA compliant prior to sale. [251] Enclosed with that letter was a "cost comparison" which indicated that buying the meter was economically preferable, in terms of comparative net present values, than leasing the meter. There were few, if any, sales by BOPE of meters to competitors under this policy. [252] Comparatively little evidence was provided by the Commission of interactions between BOPE and retailers in the period from the beginning of 2002 onwards. Evidence from BOPE did establish entry by a number of retailers into the Eastern Bay. The significance of that entry was disputed by BOPE and the Commission. The Commission's position was that its relatively limited extent was evidence of the efficacy of BOPE's approach to access to meters in limiting competitive activity. BOPE cast it in the opposite light. That is, BOPE saw this as evidence of BOPE's competitive conduct in keeping prices low and service quality high.[253] The question of BOPE's approach to the issue of whether it would, were it to win a customer back, lease or buy a new meter installed by a competitor is not as clear. [254] This is not an aspect of BOPE's behaviour or of its policy that was referred to directly in the Commission's pleadings. Mr Brown did argue that a policy of not leasing, or not purchasing, meters installed by its competitors already came within the Commission's description of BOPE's policy of being one of "not entering into meter leasing agreements". We find that argument difficult to accept. We think the pleadings make clear that the policy referred to is one of BOPE declining to provide electricity metering services to competitors (see paras 13 and 14 of the Commission's Statement of Claim). [255] Some of the Commission's industry witnesses had, nevertheless, referred to this matter. [256] Mr Reilly referred to the possibility of stranded assets, given the risk that a customer might switch back to BOPE after a short period of time and Mighty River might be left with a stranded asset. Mr Reilly did not, however, give evidence as to whether BOPE, at the time of his dealings with it, had ever communicated, or even had, a "position" on this matter. [257] Mr Pickup did not refer to this issue directly as being a matter of concern for Genesis. In negotiations with BOPE for the meter sales agreement, however, he commented to Mr Bulley that it was not Genesis' intentions to acquire BOPE's meters on a permanent basis. Genesis therefore proposed that if it lost customers it would "be happy for the meters to return to BOPE ownership" (CB 6/1671). When that agreement was signed in December 2000, cl 11 gave Genesis the contractual right, where it lost a customer, to require BOPE to purchase the meter involved. [258] That BOPE in December 2000 agreed to this contractual provision would appear to indicate that, at that time, it did not have a "policy" of not buying meters back.[259] Mr Harnett referred to the stranding risk. He spoke of a risk that a TrustPower customer would switch back to BOPE and BOPE would then install its own meter rather than leasing the existing TrustPower meter. Like Mr Reilly, however, Mr Harnett did not give evidence as to whether BOPE, at the time of his dealings with them, had communicated or even had a "position" on this matter. [260] Mr Treadwell referred to the matter in a similar way, noting that there was a risk of meters in effect becoming stranded assets if customers subsequently switched back to BOPE and BOPE refused to lease the meters from Meridian. Again, however, Mr Treadwell did not say whether or not that issue had been raised by Meridian with BOPE, or whether BOPE had indicated any policy or view on it. [261] The stranding risk issue was not referred to in the material relating to dealings by BOPE with EOL or Empower. We do note, however, that in negotiations with Empower BOPE had requested that it be provided with a buy back option. That it was interested in such an option would not appear to be consistent with it having a policy, at that time, of not acquiring MARIA-compliant meters installed by competitors. [262] Contact raised the issue in July 2005. At that time BOPE indicated that it would purchase meters installed by Contact if Contact lost a customer back to BOPE. This is also at odds with BOPE having a policy not to do so. [263] Evidence relating to the stranding risk was also provided by Commission witnesses from two companies which operated meter leasing businesses. [264] Mr Barnett, the metering business supervisor for The Lines Company, stated that since July 2001, when The Lines Company had started operating metering businesses in network areas throughout New Zealand, the company had installed 28 ICPs at consumer sites in the Horizon Area. At seven of those sites, the customer had subsequently switched to BOPE and BOPE had removed the meters at four of those sites. [265] Mr Barnett's evidence was that the only other company that had removed meters in that situation was TrustPower in its incumbent areas. Mr Barnett'sevidence was that although The Lines Company was prepared to install small numbers of meters in the Horizon Area, it would not – because of the stranding risk – install large numbers without getting protection from its customer for that risk. Mr Barnett considered it unlikely that a customer would be willing to take on that risk. [266] There was some challenge to the accuracy of Mr Barnett's records as regards The Lines Company's meter assets in the BOPE area, but we do not think that challenge affects this aspect of Mr Barnett's evidence. Mr Barnett also confirmed that, relative to other areas where The Lines Company operated, it had a very small number of meters installed in the BOPE incumbency. Contact was its main customer, and Mr Barnett's understanding was that it had problems in the past and so did not go into that area. [267] Messrs Congalton and Dennehy, employees of Vector Limited ("Vector") also gave evidence. Vector was the owner of NGC and they had been employed by Vector following its acquisition of NGC's metering business. They also referred to BOPE's practice of installing its own meters when it won a customer previously serviced by NGC owned meters. NGC only dealt with Genesis as regards the Horizon Area. Were another retailer to ask NGC to install a meter in the Horizon Area, NGC would have to consider a number of factors, including the stranding risk, before deciding whether or not to accept the installation. [268] NGC witnesses acknowledged that NGC was currently in negotiations with BOPE regarding the possibility of NGC purchasing BOPE meters. [269] Neither Mr Tweedie nor Mr Bahirathan themselves referred to BOPE having any policy on this matter, and the issue was not put to them by the Commission. [270] The issue was raised with Mr Doidge. His evidence was that there was no general policy for BOPE not to purchase or lease existing meters from its competitors, rather it was decided on a case by case basis. In the case of some TrustPower meters, BOPE had simply taken over the TrustPower meters and given TrustPower a couple of its meters in exchange for them. In other cases, BOPE had replaced meters either because they were not MARIA certified or because they were otherwise not suitable for BOPE's purposes. Aspects of Mr Doidge's evidence,which appeared to identify the need for BOPE to replace meters installed by other retailers by reference to BOPE having a different standard metering configuration than those retailers, appeared inconsistent with aspects of Mr Bahirathan's evidence as to when BOPE moved to a "single meter" policy. [271] That the significance of stranding risk (but not necessarily an assertion of a BOPE policy giving rise to such risk) had been put in issue by the Commission's case, as reflected in the evidence of Mr Treadwell and Mr Reilly, was acknowledged for BOPE in Mr Murray's witness statement and his overview. In Mr Murray's view this was a risk associated, not with any BOPE specific behaviour, but with competition for Consumers. That is, whoever owned a meter, and serviced a Consumer, risked losing that Consumer to another retailer. There was no certainty the new retailer would wish to lease that meter. Even in that situation, the initial investment in the meter was not lost, as the meter could be removed and re-used, acknowledging there that installation costs would be lost. Mr Murray provided evidence quantifying that stranding risk. [272] In our view, therefore, and notwithstanding that this issue may not have been directly referred to in the Commission's pleadings, we think it is an issue which was raised by this case and of which BOPE was aware of as such. We proceed on that basis. [273] On the basis of the evidence summarised above, however, we think it is difficult to conclude that the Commission has established that BOPE had, at a relevant time, a formal policy of not being prepared to buy or lease new meters from competing retailers who had installed them. What can be said, we think, is that there was evidence that some of BOPE's competitors, and other meter leasing retailers, were aware of the risk that that might occur, and that on a number of occasions this is what in fact occurred. Moreover, and given BOPE's general approach on the question of leasing meters, it would appear to make sense for BOPE, where it won a customer back, to install its own meters rather than lease them from a competitor. [274] We consider other aspects of the Commission's factual assertions, and the parties' evidence on those matters, when analysing the application of ss 36 and 27.The Legal Framework[275] The plaintiff alleges breaches by the defendant of ss 36 and 27 of the Commerce Act 1986.Section 36[276] The inquiry called for where a breach of s 36 is alleged was described by the Privy Council in Telecom Corp NZ Ltd v Clear Communications Ltd [1995] 1 NZLR 385 in the following terms:(1) In order to show a breach of s 36, three elements have to be present: (i) a person who has a dominant position in a market; (ii) who has used that dominant position; (iii) for the purpose of the matters referred to in paras (a), (b) and (c) of subs (1). (p 402, lines 21-25)[277] It is, therefore, also necessary to establish the existence of the relevant market. [278] The law as to the individual elements of s 36 is reasonably well established. Counsel for the plaintiff and the defendant were in general agreement on those matters. Given, however, the difficulties that inevitably arise in cases under the Act in applying that law to the facts, and the number of matters which did give rise to differences of view as between the plaintiff and defendant, we will record our understanding of the relevant legal principles. [279] We turn now to the individual elements of s 36.Market[280] In addition to noting the s 3(1A) statutory definition, including its references to substitutability and commercial common sense, we refer particularly to the discussion of market issues in Port Nelson Ltd v Commerce Commission [1996] 3 NZLR 554 (CA) at 559-562.[281] We note the following elements of that discussion, including as taken by the Court in Port Nelson (CA) from – as indicated by the bracketed references in our text – Tru Tone Ltd v Festival Records Retail Marketing Ltd [1988] 2 NZLR 352 and the Australian decisions of Singapore Airlines v Taprobane Tours WA Pty Ltd[1992] ATPR 41-159 and Queensland Wire Industries Pty Ltd v Broken Hill Proprietary Co Ltd (1989) 167 CLR 177: (a) The focus in s 3(1A) on distinguishability as a matter of fact and commercial common sense avoids giving prominence to any particular criterion, including substitutability in response to price change. The test is not substitutability as such, although that is an important criterion. Market is a multi-dimensional concept, including dimensions of product, functional level, space and time. (Tru Tone Ltd) (b) The concept of market has a descriptive and purposive role. Fact finding, together with evaluative and purposive selection, are involved in the market designation exercise. (Singapore Airlines) (c) The market designation imposes on the activities which it encompasses limits set by law for the protection of competition. It involves a choice of the relevant range of activity by reference to economic and commercial realities and the policy of the statute. To the extent that it must serve statutory policy, the identification will be evaluative and purposive as well as descriptive. (Singapore Airlines; as also cited by Barker J in Power New Zealand Ltd v Mercury Energy Ltd [1996] 1 NZLR 686 at 705). (d) Generally a market will be identified by reference to the activities of those engaged in commerce, the structures underlying their activities and the perceived susceptibility to change in the medium-term future. (e) The identification of markets involves value judgements about which there is room for legitimate differences of opinions. (Queensland Wire)(f) A market can exist where there is potential for, but no actual, competition. A market may also exist for particular goods even if there is no supplier of, nor trade in, those goods at a given time. [282] The need, in the market designation process, for a purposive approach, as referred to in Port Nelson (CA), and the role of substitutability on the demand and supply sides, were also matters emphasised by each of the economic experts who provided evidence. As usefully summarised by Mr Farmer in his closing submissions, that evidence was to the effect that: (a) The market definition exercise is not an end in itself, but is rather a tool for analysing the competitive issues of interest; (b) Substitution, in either demand and/or supply, is what defines the area of close competition; and (c) The principle of a small but significant and non-transitory increase in price (SSNIP) has been adopted as the pedagogical tool for determining market boundaries by most competition authorities.Dominance/substantial market power[283] The time period relevant to the Commission's s 36 cause of action straddles the amendment to that section which came into force on 26 May 2001. The Commission separately pleads breaches of s 36 by reference to both expressions of the section. It is necessary therefore for the Court to have regard to both of those expressions. [284] As relevant to these proceedings, prior to 26 May 2001 s 36 provided as follows:36 Use of dominant position in a market(1) No person who has a dominant position in a market shall use that position for the purpose of – (a) Restricting the entry of any person into that or any other market; or(b) Preventing or deterring any person from engaging in competitive conduct in that or in any other market; or (c) Eliminating any person from that or any other market.[285] The Act defined, or otherwise provided interpretative guidance as regards, a number of the terms used in s 36:(a) Market: s 3(1A): Every reference in this Act to the term "market" is a reference to a market in New Zealand for goods and services as well as other goods and services that, as a matter of fact and commercial common-sense, are substitutable for them. (b) Dominant Position: s 3(8): For the purposes of ss 36 and 36A of this Act, a dominant position in a market is one in which a person as a supplier or an acquirer of goods or services either alone or together with any interconnected body corporate is in a position to exercise a dominant influence over the production, acquisition, supply, or price of goods or services in that market and for the purposes of determining whether a person is in a position to exercise a dominant influence over the production, acquisition, supply, or price of goods or services in a market regards shall be had to— (a) The share of the market, the technical knowledge, the access to materials or capital of that person or that person together with any interconnected body corporate: (b) The extent to which that person is constrained by the conduct of competitors or potential competitors in that market: (c) The extent to which that person is constrained by the conduct of suppliers or acquirers of goods or services in that market. (c) Purpose: s 2(5)(b): A person shall be deemed to have engaged, or to engage, in conduct for a particular purpose or a particular reason if – (i) That person engaged or engages in that conduct for that purpose or reason or for purposes or reasons that included or include that purpose or reason; and (ii) That purpose or reason was or is a substantial purpose or reason. (d) Substantial: s 2(1A): In this Act (except ) substantial means real or of substance.[286] On and from 26 May 2001, the section – again as relevant – provides as follows:36 Taking advantage of market power (2) A person that has a substantial degree of power in a market must not take advantage of that power for the purpose of – (a) restricting the entry of a person into that or any other market; or (b) preventing or deterring a person from engaging in an competitive conduct in that or any other market; or (c) eliminating a person from that or any other market.[287] The Act continues to provide definitions or interpretational guidance – as before – for the terms "market", "purpose" and "substantial". The definition of "dominant position" was deleted along with the discontinuance of that term as from 31 May 2001. Its replacement, the term a "substantial degree of power", is not defined or otherwise explained and the definition of "substantial" provided in the Act expressly does not apply to s 36. [288] Taken overall therefore: [a] prior to 31 May 2001 the prohibition was on the use of a dominant position in a market for a prohibited – subsection (a), (b) or (c) – purpose; whereas [b] on and from that date the prohibition is on taking advantage of a substantial degree of power in a market for a prohibited purpose. [289] Neither party in this case argued that the two tests were in any material way different. [290] We recognise that, prior to the enactment of the 2001 amendments, there had been intensive consideration of the various expressions by the Court of the "dominance" standard. The discussion, by McGechan J in Commerce Commission v Port Nelson Ltd (1995) 6 TCLR 406 (HC), of the Court of Appeal discussion inTelecom Corp of NZ Ltd v Commerce Commission [1992] 3 NZLR 429 ("the AMPS- A decision"), and the Court of Appeal's subsequent decision in CommerceCommission v Southern Cross Medical Care Society (2001) 10 TCCR 269, were important parts of that process. [291] Notwithstanding what was categorised by some as the "dictionary" approach taken in AMPS-A, the Court of Appeal in Southern Cross, prior to the 2001 Amendment, commented on dominance in terms which showed no hesitation in embracing an economic approach. [292] That approach: (a) reflected Richardson J's concerns in AMPS-A to take appropriate account of the plain meaning of the word dominance, and to avoid unhelpful synonyms; (b) recognised at the same time the economic context, and statutory purpose, of s 36 and therefore incorporated appropriate economic principles into the understanding and application of the terms; and (c) was consistent with the approach taken by the Australian Courts in applying the concept of substantial market power. [293] Turning now to the formulation of "a substantial degree of power in a market" adopted when the Commerce Act was amended in 2001, we note immediately that this is the same formulation which was inserted into s 46 of the Australian Trade Practices Act in 1985. Australian authority on that section therefore becomes directly relevant. [294] In terms of the words of the section, and that authority, there is no doubt that the substantial degree test is cast explicitly in the language of market power. Accordingly there can no longer in our view be any dispute as to whether the Court should apply an ordinary meaning approach, or one based on the economic principles that have been developed over time. Given our conclusion that inSouthern Cross the Courts had returned to an interpretive approach based on economic principles, this clarification of the interpretive approach does not, in our view, move matters very far.[295] Rather, the issue which is left is whether the threshold of market power represented by "substantial degree" is in any way different to the threshold represented by dominance, as so understood. [296] There are a number of indicators in the extrinsic materials surrounding the New Zealand amendments that suggest that substantial degree of power in a market was intended to establish a lower threshold than dominance. We refer to a Cabinet paper of 28 March 2000 from the Acting Minister of Finance referred to in Gault on Commercial Law at CA36.04, to the Parliamentary debates on the amending bill, for example, at [91] NZPD 8734, and to a discussion document released by the Ministry of Commerce in April 1999. [297] That said, it seems to us that the primary concern of Parliament was the high threshold, and even more particularly the interpretive methodology, represented byAMPS-A. As we have said, in Southern Cross the Court took an economic approach on the interpretive question which, to a certain but not clearly determined extent, impacted also on the threshold issue. It may be, therefore, that by the time the new wording was introduced by Parliament, a lesser degree of change to the status quo than had been anticipated was effected. [298] Therefore, taken overall, and whilst it may in our view be possible to conclude that substantial degree of market power was originally intended to set a slightly lower threshold than dominance, in general the question of degree is so slight as to prevent clear enunciation. In this case, and as accepted by both parties, we will therefore proceed on the basis that there is no material difference between the two tests and will analyse the factual situation accordingly. We refer variously to dominance and a substantial degree of market power similarly. [299] For the period until 26 May 2001, the "dominant position" definition in s 3(8) is directly applicable, including its references to: (a) market share; and(b) the constraints on the allegedly dominant person provided by the conduct in the relevant market of actual and potential competitors, and of suppliers or acquirers of goods and services. [300] We think it is accepted that, although the s 3(8) definition of a dominant position has been repealed, market share and the effect of potential competition continued to be important considerations after 26 May 2001. [301] The final factor that we think we should refer to, as regards determining whether a person is dominant or has a substantial degree of power in a market, is the matter of barriers to entry. In terms of the applicable general legal principles, the defendant put particular emphasis on the significance of barriers to entry. [302] As to barriers to entry, in Queensland Wire it was said:A large market share may well be evidence of market power but the ease with which competitors would be able to enter the market must also be considered. It is only when for some reason it is not rational or possible for new entrants to participate in the market that a firm can have market power: see Continental Can. There must be barriers to entry. As Professor F. M. Scherer has written, "significant entry barriers are the sine qua non of monopoly and oligopoly, for sellers have little or no enduring power over price when entry barriers are nonexistent": Scherer, Industrial Market Structure and Economic Performance, 2 nd ed. (1980), p 11. (pp 189-190)[303] The importance of barriers to entry has been recognised in New Zealand: seeMagic Millions Ltd v Wrightson Bloodstock Ltd [1990] 1 NZLR 731 at 755, Carter Holt Harvey Building Products Group v Commerce Commission (2001) 10 TCLR 247 ("Carter Holt (2001)") at [22], and Southern Cross at [68] and [69]. [304] From those authorities we take the following propositions: (a) Market share is not the sole determinant of the presence or absence of dominance or market power. The most that can be said is that dominance is frequently attended by a substantial market share but all other relevant factors must be brought to account. For example, a substantial market share without barriers to entry will seldom, if ever, be indicative of dominance. (Magic Millions)(b) Although the concept of barriers to entry is not mentioned expressly in s 3(8), it is generally accepted that, because potential competition can be an important constraint upon market participants, a careful examination of barriers to entry to the relevant market is an important part of the test for dominance. (Carter Holt (2001)) (c) Market share is relevant to the level and significance of market power but it is not in itself the determinant of market power. What level of market power a firm has, as a result of its market share, will depend substantially on the level of barriers to entry and expansion which apply to the market. If the barriers are low, a high market share is unlikely to result in a substantial degree of market power. Conversely, if the barriers are high, a high market share is likely to lead to such a result. The level of market share and the level of market power have no direct relationship in themselves. (Southern Cross.)Use/taking advantage[305] It was accepted in this case that the test for use and taking advantage is the so-called "counterfactual" test, as outlined by the Privy Council in Telecom v Clear Communications. There the Privy Council, having referred to the possibility of inferring purpose from use, but of the danger in inferring use from purpose, set out what it considered to be the correct approach to determining whether a person with a substantial degree of market power has used that power:Both the High Court and the Court of Appeal proceeded on the basis, with which Their Lordships agree, that if the terms Telecom were seeking to extract were no higher than those which a hypothetical firm would seek in a perfectly contestable market, Telecom was not using its dominant position. (at 403, lines 17-21) In Their Lordships' view it cannot be said that a person in a dominant market position "uses" that position for the purposes of s 36 unless he acts in a way which a person not in a dominant position but otherwise in the same circumstances would have acted. (at 403, lines 53-56)[306] Although this had not always been the accepted approach, any uncertainty in this area would appear, however, to have been resolved by the Privy Council on appeal from Carter Holt (2001). [307] In Carter Holt Harvey v Commerce Commission [2006] 1 NZLR 145 ("Carter Holt (2006)"), the critical question was use. The Privy Council noted that Carter Holt had framed its appeal as being that although the counterfactual test was referred to by both the High Court and the Court of Appeal, neither Court had applied that test to the facts. As a result the Court had reached a conclusion contrary to what s 36 required, that a trader in a dominant position was under a duty to act in a way which would not be required of a trader who was non-dominant. [308] In reasserting the legitimacy and necessity of the counterfactual test, their Lordships set out its general rationale in the following terms:[51] The starting point is to be found in the principle that competition is in the public interest. It lies at the heart of the 1986 Act. The object of s 36, like its counterpart in Australia, is to protect the interests of consumers. It is predicated on the assumption that competition is a means to that end:Queensland Wire Industries Pty Ltd v Broken Hill Pty Co Ltd at p 191 per Mason CJ and Wilson J. A dominant firm is as free to compete in the market as a firm that is non–dominant, so long as it does not act in an anticompetitive manner by abusing its position of dominance. With this in view, the section is carefully worded. The word "use" requires that a causal relationship is shown between the conduct which is alleged against the dominant firm and its dominance or market power. Only if that connection is shown can it be said that its conduct is a use of that dominance: Natwest Australia Bank Ltd v Boral Gerrard Strapping System Pty Ltd (1992) 111 ALR 631 at p 637 per French J. [52] It follows that if a dominant firm is acting as a non–dominant firm otherwise in the same position would have acted in a market which was competitive it cannot be said to be using its dominance to achieve the purpose that is prohibited. That is the basis on which the counterfactual test is founded. As Wilcox J said in Eastern Express Pty Ltd v General Newspapers Pty Ltd (1991) 103 ALR 41 at p 65, it would be surprising if Parliament intended to proscribe conduct by a company with sufficient resources to compete effectively. Something more than that is required.[309] In our view, the position therefore is clear. The counterfactual test is, of necessity, to be undertaken and is the test of use. [310] We do note, however, that – as recognised by the minority of the Privy Council in Carter Holt (2006) – that application of that test is not always without itsdifficulties. This case is an example of that, as was acknowledged by both Professor Sweeney and Dr Pleatsikas when they gave evidence. [311] For completeness, we note that the 2001 amendments to s 36 substituted the phrase "takes advantage of" for "use". That is the phrase that is used in the Australian legislation (since 1986) and which formed the basis for the counterfactual test in Queensland Wire. As the Privy Council affirmed the counterfactual test in part reliance on Queensland Wire, we think the counterfactual test enunciated inCarter Holt (2006) continues to be the law in New Zealand at present, notwithstanding the statutory language applicable when that case was decided referred to "use" rather than "takes advantage". We acknowledge, however, the relevance of High Court of Australia authority subsequent to Queensland Wirewhich has arguably taken a more expansive approach to the counterfactual test under the "takes advantage" element of the section. We turn now to consider those authorities as they were advanced to us by Mr Farmer. [312] We mention, at this point, the authority referred to by Mr Farmer in support of his proposition regarding the relevance of BOPE's legitimate business rationale as regards the test of use. In his submission, the Australian decisions of Australian Competition and Consumer Commission v Boral Ltd (1996) 166 ALR 409 andMelway Publishing v Hicks [2001] HCA 253 at 269 para 66, recognised that if the impugned conduct had a business rationale that was a factor which pointed against any finding that the conduct constitutes a taking advantage of market power. Mr Farmer referred to the following statement of Heerey J in Boral:If a firm with no substantial degree of market power would engage in certain conduct as a matter of commercial judgment, it would ordinarily follow that a firm with market power which engages in the same conduct is not taking advantage of its power. (at [158])[313] Mr Farmer went on to submit that those passages, and the requirement for use/taking advantage, generally recognised that when a firm has a legitimate business rationale for refusing to deal, which it would adopt whether it had substantial market power or not, then it could not be found to be an abuse of that market power if the firm does so refuse to deal.[314] Dr Pleatsikas would appear to have, from an economic perspective, a similar understanding. In his witness statement he outlined a general economic approach that could be used to assess whether a firm has taken advantage of a substantial degree of market power or dominance. [315] Such an economic test had at least three elements. First, one must establish that the firm has (or had) substantial market power. Second, assuming the first element has been established, one must identify whether the conduct in question is consistent with behaviour that would likely have occurred if the firm did not have substantial market power. Third, if the observed conduct is inconsistent with conduct that would likely have occurred if the firm operated in a workably competitive market, one must determine whether such conduct is causally related to some factor other than substantial market power – for example, regulatory policy. If it was related to some other factor, then the pertinent conduct may not be the result of taking advantage of a substantial degree of market power. [316] It is that third element that appears similar to Mr Farmer's "legitimate business purpose" rationale. [317] Whilst we acknowledge the relevance of that type of inquiry, in our view the core question remains whether the firm would rationally engage in the conduct in question if it did not enjoy dominance or possess a substantial degree of market power. We think there may be some risk of begging the question in the approach taken in the cited extract from BORAL Besser. That is, it must be accepted that conduct which may be legitimate for a firm not possessing market power (and, given that it was undertaken by such a firm, the presumption has to be that there is a profit maximising business rationale for such conduct), can nevertheless be illegitimate if carried out by a firm enjoying dominance and/or a substantial degree of market power, for an illegitimate purpose. [318] We also note in this case that there was some discussion by counsel of the could/would debate regarding the counterfactual test. We do not find that debate particularly enlightening. In our view, what the test addresses is the question of whether a firm with dominance and/or a substantial degree of market power would rationally, in the sense that it could and could do so profitably – over a relevant timeframe – engage in the impugned conduct if it did not possess that dominance or substantial market power.Purpose[319] The final element of s 36 requires the presence of a prohibited purpose. That is, a person must not take advantage of the substantial market power for the purpose of: [a] Restricting the entry of any person into that or any other market; or [b] Preventing or deterring any person from engaging in competitive conduct in that or in any other market; or [c] Eliminating any person from that or any other market. [320] We first note that for these purposes: (a) A person shall be deemed to have engaged, or to engage, in conduct for a particular purpose or a particular reason if – (i) That person engaged or engages in that conduct for that purpose or reason or for purposes or reasons that included or include that purpose or reason; and (ii) That purpose or reason was or is a substantial purpose or reason (s 2(5)(b)); and (b) In this Act (except ) substantial means real or of substance. (s 2 (1A)) [321] The purpose requirement was discussed in the following terms in Telecom v Clear Communications:(2) Before considering the "use" requirement, it is convenient to consider the "purpose" requirement. If a person has used his dominant position it is hard to imagine a case in which he would have done so otherwise than for the purpose of producing an anti-competitive effect; there will be no need to usethe dominant position in the process of ordinary competition. Therefore, it will frequently be legitimate for a Court to infer from the defendant's use of his dominant position that his purpose was to produce the effect in fact produced. Therefore, as the Court of Appeal in the present case accepted, use and purpose, though separate requirements, will not be easily separated. Although it is legitimate to infer "purpose" from use of a dominant position producing an anti-competitive effect, it may be dangerous to argue the converse ie that because the anti-competitive purpose was present, therefore there was use of a dominant position. Telecom's relationship with Clear is not only that of supplier and customer; Clear will also be Telecom's competitor in the provision of telecommunications in the CBDs. It is unavoidable that, as a competitor, Telecom will be seeking in one sense to "deter" Clear from competing successfully. A monopolist is entitled, like everyone else, to compete with its competitors: if it is not permitted to do so it "would be holding an umbrella over inefficient competitors". it does not follow from the existence of Telecom's anti-competitive purpose that Telecom has used its dominant market position. In Their Lordships' view that is the critical question. (at 402-403)[322] Their Lordships can, therefore, be seen to have taken both an objective and a subjective approach to the identification of the necessary purpose, with the latter being relied upon to support conclusions reached based on the former. [323] The possibility that a prohibited purpose may exist alongside a legitimate purpose – recognised by s 2(5)(b) – was referred to in Port Nelson (CA) in the following passage:The distinction Mr White sought to draw between the purpose of PNL in protecting its own pilotage business and the purpose of deterring or eliminating TBMP is fine indeed when the means adopted to achieve that purpose was to refuse to supply tugs where TBMP pilots were engaged. The short answer is that each plainly was a substantial purpose. There was ample evidence, as reviewed in the judgment of McGechan J, to support the finding as to PNL's purpose. (at 578, lines 34-39)[324] Remarks to a similar effect appear in the decision of Tipping J in Magic Millions where, referring to s 2(5)(b) and the statutory definition of substantial His Honour noted:Human conduct is often motivated by more than one reason or purpose. (at 763, lines 45-46)[325] Based on that discussion, we think the primary enquiry is an objective one, but that evidence of subjective statements of purpose and intention can be relevant.We also note, in this context, the significance of the statutory recognition provided by s 2(5)(b) that a prohibited purpose need not be the sole purpose.Section 27[326] Section 27 relevantly provides:27 Contracts, arrangements, or understandings substantially lessening competition prohibited(1) No person shall enter into a contract or arrangement, or arrive at an understanding, containing a provision that has the purpose, or has or is likely to have the effect, of substantially lessening competition in a market. (2) No person shall give effect to a provision of a contract, arrangement, or understanding that has the purpose, or has or is likely to have the effect, of substantially lessening competition in a market.[327] Section 27 is a broad provision designed to prevent cartel conduct or collusion between firms or individuals within a market. It is broad in the sense that not only formal contracts and agreements are captured by the provision, but also less formal arrangements or understandings between firms or individuals. It is also broad in the sense that it forbids collusive behaviour that has the purpose of substantially lessening competition, notwithstanding it may not achieve that end, or that is likelyto substantially lessen competition, even if it has not in fact done so. [328] The Commission's case rests on the allegation that certain parties entered into an arrangement, or arrived at an understanding, a provision of which was that BOPE would not lease its meters to potential competitors. BOPE admitted that, at all relevant times (1999 to date) it generally did not enter into meter leasing agreements, and had a policy of not doing so. Furthermore, it was accepted that the BOPE Joint Venture parties made that decision, and adopted that policy, and that BOPE gave effect to it. [329] Therefore the elements of the s 27 causes of action relying on the existence of an arrangement or understanding, and BOPE having given effect to that arrangement or understanding, were not in dispute, and do not require further discussion here.Purpose[330] The wording of s 27 indicates that it is the allegedly anti-competitive provision itself that must have the purpose of substantially lessening competition, rather than the parties that entered into the arrangement or understanding: Port Nelson (CA) at 563. This is in contradistinction to the purpose assessment under s 36, which attaches directly to the defendant and its conduct. However, the inquiry as to the purpose of the provision must be undertaken having regard to the deeming provision in s 2(5) of the Act, which reads:(5) For the purposes of this Act—(a) a provision of a contract, arrangement or understanding, or a covenant shall be deemed to have had, or to have, a particular purpose if—(i) the provision was or is included in the contract, arrangement or understanding, or the covenant was or is required to be given, for that purpose or purposes that included or include that purpose; and (ii) that purpose was or is a substantial purpose.[331] That deems a provision to have the purpose if it was required to be included (by the person or persons so requiring) for that purpose. The Court of Appeal inPort Nelson held that this indicates plainly that not all parties need be shown to share the purpose; a unilateral purpose is sufficient. [332] We also note that the purpose in question need not be the only purpose behind the inclusion of the provision, as long as it is a substantial purpose. Equally, however, the purpose must be substantial, which by s 2(1A) is defined as "real or of substance". This is the formulation that was adopted by McGechan J in Union Shipping NZ Ltd v Port Nelson Ltd [1990] 2 NZLR 662 at 707 and by Cooke P inTui Foods Ltd v New Zealand Milk Corporation Ltd (1993) 5 TCLR 406 at 410. [333] On the further question of whether the purpose is to be ascertained subjectively or objectively there is some dispute. The Court of Appeal in Port Nelson noted that in most cases there will be very little difference between ascertaining subjective purpose by inference from what was said and done and ascribing objectively a purpose from evidence of what was said and done.Furthermore, the Court observed that it is difficult to see how the purpose of a provision can be ascertained subjectively. On the other hand, the purpose for which a provision is included (s 2(5)(a)) could be ascertained subjectively. The Court ultimately preferred to simply adopt the comments by Cooke P in Tui Foods at 409 as follows:I am disposed to think that, if a purpose is discernable on the face of a contract or arrangement having regard to the express terms considered in the light of any relevant surrounding circumstances, such a purpose will qualify under the statute. That might be described as an objective approach. But it is at least conceivable that there may also be cases where, although the purpose is not so apparent, it can be shown by evidence dehors a contract or arrangement that the intention of the party who sought the inclusion of the relevant provision was of a kind falling within the prohibition in s 29, and it may be that in such a case what may be called a subjective test is sufficient.[334] The matter was revisited by the Court of Appeal in ANZCO Foods Waitara Ltd v AFFCO New Zealand Ltd [2006] 3 NZLR 351. Glazebrook J traversed the discussions in Tui Foods and Port Nelson (CA) and opined that the matter was really an evidential rather than a substantive question. In her view, the test for s 27 is an objective one but evidence of subjective purpose can be adduced and taken into account in assessing objective purpose. [335] The issue of objective versus subjective analysis of purpose also has implications for the relationship between 'purpose' and 'effect' or 'likely effect'. William Young J pointed out in ANZCO that, on a rigorously objective approach, the Judge's conclusions as to effect and likely effect would be determinative of purpose, on the basis that a provision, objectively construed, must have the purpose of achieving the effect that it does in fact achieve, or is likely to achieve. William Young J therefore observed that a better approach might be to assess the purpose of a provision objectively by reference to its actual or likely effect should it be acted upon. That said, however, a number of indications in the Act militated against that approach. On a purely objective view, there would be little obvious reason for the legislature to introduce the concept of purpose at all, because on an objective standard the words effect and likely effect would do all the work. Further, a purely objective approach to purpose is not consistent with the s 2(5)(a)(i) definition, which deems the subjective purpose of the parties to be the purpose of the provision in some cases. Finally, William Young J made the point that the assessment of anti- competitive effect in circumstances of imperfect information is necessarily uncertain.It is therefore sensible that the law prohibits a provision the purpose of which is anti- competitive, notwithstanding no actual or likely anti-competitive effect. Accordingly, the possibility of taking a subjective approach to purpose (in some cases) demonstrates that purpose is to be determined independently of effect or likely effect, and may conflict with findings of effect or likely effect. [336] Those comments must also, however, be read in light of Glazebrook J's discussion of purpose in the same case (at [250] to [265]). As already mentioned, Glazebrook J took the view that purpose was primarily an objective question but that subjective evidence might also be taken in making the objective assessment. Glazebrook J therefore agreed with William Young J that purpose is not the same as effect or likely effect. However, the purpose which must be proved is one that has, as an end in view, the substantial lessening of competition. Accordingly, Glazebrook J held that where it is obvious that substantial lessening of competition could not be achieved if the provision were implemented then, assessed objectively, the provision cannot have that purpose. Glazebrook J would therefore restrict the subjective assessment to "borderline cases" where there is evidence of a subjective anti-competitive purpose and the evidence is equivocal as to anti-competitive effect. Where there is clearly no actual or likely anti-competitive effect there is limited utility in embarking on a subjective assessment of purpose. [337] William Young J responded to that analysis at [154], recording his agreement that in assessing whether a provision has the purpose of substantially lessening competition, the question whether it could have such an effect is necessarily relevant. However, he did not go as far as to hold that the impossibility of the provision having an anti-competitive effect necessarily controls the application of the purpose test. [338] Anderson P agreed with Glazebrook J on the essential point on the appeal: that the Judge had not erred in finding there was no anti-competitive purpose. However, he did not express an opinion on the disagreement between Glazebrook and William Young JJ as to whether an anti-competitive purpose can be identified in circumstances where the provision was incapable of having that effect. However, the President did hold that there can be an anti-competitive purpose without itresulting in an anti-competitive effect where, for example, the parties are mistaken as to the likely effect. [339] Ultimately, we think the difference in approach is one of degree more than substance. Both William Young J and Glazebrook J are in agreement that: [a] Purpose is primarily (and perhaps preferably) to be assessed objectively, but subjective evidence of an anti-competitive purpose will be relevant to the inquiry. [b] As a result, purpose cannot be equated with actual or likely effect – there may be a prohibited purpose notwithstanding no actual or likely effect of substantially lessening competition. [c] However, where the provision in question is incapable of having the effect or likely effect of substantially lessening competition, that will be relevant, and may be determinative, of the purpose inquiry. [340] Beyond that, the Judges' disagreement turns on the weight they would give to those competing factors in their consideration of the issue of purpose. In most cases, therefore, the outcome will more likely turn on the nature and quality of the evidence rather than the strict application of either an objective or subjective approach. That likelihood was acknowledged in both Tui Foods (at 409) and in the Court of Appeal's decision in Port Nelson (at 564). [341] The final point we make in regards to purpose is that the different tests for purpose under s 36 and s 27 mean that it will not be appropriate to reason directly from a finding of prohibited purpose under s 36 to prohibited purpose under s 27. InUnion Shipping McGechan J found that, although the cause of action under s 36 was made out, the plaintiff had not fulfilled the onus of establishing purpose under s 27.Effect / likely effect[342] Whether a provision has the actual or likely effect of substantially lessening competition is a question of fact: ANZCO at [135]. Actual results are relevant whenone is considering the effect of a provision, whereas likely effect involves considering results that may happen. [343] In order to establish the effect of substantially lessening competition, there must be a causative link between the impugned provision and the substantial lessening of competition. Concepts of causation may differ, but McGechan J in Port Nelson referred to the "direct and immediate effect" of the provision: at 429. The Court of Appeal did not differ from him on that point. That is also the test under the Australian Trade Practices Act, per Deane J in Re Application by Concrete Carters Association (Victoria) (1977) ATPR 40-042. That said, the reference to an "immediate" effect denotes an effect that follows directly from the provision without an intervening cause, rather than an effect which occurs immediately in time upon the promulgation or implementation of the provision. That follows from s 2(3) of the Act, which provides that an anti-competitive provision is unenforceable at any time at which it can be demonstrated that it will have the prohibited effect or likely effect, notwithstanding that at an earlier time the provision did not have that effect or likely effect. Gault on Commercial Law suggests at CA27.12 that that provision applies equally to s 27(2) in relation to giving effect to a prohibited provision. [344] The authority on what constitutes a likely effect is the Court of Appeal decision in Port Nelson. The Court held that "the appropriate level is that above mere possibility but not so high as more likely than not and is best expressed as areal and substantial risk that the stated consequence will happen" (emphasis added).Substantially lessening competition[345] A number of definitions in the Act are relevant. First, pursuant to s 3(1), competition means workable and effective competition. [346] Secondly, lessening of competition includes references to hindering or preventing competition: s 3(2). McGechan J in Port Nelson made the following comments on the extended definition of the term 'lessening', which were later affirmed in ANZCO:One may or may not, normally, 'lessen' when one 'hinders'. The word 'hinder' (Shorter Oxford English Dictionary (3rd ed), vol 1, p 865) coverssenses which include 'do harm to' and 'prevent'; but also 'to keep back; impede, deter, obstruct', and 'delay or frustrate action, by an obstacle or impediment'. One can 'hinder' by merely delaying or obstructing for the immediate time. That, no doubt, is the extended sense intended. There would be little point, otherwise, in the extension. The inclusion of 'hindrance', in that sense accords entirely with the overall policy of the Act to remove obstacles in the way of free competition.[347] Thirdly, s 2(1A) provides that substantial in this context means real or of substance. McGechan J in Port Nelson paraphrased this as meaning something more than insubstantial or nominal. That approach was recently reaffirmed by the Court of Appeal in ANZCO. [348] Finally, ss 3(5) and (7) state that s 27 will be breached in circumstances where a connected set of provisions or a connected range of conduct taken as a whole have the actual or likely effect of substantially lessening competition. [349] The question whether a substantial lessening of competition is an actual or likely effect is determined by means of a counterfactual analysis comparing: [a] in the case of an allegation of 'effect', the actual state of competition in the pleaded market with the state of competition that would have obtained in the absence of the impugned provision (ANZCO at [150]); [b] in the case of an allegation of 'likely effect', the likely state of competition if the provision is given effect to with the likely state of competition if it is not (Commerce Commission v NZ Bus (2006) 11 TCLR 679 at [121]). [350] An oft-cited analysis of the statutory test is that propounded in Dandy Power Equipment Pty Ltd v Mercury Marine Pty Ltd (1982) 64 FLR 238, adopted in Tru Tone at 362, and re-affirmed recently in ANZCO as follows:To apply the concept of substantially lessening competition in a market, it is necessary to assess the nature and extent of the market, the probable nature and extent of competition which would exist therein but for the conduct in question, the way the market operates and the nature and extent of the contemplated lessening. To my mind one must look at the relevant significant portion of the market, ask oneself how and to what extent there would have been competition therein but for the conduct, assess what it leftand determine whether what has been lost in relation to what would have been, is seen to be a substantial lessening of competition. I prefer not to substitute other adverbs for 'substantially'. 'Substantially' is a word the meaning of which in the circumstances in which it is applied must, to some extent, be of uncertain incidence and a matter of judgment. There is no precise scale by which to measure what is substantial. I think in the context, particularly the penalty and other remedies for contravention of the Act, and the nature of trade which is the subject of the Act, the word is used in a sense importing a greater rather than a less degree of lessening. Accordingly in my opinion competition in a market is substantially lessened if the extent of competition in the market which has been lost, is seen by those competent to judge to be a substantial lessening of competition. Has competitive trading in the market been substantially interfered with? It is then that the public as such will suffer.[351] Glazebrook J made a number of other observations in ANZCO relevant to the requirement of substantial lessening of competition. She held at [240] that 'substantially' is used in a relative rather than absolute sense. Thus, the issue of whether there is a substantial lessening of competition must be assessed in terms of the particular circumstances, including the market involved in the particular case. Further, the concept of competition is distinct from the concept of rivalry. Rivalry emphasises the behaviour of individual buyers and sellers and the independent striving for custom. Competition is a process that emphasises the structural conditions under which rivalry occurs. The Commerce Act is not concerned with protecting individual competitors but with the competitive process: [242]. Analysis of the competitive process may require examination of market structure elements (see Re Queensland Co-operative Milling Association Ltd (1976) 8 ALR 481 at 516), behavioural aspects, or alterations in market power resulting from the practice in question: [243] to [244]. The Court is unlikely to find a substantial lessening of competition in circumstances where the effects operate in the short term only: [247].Market[352] Generally speaking, the same market definition issues that have been earlier discussed in relation to s 36 apply in the case of s 27. The primary inquiry is into substitutability as a matter of fact and commercial common sense.DiscussionOverview of analysis[353] We turn now to apply ss 36 and 27 to the facts of this case. [354] We discuss s 36 first. We observe at the outset that s 36 is a broadly stated provision that prohibits a range of conduct. Within the bounds of any given factual situation there may be a number of different ways of characterising the case in terms of the elements of s 36. In terms of the present allegations, we further note that there was a degree of disconnect between the case as pleaded, as argued in the submissions at trial, and as substantiated in the evidence. As a result, in the course of writing this judgment we have wrestled with a number of different approaches. The result we have ultimately reached rests in large part on our view of some necessary implications of the way the case was pleaded and argued at trial. [355] There is an unusual feature of this case. The core conduct of BOPE impugned by the Commission – BOPE's refusal to lease its meters – is said by the Commission to have been the reason for the existence of, or to have contributed to the existence of, the pleaded local markets and BOPE's dominance in them. At the same time, that conduct is alleged to have constituted the use of that dominance. [356] During the course of the trial, it became apparent that two characterisations were open on the pleadings. Those were: [a] As a case involving an unlawful refusal to supply; and [b] As a case involving conduct by a firm which created barriers to entry by raising the costs potential rivals to that firm faced to compete with it. [357] We conclude that the core conduct of BOPE which the Commission impugned, namely BOPE's – admitted – policy of not leasing its in situ meters, was neither an unlawful refusal to supply nor did it – in any relevant sense – raise costsfaced by its rivals. To that extent, therefore, the Commission did not establish its case against BOPE under s 36. [358] We go on to discuss two further alternative characterisations of the case that were suggested by the evidence and the submissions. We conclude that those characterisations were not open on the pleadings, or were not the focus of the Commission's case, such that it would be prejudicial to BOPE to decide the case on that basis. [359] Under the first of those alternatives, we note that the Commission identified as relevant BOPE's alleged conduct in obstructing efforts by potential competitors to enter the market, principally by refusing to enter, or delaying entering, into negotiations with them, by exaggerating to its customers the inconvenience of switching to a new supplier and by increasing those switching costs by not willingly co-ordinating the processes required for such switches to occur. We find, as a matter of fact, that BOPE did engage in such conduct. [360] The Commission did not, however, plead that this conduct constituted use or taking advantage. BOPE argued, strongly, therefore that we could not find that the Commission had established its s 36 case by reference to that behaviour. We agree with that submission. [361] We consider, nevertheless, what the position would be if the Commission had effectively pleaded use as regards this conduct. We conclude that, even in that circumstance, the Commission's case would fail. The additional costs that behaviour imposed, which in our view would also fail the counter-factual test, were not shown to be of a sufficient order of magnitude to delineate as required the pleaded local market or to give rise to relevant market power. We explain why we consider that to be the appropriate inquiry. [362] Under the second of our two further alternatives, we consider the possibility that BOPE might have been dominant in a relevant market independent of its cost- raising behaviour. From our conclusion that such behaviour would not pass the counter-factual test, but would appear – for at least part of the period under review – to have been engaged in by BOPE, it is possible to infer that BOPE had some degreeof market power. While this does not necessarily imply dominance or a substantial degree of market power, it is nevertheless not inconsistent with that; accordingly, we consider also whether BOPE may have had such dominance independently of the conduct challenged by the Commission in its case. We express the tentative view that this would appear to have been a possibility, on the basis that the industry reforms intended to facilitate competition were not successful in depriving incumbent firms of the market dominance they possessed prior to 1 April 1999. On that basis, the conduct we recognise that would fail the counter-factual test could constitute use of that dominance. In this context, we also record our view of BOPE's purposes as would then be relevant for a s 36 inquiry. [363] This was not, however, the case the Commission pleaded. It is, in fact, an analysis which is at odds with the view the Commission had expressed in Decision 333, and which formed the basis of the way the Commission argued its case before us. To take that approach would require a quite different factual inquiry than was conducted during the trial. We are of the view, therefore, that notwithstanding these tentative views that analysis is not a basis upon which we can find BOPE to be in breach of s 36. [364] We then consider the Commission's s 27 case. There the Commission's focus was solely on the agreement asserted regarding BOPE's policy of not leasing its in situ meters. No broader agreement was asserted. We consider that it follows necessarily from our discussion of s 36 that the Commission did not establish the pleaded local market. Nonetheless, we go on to consider what the position would be under s 27 had the Commission established the pleaded markets or, alternatively, a national market. We conclude that the Commission would have succeeded under s 27 had it established a local market on the basis that the no leasing policy was for the purpose of lessening competition.Section 36The unusual feature of this case[365] As noted in paras [276] and [277] above, there are four elements to be considered in determining whether a firm has breached s 36. Section 36 cases needto be analysed by reference to each of those elements. The inquiries as to the existence of the relevant market, and as to the alleged dominance of the firm in question, often involve similar considerations. Those inquiries will, however, generally be conceptually distinct from the inquiry as to whether or not the firm in question has used, or taken advantage of, that position of market power. [366] The Commission's case against BOPE was that – in the Horizon Area – the relevant markets existed, that BOPE had substantial market power in the Eastern Bay Electricity Metering Services market, and that BOPE took advantage of that market power, all by reference to BOPE's, admitted, policy of not leasing its installed meters to potential competitors. To the extent that the Commission did identify other sources of dominance and substantial market power, for example BOPE's position as the incumbent supplier in the Horizon Area, and BOPE's "obstructive behaviour", those were things which amplified the effect of the core behaviour, namely that of not agreeing to lease meters or, more generally, of not agreeing to enter into agreements for electricity metering services. Such other behaviour contributed, therefore, to BOPE's dominance. It was not, however, alleged to be use or taking advantage of that market power. [367] That same core behaviour – refusing to lease – is central to each of the first three inquiries under s 36 (market, dominance, use). It is, therefore, conceptually difficult to approach an analysis of the s 36 case presented by the Commission by analysing, in sequence, each of the elements of the section in light of the established facts. [368] In terms, furthermore, of any such analysis it is also apparent that if BOPE's conduct in refusing to lease is conduct that would have been available to it in a workably competitive market, then the Commissioner's case does not succeed. Such behaviour would not constitute use or taking advantage. [369] In addressing these analytical difficulties we think it is useful to recognise that the Commission's case against BOPE can be categorised, and in fact was categorised by the parties, in essentially two ways:[a] First, and as reflected in the Commission's pleadings and in a central legal argument advanced by Mr Farmer for BOPE, the case was that BOPE had refused to supply what was, in effect, an essential facility for retailers who wished to compete with BOPE. We think the Commission's central assertion, namely that it was economically infeasible for retailers to compete with BOPE if BOPE did not enter into agreements for electricity metering services, and implicitly on acceptable terms, supports that categorisation. [b] Secondly, and very much with reference to the evidence of the Commission's principal economic witness Dr Pleatsikas, the Commission's case was that BOPE had, through refusing access to its in situ deemed compliant meters, raised barriers to entry. Those barriers to entry were the additional costs competitors would incur, relative to the situation that would have existed if BOPE had leased its deemed compliant in situ meters to those potential competitors, in endeavouring to compete with BOPE for Horizon Consumers. Dr Pleatsikas explicitly referred to economic literature on raising rivals' costs, and strategic behaviour entered into for that purpose. Here the Commission was, we think, essentially asserting that BOPE was behaving in a manner that it would not have in a workably competitive market. That behaviour continued the existence of the local market that had faded away elsewhere, gave BOPE its dominance, its substantial market power - and at the same time constituted the use of that market power. [370] We therefore propose to analyse the Commission's case first by reference to each of those characterisations. In doing so, we will consider the implications of that analysis as regards the elements of s 36. We acknowledge that, under the Act, it is the application of s 36 itself that is required.Refusal to supply[371] BOPE's admitted conduct can be seen as BOPE refusing to lease, and subsequently sell, to its competitors its in situ, deemed MARIA compliant, meters.Access to such meters, implicitly on some acceptable terms, was necessary to enable firms to compete with BOPE for the sale of electricity to existing Consumers – see the Commission's central assertion discussed at [14] above. BOPE was, therefore, refusing to supply those meters whether by sale or lease. [372] BOPE responded by arguing that the Commission was, in effect, requiring BOPE to subsidise its competitors in a way which was contrary to both economic and legal principle. The Commission's case therefore fell outside the Act. The Commission was requiring BOPE to lease its meters to competitors, and to incur additional costs in so doing. As Professor Sweeney put it in his overview:Such a remedy would allow the new competitor the right to use meters that were owned by BoPE, although at some price. Because the competitors put substantial effort into trying to gain that right, one can infer that the right to use BoPE's assets would be beneficial to them. BoPE has determined that leasing their meters on the scale they envisioned would be unprofitable and would divert management attention from higher priority activities. Thus giving the competitors the right to use BoPE assets would be harmful to BoPE. In summary, the Commerce Commission's proposed remedy would harm the company that had invested in the metering assets – BoPE – and would provide benefits to the new competitors in the Bay area. The Commission's proposed remedy would tilt the otherwise level playing field in favour of the new competitors in the retail electricity market and to the detriment of BoPE or any other small incumbent (para 1.14). In short, the Commission's remedy would force a private sector owner of capital assets – meters – to make these available to its competitors on a favourable basis. When Mr Farmer referred to this as a requirement that BoPE provide a helping hand to his competitors, he correctly described the nature of the Commission's proposed remedy. Such a requirement would fly entirely in the face of competition policy as I know it in the US and in New Zealand. Competition policy has a fundamental purpose of protecting competition, not assisting individual competitors. The Commission's proposed remedy would assist BoPE's large national competitors and give them an unfair advantage. It would not promote competition. As an economist involved in competition cases here and in the US, I have reviewed much literature, including US Supreme Court decisions. The literature and the decisions make it clear that there are but very few exceptions to the principle that competition policy does not require a firm to offer its assets or services to its competitors. (paras 1.17-1.19).[373] Mr Farmer supported this line of argument by reference to what he categorised as essential facility cases (Auckland Regional Authority v Mutual RentalCars (Auckland Airport) Ltd [1987] 2 NZLR 647; Port Nelson (CA); Queensland Wire and Telecom v Clear Communications). He distinguished those cases, submitting that meters were not an essential facility of the type recognised in those cases. That they were not was demonstrated by the number of occasions on which competitors to BOPE in the Horizon Area, and other firms elsewhere in the industry (eg FreshStart), had or had been willing to install their own meters at Consumer's premises. [374] Mr Farmer argued that, for this Court to declare that BOPE had breached s 36 by refusing to lease its in-situ meters, we were required to find – in effect – that BOPE had a legal obligation to supply those meters if it were not to breach s 36. The Court would then be required to determine the appropriate price, because if the price charged for the mandatory supply was not subject to the Court's oversight, the obligation to supply could be rendered meaningless. We think the logic of those propositions must be recognised. Mr Farmer's further submission was that regulating prices is not a function the Court is well placed to perform. [375] Furthermore, BOPE was not in the business of leasing meters. Therefore to require it to lease its meters would be to establish an unfortunate precedent, contrary to the general market proposition that there is no duty to aid competitors by providing them with a helping hand, citing Olympia Equipment Co v Western Union Telegraph Co 797F 2d 370 (1986) 4 as referred to in Telecom v Clear Communications. [376] In terms of United States and Australian authority (Verizon Communications Inc v Law Offices of Curtis v Traki LLP, 134 SC 872, 881 (2004) and Australian Competition and Consumer Commission v Boral Ltd (1996) 166 ALR 409), the defendant had a legitimate business purpose, a business rationale for its conduct. Given the existence of that purpose or rationale, the defendant's conduct did not – and could not – constitute a breach of the section. [377] The remedy sought by the Commerce Commission would, in effect, involve the Court in the positive regulation of the defendant's conduct over time, a regulatory intervention in the metering sector of the electricity industry which bothParliament and the Electricity Commission had eschewed. This Court should do likewise. [378] Here, Mr Farmer referred to the outcome of the Telecom v Clear Communications litigation as illustrating the sterility of the orders sought by the Commission. In that litigation, he submitted, no Court had been able to deliver any workable price mechanism. He referred to an article by Professor W Pengilly, "Misuse of Market Power: the Unbearable Uncertainties Facing Australian Management" (2000), Trade Practices Law Journal, 56-78, at p 65 where the Professor had commented on the unsatisfactory outcome of that litigation, as well as expressing considerable criticism of the Queensland Wire decision. [379] The flavour of Mr Farmer's submissions are perhaps best captured in an extract from the judgment of Scalia J in the recent US Supreme Court decision on refusals to deal, (Verizon Communications Inc v Law Offices of Curtis V Trinko LLP, 134 S Ct 872, 881 (2004) at pp7-8 Scalia J) which Mr Farmer cited:Firms may acquire monopoly power by establishing an infrastructure that renders them uniquely suited to serve their customers. Compelling such firms to share the source of their advantage is in some tension with the underlying purpose of antitrust law, since it may lessen the incentive for the monopolist, the rival, or both to invest in those economically beneficial facilities. Enforced sharing also requires antitrust courts to act as central planners, identifying the proper price, quantity, and other terms of dealing – a role for which they are ill suited. Moreover, compelling negotiation between competitors may facilitate the supreme evil of antitrust: collusion. Thus, as a general matter, the Sherman Act "does not restrict the long recognised right of [a] trader or manufacturer engaged in an entirely private business, freely to exercise his own independent discretion as to parties with whom he will deal".[380] The Commission's response to this line of argument was to note first that Professor Sweeney's argument that BOPE competes on a level playing field, and that to grant the Commission the declaratory and injunctive relief it sought would tilt that playing field in favour of BOPE's competitors, or involve BOPE in subsidising those competitors, was essentially circular. It begged the question. We think there is real force in that submission. [381] In order to succeed, the Commission had to establish the ingredients of s 36 to the civil standard of proof as pleaded. It was not, however, required to do morethan this, and for example to prove that BOPE's actions had somehow harmed competition or was likely to create inefficiencies. The philosophy underpinning s 36, in New Zealand law, was that competition law has been harmed if the ingredients of that section are established. [382] It was, therefore, in the Commission's submission, neither necessary nor appropriate to import into New Zealand law concepts derived in the United States in a different statutory context. [383] As for the argument by reference to the alleged difficulties created by the relief sought, that is the difficulties associated with the Court becoming engaged in an ongoing regulatory oversight role as regards BOPE's conduct, and being in effect required to set or approve BOPE's prices, the Commission noted that the relief it sought in the first instance was declaratory, as to whether or not BOPE's conduct had breached s 36. The application for injunctive relief was conditional, and did not necessarily flow from a grant of declaratory relief. [384] Therefore, the correct approach for the Court to adopt was to consider the elements of the section in light of the facts as established and determine, in an orthodox manner, whether breaches had occurred. If they had, that would answer the defendant's argument as to philosophical approach, because the New Zealand statutory scheme in that context reflected the philosophy that where the elements of the section are established, competition will have been harmed. [385] In considering those competing positions we start with the following propositions: (a) A firm with lawfully acquired monopoly power is not required to assist its competitors (See cases discussed in Gault at para CA 36.21). (b) A firm in a monopoly position is entitled to compete with its competitors. It does not have to "hold an umbrella over inefficient competitors" (Telecom at 402).(c) Section 36 does not prevent the operation of a lawfully acquired monopoly, nor does it forbid the collection of monopoly profits. Rather, s36 aims to preserve or bring about circumstances of competition which, over time, will erode a monopoly position (Telecom at 407). [386] More generally, and in economic terms, it could perhaps be said that s 36 does not require firms to act efficiently but, on and subject to its terms, prohibits them from impeding competition as a way of promoting more efficient markets over time. [387] The 'essential facility/refusal to supply' cases – more common in overseas jurisdictions than in New Zealand – can be seen as qualifying the principles set out above. From those cases it would appear that where a firm controls an essential facility or an essential input product it may in certain circumstances be required to supply that product, or access to that facility, at a 'competitive' price. The case that best exemplifies that principle, in terms of the comparable New Zealand and Australian legislative schemes, is Queensland Wire.[388] In cases where such an obligation to supply has been upheld two factors have been present: [a] In order for a facility or input product to be considered 'essential' it must not be practically duplicable, there must be no close substitutes for the facility or input, and it must be a necessary facility or input for competition in the pleaded market. Where the product in question is an input for a product in a downstream market in which the plaintiff wishes to compete, there must also be no close substitutes for the downstream product. For example, in Queensland Wire, Y-bar was essential for the manufacture of star-picket fence posts and there were no close substitutes for either Y-bar or star-picket posts. [b] The defendant firm must control the facility or input. If the same facility or input is available elsewhere, then the alleged infringer would not be dominant in the pleaded market. See for example, againin Queensland Wire, the High Court's comments to the effect that there were no other steel manufacturers to supply Y-bar. [389] To the extent that the essential facility/refusal to supply doctrine can be seen to apply in New Zealand and involve – as part of the analysis under s 36 going to markets and dominance – a good or service which is not practically duplicable and enjoys no close substitutes, that would not appear to have been the position as regards BOPE's non-time of use meters. [390] If BOPE's impugned conduct is characterised as a refusal by it to supply meter leasing services, then its competitors were able, albeit in some of their cases reluctantly perhaps, to install their own meters and acquire metering data in that way. Furthermore, BOPE was at various times willing to sell MARIA compliant meters to competitors. In terms of the Telecom decision, in which the PSTN was clearly an essential facility (in the sense that it was not practically duplicable and competitors needed to obtain access to it in order to achieve the ubiquity necessary to compete in the market), it is difficult to characterise electricity meters in the same light. Similarly, using the analogy of Queensland Wire, it cannot be said that BOPE's meter leasing services enjoyed no close substitutes in the sense that Y-bar (and star- picket posts in the downstream market) enjoyed no close substitutes. As we think was clear from the evidence, it was practicable to obtain substitutes for BOPE's meter leasing services, even if some comparative cost issues and questions of competitive advantage may have arisen. [391] There was evidence on this issue of comparative costs from a number of retailers who installed their own meters. Those retailers asserted that those costs made the Horizon Area a comparatively less attractive area – relative to other possible markets. Dr Pleatsikas identified this aspect of those costs where, at para (76) of his witness statement, he referred to "equivalent customer acquisition costs in other service territories". That evidence was not, however, generally presented in a way which explicitly compared those costs to the actual costs of leasing in other areas, although we note the reference in Mr Treadwell's evidence (see [191] above) to meter leasing costs of 4.5 – 5 cents per day per meter . Nor was particular attention paid to a comparison of the costs said to be imposed by BOPE's conduct,relative to the costs BOPE itself faced in providing to its customers metering services. Yet, we note, Dr Pleatsikas's concern was whether BOPE's competitors could get access to metering services at a price which was equivalent to the price that BOPE faced. [392] Whilst, therefore, it is possible that it was cheaper for firms to lease meters from incumbent suppliers in other areas than it was for them to purchase and install their own meters in the Horizon area, the evidence did not establish that it was impracticable for them to do so. To the extent that competitors may have faced higher costs in obtaining substitutes for BOPE's meter leasing services we note that these differences in costs were not clearly enunciated and, in any event, BOPE's meter leasing services were sufficiently substitutable that the criteria set out above in [388] were not established. [393] Furthermore, evidence from BOPE, as to more recent activity, also confirmed that other retailers would appear to be installing meters. [394] It is therefore difficult, in our view, to think of BOPE's meter leasing services as an essential facility or input, and equally difficult therefore to conclude that there was an obligation on BOPE under s 36 to supply meter leasing services to potential competitors. [395] In any event, it is not clear that New Zealand courts have, as yet, adopted the essential facility/refusal to supply doctrine as such, in circumstances where a firm with dominance/a substantial degree of market power is confronted by a request from a competitor to unbundle some part of a vertically integrated supply chain, at a price that will facilitate competitive entry. We do recognise that Queensland Wire, persuasive authority in New Zealand, would appear to have been such a case, and that such an outcome was also in our view, in effect, the Commission's object – at least in part – here. [396] As regards New Zealand cases: [a] In Telecom v Clear Communications the Privy Council did not discuss the initial conceptual question of whether s36 was capable ofimposing an obligation on a firm to supply a service to a competitor. It was always accepted throughout the course of the proceedings thatTelecom was obliged to offer interconnection to Clear. The parties had been negotiating towards that end for some considerable time, and the only issue for the Board was whether the terms of Telecom's offer infringed s36. [b] In Auckland Regional Authority v Mutual Rental Cars Justice Barker discussed at length the essential facility doctrine as applied in the US. The basis for his decision, however, was that the contractual commitments agreed to by the Regional Authority had the effect of excluding other potential rental car operators from applying for a licence. The Court emphasised that the Regional Authority was not necessarily required to accept any applicant for a licence. The case did not go so far as to require it to grant Budget a licence. Rather it was the Regional Authority's conduct in binding itself to supply licences only to particular operators that contravened the Act. The ARA case did not, furthermore, involve a dominant firm dealing with a potential competitor. Rather the Regional Authority can be seen as having discriminated in the way it dealt with potential, as opposed to existing, customers. [c] In Port Nelson Ltd (CA) the Court declined to apply the essential facility doctrine, said to be recognised in the US Courts, as not being applicable in New Zealand. Furthermore, that case also did not involve an alleged refusal by a dominant firm to supply a potential competitor. Rather, the dominant firm used its market power to impose a tie on its customers, thus excluding the potential competitor from supplying one element (pilotage) of the services (tugs and pilotage) the subject of dominant firm's tie. [397] As regards Australian cases, we note that Queensland Wire provides compelling authority for the imposition of an obligation to supply in New Zealand law. To the extent, therefore, that the essential facility/refusal to supply doctrine applies in New Zealand, it is the approach adopted in Queensland Wire under theTrade Practices Act that we think is applicable to the Commerce Act in New Zealand, rather than the essential facilities doctrine as it is applied under the very different statutory provisions in the Sherman Act in the United States. Our view is that Queensland Wire is in line with the criteria we have set out at [388] above. [398] Finally, in terms of the elements of s 36, the criteria set out at [388] are part of the assessment of the relevant market and the degree of market power enjoyed by the defendant firm. Under the refusal to supply characterisation of the case, the Commission's argument was that the product dimensions of the market were limited to meter leasing services, on the basis that the other options offered by BOPE were comparatively more costly such that they were not substitutes for BOPE's meter leasing services (the Commission's 'central assertion', referred to above at [14]). Hence BOPE was dominant, or had a substantial degree of market power, within that market. As is evident from the above discussion, we do not agree that the substitutes for meter leasing were so costly as to fall outside the relevant market. We go on now to consider in greater detail the cost comparison in terms of the characterisation of the case as one of cost-imposing behaviour raising barriers to entry.Raising rivals' costs: the creation of barriers to entry[399] The Commission also characterised its case as involving "cost raising" behaviour by BOPE. Rivals' costs were raised because they were forced, by BOPE's refusal to lease meters, to install their own meters. They faced additional costs that resulted directly from having to install their own meters; they also faced indirectly the costs customers incurred as a result. The costs incurred by customers were the actual and perceived costs of the interruption in their electricity supply, and associated inconvenience, made necessary by the installation of new meters. Those costs were barriers to entry into the local markets. It was BOPE's behaviour in refusing to lease its in situ, deemed compliant, meters that was the ultimate source of those costs. That cost imposing behaviour maintained those markets, was the source of BOPE's market power, and represented the use of that market power. [400] This was the characterisation that Dr Pleatsikas preferred. In terms of Mr Farmer's characterisation of the case as an, albeit inappropriate, attempt by the Commission to call into play the essential facility/refusal to supply doctrine, theCourt asked Dr Pleatsikas whether he saw this case in a similar light. He responded, by saying that he did not think about the case as involving essential facilities or natural monopolies, but rather one where the behaviour in question could be seen as imposing on competitors unavoidable additional costs not faced by the incumbent. It was for that reason that he had referred in his evidence to economic sabotage literature, economic sabotage being essentially conduct aimed at raising rivals' costs and reducing rivals' revenues. BOPE, he said, had many of the characteristics that one would associate with that kind of behaviour: large market share, inelastic demand, tight vertical integration and a number of others. [401] In other words, it was not BOPE's refusal to lease its meters in and of itself that gives rise to the competition concerns, but rather steps which BOPE may have taken – not available to it in a workably competitive market– which imposed on BOPE's competitors additional costs which BOPE itself did not face and which therefore could potentially give it an advantage over those competitors. [402] On that basis, the Commission's allegations, particularly as they developed at trial based on the evidence, were not so much that BOPE's conduct constituted an unlawful refusal to supply. Dr Pleatsikas' evidence, as to the economic equivalence of sale and leasing – which we will turn to shortly – was important in that context. Rather the Commission's case as characterised by Dr Pleatsikas was that BOPE's conduct imposed additional costs on potential competitors, relative to those faced by BOPE, in obtaining access to meters as required to compete with BOPE. [403] According to Dr Pleatsikas, the additional costs consequent on BOPE's policy of refusing to lease meters comprised: [a] Installation (but not purchase) costs of meters, whether as regards meters installed by the new entrant itself, or associated with the installation and sale of new MARIA compliant meters by BOPE. [b] The costs of customer churn. Dr Pleatsikas did not pay a lot of attention to customer churn, in and of itself. Costs associated with customer churn reflect the risk that a customer won by a competitor may subsequently be lost before any investment required to win thatcompetitor has been paid off. It is difficult to see how BOPE's behaviour can be seen as imposing customer churn costs in and of themselves, except as linked to the "extra" costs of meter installation. Subsequently, there was considerable discussion of the implications in this context of BOPE's alleged policy of not leasing meters installed by competitors when it won a customer off those competitors. [c] Service interruption costs. [404] In Dr Pleatsikis' view, these additional costs – of installation, electricity interruption and churn – raised substantial, sunk cost, barriers to entry. Because they represented such a substantial fraction of expected net margins, they would facilitate the ability of the incumbent to earn a rate of return substantially in excess of its cost of capital, or – if inefficient – to charge supra-competitive prices whilst earning normal returns. [405] For each of the Commission and Dr Pleatsikas, those increased costs were also central to their conclusion that BOPE enjoyed dominance/substantial market power in the local market they had identified. Essentially it was those costs, together with BOPE's historic incumbency and market share, that resulted in it having that market power. [406] Finally, for each of the Commission and Dr Pleatsikas, the behaviour they identified as constituting use and taking advantage was the same cost imposing behaviour already identified as maintaining barriers and giving rise to dominance/substantial market power. [407] There was, however, a significant difference of approach between the Commission and Dr Pleatsikas at this point. As we have noted, the key element of the additional costs identified by the Commission in argument were the costs, relative to leasing in-situ, deemed compliant, meters, of acquiring and installing new MARIA compliant meters.[408] As developed at trial, Dr Pleatsikas's evidence did not support that contention. [409] That there was this difference of view between the Commission and its expert economic witness became clearer during the "hot tub", which was conducted amongst the expert economic witnesses. [410] First, the Commission regarded the costs associated with acquiring meters as being relevant, as well as the costs associated with installing those meters. For Dr Pleatsikas, it appeared initially to have only been the installation costs that were of concern. Dr Pleatsikas had analysed the costs – for competitors – associated with BOPE's behaviour "ignoring the cost of the meter itself" (see para (77)) of his witness statement). [411] Furthermore, in his testimony during the "hot tub" process, Dr Pleatsikas said he saw no distinction for competition purposes in economic terms between a sale or a lease of existing, deemed compliant, installed meters. By Dr Pleatsikas' account, selling installed meters was economically equivalent to leasing installed meters. A refusal to lease was not, therefore, significant in economic or competition terms if BOPE was prepared to sell those meters. [412] There was the further question, however, of BOPE's policy on MARIA compliance. Specifically during the course of its negotiations with Genesis, and subsequently more generally, BOPE indicated that where it sold meters to competitors it would, prior to the sale, upgrade those meters so that they were certified MARIA-compliant. Again, as we understood his testimony, Dr Pleatsikas acknowledged (see p 398 of the transcript, line 5 and following) that a firm owning a deemed MARIA-compliant installed meter would, in a workably competitive market, price that meter for sale at a price that left its purchaser approximately indifferent between purchasing that meter or purchasing its own certified MARIA- compliant, meter. Accordingly, leasing in situ meters was economically equivalent to selling new certified compliant meters. BOPE's insistence on selling only certified MARIA-compliant meters could not therefore be considered an addition to its putative competitors' costs: selling the installed deemed compliant meter instead would, in a workably competitive market, make the competitor no better off.[413] Taken overall therefore, we understood Dr Pleatsikas to confirm that he did not regard acquisition, as opposed to leasing, and differential costs as between non- MARIA-compliant and MARIA-compliant meters, as relevant. His reason for this was that in a workably competitive market, all these were simply alternative ways of obtaining a meter and competition would lead such prices to converge, including as between non-MARIA-compliant and MARIA-compliant meters. [414] In terms of this analysis, leasing in situ meters was equivalent to selling in situ meters. Further, in a workably competitive market an incumbent could price its in situ meters up to the point where an entrant was indifferent between buying them and installing its own certified compliant meters. Accordingly, leasing in situ meters was equivalent to selling new certified compliant meters, and vice versa. [415] We agree with that assessment. Irrespective of Dr Pleatsikas's evidence, in our judgment the reason for this is that one would expect the terms of a lease and a sale to be such the parties would be indifferent between the two modes of transfer. If not, then relative demand for the more attractive mode would rise and push up the relative price until the point of indifference was again reached. [416] As we have noted, these matters were explored by the Court with Dr Pleatsikas during the "hot tub" procedure. We think Dr Pleatsikas was reasonably clearly of the view that there was no significance in economic or competition terms, as regards meter acquisition costs, in BOPE's practice and policy of not leasing meters, but offering to sell MARIA-compliant meters. This can be seen as being consistent with his evidence in chief, where his focus had very much been on the costs of installing, as opposed to acquiring, meters. It was the imposition of those installation costs which he appeared to accept created the barriers to entry and which was use, or taking advantage, of BOPE's market power. [417] We acknowledge that it was perhaps less clear whether, during those exchanges, Dr Pleatsikas explicitly accepted that the same considerations applied as regards installation, as opposed to acquisition, costs. In our view, however, that conclusion is implicit in Dr Pleatsikas's recognition that in a workably competitive market the price of a non-compliant in situ meter would rise to the indifference point relative to the cost of a new, MARIA-compliant meter.[418] That is, in any event, the view we have formed, having considered the evidence we heard. In our view just as acquisition costs would converge, so would the cost to lease a non-compliant in situ existing meter converge with the price at which BOPE would install and lease or sell a new MARIA-compliant meter. If the alternative for the new entrant is to install a meter, and that meter has to be MARIA- compliant, then in a workably competitive market the cost to lease an in situ deemed compliant meter would tend to converge to a point of economic indifference with the cost of a sale or lease (including installation costs) of a new MARIA-compliant meter. That is not of course, to say the costs would be the same. Rather, it is to say they would be economically equivalent. [419] In terms therefore of asserting that substantial market power derived from the imposition of those costs, and the barrier to entry they represented, the Commission in our view did not establish its case. Our conclusion, supported we think explicitly (as regards acquisition costs) and implicitly at least (as regards installation costs) by the Commission's witness Dr Pleatsikas, is that the Commission did not establish that these were "extra" costs. The costs involved in acquiring access to meters were not – in the economic terms we think are relevant given the nature of the s 36 case asserted by the Commission – increased because BOPE did not lease in situ deemed compliant meters. [420] In other words, BOPE's "no-leasing" policy in and of itself did not maintain or create a barrier to entry. [421] The Commission's case was based on the existence of a local market for electricity metering services having the geographic dimensions of the Horizon service territory. As discussed in the Commission's submissions and in the evidence of Dr Pleatsikas, the dimensions of that market, both its product dimensions and its geographic dimensions, were delineated with reference to the additional costs involved in entering the market other than by way of acquiring metering services from BOPE. Accordingly, the no-leasing policy was the basis both for the existence of the market and for BOPE's dominance or substantial degree of market power within it. We have found that the additional costs imposed on competitors by the no- leasing policy were not economically significant. In terms of s 36, therefore, the Commission failed to establish – by reference to BOPE's policy of not leasing its insitu deemed compliant meters – that the pleaded local market for metering services existed, or that BOPE was dominant, or had a substantial degree of market power, within that market. [422] In our view, this conclusion is also confirmed by a number of more general observations regarding the case the Commission presented based on BOPE's no- leasing policy. [423] The underlying assertion made by the Commission was that the margins available to firms selling electricity in the retail market were so "thin" that, unless retailers were able to lease meters from BOPE, it was not attractive for them to enter the Eastern Bay Retail Electricity Market. [424] On this point Dr Pleatsikas observed, however, that this was not a situation where no customer would be profitable with the cost penalty the Commission asserted had been imposed. [425] Dr Pleatsikas also commented that his assumption was that BOPE would sell or lease meters at a price which fully compensated it. It was not therefore, as Professor Sweeney had asserted, a question of BOPE having to subsidise someone. Rather, it was a question of whether a new entrant could get access to the market at a price which was equivalent to the price that BOPE faced. [426] The inference which we think is to be drawn from these remarks, and from the Commission's concern with the effect – relative to available margins – of BOPE's allegedly anti-competitive behaviour, was that even where a new entrant wished to compete with BOPE on a relatively limited scale (for example – as appeared as regards a small number of customers targeted not directly, but only as a result of a national promotion) new entrants should expect to enjoy substantially the same margins for supplying those few customers in the Horizon Area as BOPE enjoyed for its much larger customer base. We are not persuaded by that proposition. [427] The Commission and its witnesses referred to the anti-competitive effect of BOPE's behaviour, principally in the context of evidence of the relatively limitedinterest BOPE's competitors had, as a matter of fact, in competing with BOPE in the Horizon Area. We refer, in particular, to the evidence of the various retailers that, BOPE's conduct aside, they had little interest in competing for retail customers in the Eastern Bay, but only did so in support of national promotions. [428] Referring to that evidence, and in the context of his discussion of market issues, Dr Pleatsikas commented as follows:These [witnesses] indicate that, for meters and metering services, the financially [sic] feasibility of a non-incumbent to provide competitive non- time-of-use metering services in the Horizon service territory in response to supra-competitive prices charged by the incumbent is quite low (e.g., because a non-incumbent would incur much higher costs to install a few meters in the Horizon service territory and would be at risk of incurring stranded metering investment if a customer were switch retail service back to BOPE). In other words, any supply-side response by competitive providers of electricity retailing services to a price increase in these services by the incumbent would be insufficient to make a SSNIP unprofitable in the Horizon service territory. (Pleatsikas Brief, para 71)[429] This reference to the installation of a "few meters", where the underlying assertion was that the barriers to entry created by BOPE were sufficient to shelter a SSNIP in the downstream electricity retail market, struck us as somewhat surprising. [430] We were not persuaded that, for example, an efficient competitor who entered BOPE's territory in response to a SSNIP with a competitive pricing or service proposal would – almost by definition as seemed to be assumed by the Commission – only succeed in attracting a "few" customers. Rather, it would seem at least possible that such a firm could attract more substantial numbers of customers, and would therefore be able to co-ordinate the switching of those customers, and in that way reduce the costs that were involved in that switching process. [431] Perhaps expressed most simply we were not persuaded that – irrespective of what the position may be as regards margins in the electricity retail sector – a firm wishing to compete to supply retail customers should not expect to make an appropriate up-front investment, including the capital cost of acquiring and installing meters and other costs associated with presenting a competitive offer, so as to attract new customers.[432] Taken overall, therefore, we think there was some force in BOPE's response, as articulated by Professor Sweeney, that – as regards the costs of acquiring and installing meters – the Commission was asking BOPE to "help" its competitors enter the market and compete against it. [433] We also note Professor Sweeney's evidence that "barriers to entry are substantial costs that new competitors must face but an existing firm neither has faced nor would face in order to compete as a seller". [434] Whilst that analysis may not apply to some additional costs, particularly where imposed by strategic behaviour, we think it is appropriate to note that BOPE itself had faced the costs of acquiring its metering assets. That cost was identified as being in the vicinity of NZ$12 million. BOPE therefore had faced meter acquisition costs. We also observe, although this was not referred to directly in evidence, that BOPE continues to face meter acquisition and installation costs where it wins a contract to supply electricity in the retail market to a "new" address. In that situation, our understanding is that the rate BOPE charges for its electricity will be the same as it charges existing customers. [435] Moreover, given industry requirements for gradual MARIA compliance, more and more of BOPE's customers will have new MARIA-compliant meters installed, and yet we understood the price for delivered electricity as between customers with existing deemed compliant "old" meters and "new" MARIA- compliant meters is the same. [436] It is therefore difficult to conclude that incurring those acquisition and installation costs makes supply so unprofitable as to preclude competition, or that a new entrant should not be expected to bear its own costs in this area, even if in terms of initial entry that may have made early customers less profitable for it than would be the case for BOPE with its existing customer base. [437] If, in other areas of New Zealand, incumbent firms were leasing meters on a basis that appeared to make entry into their service territories relatively more attractive than in the case of BOPE's offer then, on the basis of the evidence we heard, two observations are possible:[a] First, for firms which had an interest – which BOPE did not appear to have in any material way – in competing outside their incumbency, there may have been a business rationale in such reciprocal arrangements. [b] Secondly, there was some evidence that that behaviour was adopted by firms in response to the government's expectation as to the industry's co-operation in promoting competition in the retail electricity market. In this regard: [i] We refer to the evidence of Mr Reilly where he observed that firms entered into reciprocal meter leasing arrangements to enable competition for small, individual customers. [ii] Dr Pleatsikas also acknowledged this possibility. [iii] The Commission itself in Decision 333 observed that the industry faced the threat, if it had been unable to introduce retail competition, of the government introducing a "mandatory default" system by regulation. [438] We do not place great weight on this particular part of our analysis. Nevertheless, it does suggest that the behaviour of other firms with which BOPE was being compared unfavourably by the Commission may not necessarily have been behaviour that those firms would have adopted in the absence of the very clear government expectations as to the way in which the industry was, effectively, to work together to facilitate retail competition. [439] Whilst, therefore, we acknowledge that BOPE's policy did require potential competitors to incur costs of meter installation, we do not consider that those costs should be regarded as "additional" in the manner asserted by the Commission. BOPE's conduct imposed no additional costs on competitors that they would not otherwise have faced in a workably competitive market, because no competitive force could prevent BOPE from pricing installed meters up to the point of indifference with MARIA compliant meters. To the extent therefore that it was asregards those costs that the Commission asserted BOPE had created barriers to entry, and had thereby continued and delineated the pleaded local markets and its dominance/substantial market power in them, our conclusion is that the Commission did not establish its case. [440] Finally, we comment briefly on some of the other evidence that was before us in the trial. In addition to the cost comparisons discussed above, the Commission adduced evidence of a range of other factors that it said illustrated BOPE's dominance within the pleaded markets. We refer to the evidence of Mr Westergaard and Mr Taylor, in which they discussed, inter alia, comparisons between BOPE's gross retail margin and the national weighted average, BOPE's retail margins versus the national margins, BOPE's prices in the Horizon Area versus its prices in other service territories, relative market share, churn rates and average wholesale prices. This evidence was said to be indicative of BOPE's dominance within the pleaded local markets. In his closing submissions Mr Brown relied on this evidence (as well as the other evidence of costs that we have already discussed above) to argue that the circumstances of the retail electricity market in the Eastern Bay of Plenty region "spoke for themselves" in suggesting anti-competitive conduct on the part of BOPE. Our view is that much of this evidence, while it might have suggested the existence of market power in an ex post sense, was not helpful in determining the source of BOPE's market power and the nature of that power. Given that the s 36 analysis, specifically the counterfactual test, requires the Court to investigate the connection between the alleged market power and the conduct said to constitute use or taking advantage, it will generally not be enough to point to circumstances said to constitute the results of dominance. What is more helpful for the Court is evidence that establishes the source of the alleged market power and its nature. That is the starting point from which the counterfactual test proceeds.First alternative analysis – cost imposing obstructive conduct[441] In addition to the costs associated with the acquisition and installation of meters, the significance of which we have already considered, the Commission also asserted that BOPE's policies and behaviour imposed other relevant costs on its competitors. Dr Pleatsikas shared that view.[442] We think these costs can be described, in terms of the case the Commission put forward, as being costs necessarily incurred where a new meter is installed when a customer switches supplier together with the additional costs associated with BOPE's conduct of: [a] Obstructing efforts by potential competitors to enter the market, principally by refusing or delaying to enter into negotiations with them; [b] Exaggerating to its Consumers the inconvenience of switching to a new supplier; and [c] Increasing those switching costs by not willingly coordinating the processes required for such switches to occur so that, for example, a Consumer was forced to incur a greater interruption to its electricity supply than would otherwise have been necessary. [443] We think it is the extent to which those costs were raised by BOPE's conduct that is relevant. In other words, if BOPE had willingly co-ordinated the process whereby a new meter was to be installed (as it does for itself in terms of its MARIA upgrade programme) then, in our view, those were costs which a new entrant could be expected to meet in order to compete. We think, however, that the additional costs, and the behaviour that gave rise to them, is of potential relevance. [444] The clearest evidence of such behaviour was, perhaps, in the context of BOPE's dispute with Meridian, where BOPE went so far as to write to customers advising them that their power would be disconnected because, in effect, they had accepted Meridian's "Sky" offer. [445] In this context, of possible relevance also is the matter on which considerable attention was focussed during the trial, namely BOPE's alleged policy of not re- leasing meters installed by other retailers when BOPE won back a customer from such a retailer.[446] We have analysed BOPE's behaviour in detail earlier in this judgment, and have recorded our conclusions as to the nature of that conduct. We have concluded that BOPE did behave in ways which gave rise to delays for competitors in even being able to negotiate with it on the question of meter access. BOPE did exaggerate switching costs, which did result in the costs faced by consumers in switching supply from BOPE to a competitor being greater than they needed to be. In general, we confirm that this behaviour would appear to have occurred in the early period of retail market deregulation. On the evidence before us there was no indication that such conduct continued after BOPE had finally settled on its "standard" response to requests for switching in late 2002. [447] In our view, however, the Commission did not base its case on the proposition that such conduct constituted use of BOPE's alleged dominance or substantial market power. Rather, and we refer in particular to paras (21) and (22) of the statement of claim, such conduct was identified as contributing to BOPE's dominance, its substantial market power. On use and taking advantage, the Commission's pleaded case was clear: The use of BOPE's dominance, of its substantial market power, was BOPE's policy of not entering into meter-leasing agreements. Whilst it could be argued that the particulars of that conduct overlap with the other cost raising behaviour pleaded by the Commission at paras (21) and (22) of its statement of claim, and also identified by Dr Pleatsikas in his evidence, we think it would be going too far to say that the Commission had advanced its case against BOPE on the basis that this conduct also constituted use or taking advantage. We think this conclusion is supported by the terms of the declaratory order sought by the Commission, namely that the conduct of BOPE in refusing access to its meters was in contravention of s 36 and, in the event of such a declaration being granted and BOPE continuing to act in contravention, orders whether in mandatory or restraining form to facilitate the provision of electricity metering services. [448] For that reason alone, and as argued strongly by Mr Farmer in his closing, we do not think it appropriate to restructure the Commission's case and analyse it from the perspective of whether any additional costs associated with that behaviour were significant enough, in and of themselves, to delineate the pleaded local markets, continue or establish BOPE's dominance, and also constitute use and taking advantage as had been pleaded for BOPE's "no leasing" policy.[449] We accept, however, that situations do arise where a case will develop and change during the course of a trial, but where nevertheless the case – as so developed and changed – can fairly be seen as coming within the case originally pleaded. In those situations, it would be unrealistic not to determine the substantive dispute as it emerged at trial. [450] In this instance, it is fair to say that considerable attention was paid, during the trial, to this other behaviour of BOPE associated with – and perhaps from the Commission's perspective, however its pleadings were structured, comprising part of – the overall conduct challenged by the Commission as being in breach of the Act. [451] We therefore propose to consider what the position would have been if, contrary to the finding we have just recorded, we did think it fair to conclude that the Commission had effectively based its case against BOPE on this conduct also constituting use, or taking advantage, of the position of dominance or substantial market power that the Commission asserted. [452] Were that to have been the case, then we think the sensible place to start in any analysis of this behaviour is with the counterfactual test. We say this because as before, if the conduct in question could not, because it passes the counterfactual test, constitute taking advantage or use, then the case cannot succeed by reference to that behaviour. [453] We have concluded that such behaviour would not have been available to BOPE in a workably competitive market. [454] That is, as regards increased disconnection, power supply interruption, and other "inconvenience" costs, caused by the behaviour we describe at [442] above, we considered that those are not costs that a firm could impose on its rivals, either directly or indirectly through customers, in a workably competitive market. A firm which attempted to do so would over time face the loss of its own customers who, to avoid the adverse consequences for them of not easily being able to switch away from that firm to a competitor, would choose to be served by those competitors rather than by the firm that attempted to impose those costs on them. As regards this behaviour, therefore, we accept Dr Pleatsikas's evidence.[455] We therefore go on now to consider whether this cost imposing behaviour raised barriers to entry so as, in turn, to delineate the pleaded local markets and give BOPE dominance/substantial market power in those local markets. [456] In our view this is an assertion that the extra costs involved were sufficient to delineate the relevant market. For that to be the case, those additional costs, because they are said to be capable of delineating that market – and in turn to shelter BOPE from competition in the downstream Eastern Bay Retail Electricity Market which was the asserted target of BOPE's behaviour – would have by definition to be sufficiently large to enable BOPE, had it desired to, to impose a SSNIP in that downstream market. [457] We reach that conclusion for the following reasons. The usual role of a SSNIP analysis is to assist in the determination of the boundaries of a market. A market will be the smallest area within which suppliers of the relevant goods or services could, by colluding, successfully impose a SSNIP without being disciplined by competition from other areas. Here, the Commission's assertion is that it was BOPE's ability to impose additional costs, through its impugned behaviour, that defined the market, essentially by creating barriers to entry. In our view that suggests an inquiry as to whether the increased costs were such that they represented a barrier behind which BOPE might subsequently (irrespective of whether or not it did in fact) have imposed a SSNIP. Those additional costs themselves, therefore, would have to be in the same order of magnitude as a SSNIP to constitute such a barrier. Therefore, an enquiry into the quantum of those costs is relevant. [458] We acknowledge that this is not the general position under s36. Generally s36, as a matter of law, directs the Court to determine whether or not a firm has used its dominance/taken advantage of its substantial market power, for an illegitimate purpose. The extent of any anti-competitive effect is not a direct focus of the section. Rather, the counterfactual test is the principal methodological tool to determine the questions of use/taking advantage. [459] In this case, however, and in terms both of the Commission's central assertion, and its pleadings as to the existence and boundaries of the relevant markets such an enquiry is, in our view, called for.[460] Were we wrong, therefore, to conclude that the Commission had not effectively based its case against BOPE on this other cost raising behaviour also being use or taking advantage, it would be necessary to consider the costs potentially associated with that behaviour. [461] The principal sources of evidence as regards the costs of the interruption to power supply associated with the replacement of an old, deemed MARIA-compliant meter, with a new, certified MARIA-compliant meter (as required by the defendant's policy to sell only such meters) were Mr Westergaard for the plaintiff and Mr Murray for the defendant. [462] They both cited a number of international studies that attempted to discern the actual costs of power disruption. Many of these concerned fairly lengthy disruptions, and the accuracy of simply taking a proportionate fraction of these numbers to get the equivalent costs for a shorter disruption was discussed. Mr Murray argued that the average cost of a disruption was likely to increase over the time of the disruption (citing the example of thawing frozen foods which will not be a problem for a half hour disruption, but will be for a four hour disruption). Other examples, such as resetting digital clocks and VCRs, suggest that average cost decreases over the time of disruption. [463] A number of these studies looked at unplanned disruptions, and it was in our view reasonably suggested that the costs of a planned disruption would likely be less. [464] In other relevant evidence the Court heard from Mr Crosby that the electricity to a house, "is usually off for only 10-30 minutes" when a meter is changed. [465] We also received evidence from BOPE Consumers who did switch suppliers. In Mr Van der Weff's case the meter switch took approximately 1-2 hours, during which the power was turned off. In that case, however, it is not clear that the power was off through the entire switching process and it is also not clear that the meter concerned was a "standard" single-phase meter. Mr Sedgman did not refer to any inconvenience when his new meter was installed.[466] On the assumption, which we consider reasonable in light of all the evidence, that a power disruption of an hour is at the high end for a typical Consumer installation, Mr Westergaard's numbers for a four hour disruption suggest the average actual costs of a one hour disruption would not exceed an amount in the vicinity of $3.50 for a small domestic user. [467] The Commission's case was that these numbers were uninformative, except as lower bounds to the true perceived costs to Consumers, because of the uncertainty that Consumers faced in determining the true costs of disruption, particularly given communications from the incumbent that allude to inconvenience and disruption. The Commission referred in particular, in cross-examination of Mr Murray, to the evidence of Miss Spelmeyer, who could not take the risk of being without electricity and therefore decided not to shift to BOPE, and Mr Nunweek who also did not go ahead with the Sky offer. [468] The Commission also pointed to BOPE publicly emphasising to Consumers the need for disconnection, including comments in the context of the Meridian dispute, acknowledged by Mr Tweedie to have been excessive. [469] However, no attempt was made by the Commission to quantify these alleged effects or perceived costs. Indeed, in closing Mr Brown initially argued that it was not necessary to place a dollar value on perceptions of these costs, as the facts spoke for themselves, a reference to evidence of comparative switching or "churn" rates in the Eastern Bay. In our view, however, attempting to infer switching costs from observed switching – or churn – rates is an unreliable exercise, as there are a number of other possible explanations for the apparently low churn rates in the Eastern Bay. We think this observation is supported by evidence from retailers who commented that, even absent BOPE's policies, they had little interest generally in competing in the Horizon Area. Furthermore, Mr Westergaard himself was unable to say that the apparently low churn rates in the Horizon Area were not attributable to factors particular to that area. [470] On the basis of the evidence available, we therefore conclude that the figure of $3.50 is an appropriate indicator of the order of magnitude of service interruptionand related inconvenience costs, noting, however, that it is the extent to which that cost was increased by BOPE's behaviour that is the relevant consideration. [471] It is difficult, therefore, on the basis of the evidence available to us to reach firm conclusions in this area. [472] Considerable attention was also paid at trial to the question of stranding risk, that is the risk, where a firm has installed a meter to acquire a customer, and it subsequently loses that customer, that it may also lose all or a significant part of the investment in that meter. This was highlighted, in the Horizon Area, by reference to what was alleged to be BOPE's policy of not leasing such meters where it won a customer back. [473] Dr Pleatsikas referred, without quantification, to the costs of removing a stranded meter and to further costs, for example testing charges, involving re- installing such a meter later. More particular evidence was provided by Mr Taylor and Mr Westergaard on this point, and Mr Murray paid this issue considerable attention. [474] In Mr Westergaard's opinion, BOPE could expect to amortise meter costs over a period of 15½ years or longer. On the basis that a competitor to BOPE was likely to have to remove a meter if the customer switched back to BOPE, but not if the customer switched to another retailer who was willing to purchase or lease the meter, a competitor would only be able to amortise the cost of the meter over 11.3 years. Furthermore, if other retailers adopted the same approach as BOPE (that is not purchasing or leasing meters but installing their own) then the period of time over which a retailer competing with BOPE had to amortise its meters would drop to something approximately 8 years. Mr Westergaard did not monetise these differences. [475] Mr Taylor also approached this issue with reference to the question of a firm's ability to recover the capital costs of a new meter, as affected by the period of time for which it could expect that meter to be installed. Mr Taylor prepared a chart which showed that if a meter remained installed for a 30 year assumed life, the daily charge would need to be approximately 9.5 cents. If, however, the period ofinstallation and therefore use was reduced to two years, that daily charge increased to 41 cents. On that basis, if due to stranding risk the expected use period of the meter was reduced, then costs would obviously rise. Mr Taylor also referred to the costs of reinstallation plus forgone return on investment, even if a meter could be re-used. Mr Taylor did not, however, specifically apply his table on the basis of actual expectations of retailers entering BOPE's market. [476] Mr Murray opined that the risk of stranding, which existed whoever owned a meter, appeared to have been overstated. He suggested that the value to a new retailer of the risk of stranding their meter, were they to purchase a meter for a customer and subsequently lose that customer back to BOPE, could be quantified at $1.25. He reached that view on the basis of a very detailed net present value exercise, designed to assess the comparative benefits of leasing meters as opposed to buying them. Following questions from Mr Brown and the Court, he revisited the methodology for that calculation and, in a document handed up to the Court, adjusted it to $3.89. [477] In our view, the correct way to calculate the stranding risk is to calculate the net present value of meter purchase in the presence of such a risk and compare it to the net present value of meter purchase absent that risk. The difference would be the present value of the risk of stranding. The revised calculation by Mr Murray came closest to being correct methodologically, as it allowed for – as suggested by Mr Brown for the Commission should have been the case – the probability weighted loss to a retailer over time from a customer switching to BOPE. However, in our view it also omitted a further risk, namely the one that stems from a customer switching to a non-BOPE rival and subsequently switching back to BOPE. Such a pattern would also strand the initial retailer's asset, even though it might be initially leased to the non-BOPE rival. [478] Reflecting on the variety of data presented to the Court to indicate the sort of costs associated with meter purchase and leasing, and the sensitivity of any calculation of this sort to the particular data used, our assessment is that were this calculation to be done in the manner we consider to be methodologically correct, a range of values – considerably in excess of the point value derived by Mr Murray, would be likely to be derived.[479] Having said that, and although we therefore have some scepticism that Mr Murray's estimate was especially accurate, the Commission did not provide any firmer evidence in this area. [480] More significantly, however, and as we have already concluded, the Commission did not in our view establish that BOPE had a practice of not leasing competitors' meters. Even if, therefore, we were considering the Commission's case on the basis that these various additional cost raising behaviours had been said to be use or taking advantage, the Commission did not make out a case against BOPE that it had a practice of not leasing meters installed by competitors where BOPE won customers back from those competitors. [481] The final element of this behaviour was BOPE's behaviour, in effect, of avoiding and delaying dealing with competitors who sought to engage it on the question of leasing meters. No effort was made to quantify these costs. Whilst that does not take away from the reality, in our view, that such conduct did occur, it does limit our ability, in this context, to assess its significance. We also note that this conduct would appear to have occurred when – as recognised by a number of witnesses – the market itself was in something of a chaotic state and the whole issue of switching was causing difficulties across the industry. [482] Taken overall, therefore, had we been considering whether or not the additional costs properly to be regarded as associated with this cost increasing behaviour were sufficient to delineate the local markets identified by the Commission in its case, and to give BOPE dominance and substantial market power in those markets, we would have concluded that the Commission had not established that they were of a sufficient order of magnitude to be considered, on the basis of the relevant standard of proof, as more likely than not to have had those effects.Second alternative analysis – market power independent of cost-imposing conduct[483] Before turning to the Commission's s 27 case, we think it is appropriate to note that, in our view, there may have been an alternative analysis of this factual situation which could have resulted in a finding that, although not to the extent asserted by the Commission, BOPE did breach s 36. [484] We have concluded that BOPE's additional cost raising behaviour, that is the cost raising behaviour which was asserted by the Commission to contribute to barriers to entry and therefore to dominance and substantial market power, was not conduct that BOPE could have engaged in, in a workably competitive market. We have also concluded that to the extent that such conduct did raise costs faced by potential entrants, and therefore contribute to barriers to entry, the costs imposed were not established to have been of an order of magnitude sufficient to delineate the pleaded local markets and to give rise to dominance or substantial market power. BOPE did, however, engage in such conduct. [485] It might be inferred from this conclusion that BOPE had some degree of market power. Whilst this does not necessarily imply dominance or a substantial degree of market power, it is nevertheless not inconsistent with that conclusion. On that basis, we have considered whether BOPE might have had dominance or a substantial degree of market power independently of the conduct challenged by the Commission in this case. Were that to be the case, then the conduct we have identified as failing the counter-factual test could constitute use of that market power, irrespective of the question of the quantum of the costs imposed by that behaviour. It is appropriate to record that we have undertaken this inquiry – at least in part – because of the views we reached, and record below, as to BOPE's purposes in this context. [486] To consider whether – by reference to this behaviour alone – BOPE could be said to have had and used substantial market power in a local meter wholesaling or electricity retail market, would therefore require the case as argued by the Commission to be restructured so that it was argued that:[a] BOPE was dominant, or had that substantial market power, independently from the activity which imposed these costs; and [b] The imposition of these costs represented the use of that pre-existing dominance or substantial market power. [487] In our view, this case was very much a reflection of the various ways in which the industry, in response to the government's threat of regulation, had moved to introduce competition at the retail level. What can be noted, in that context, are the following factors: [a] The voluntary switching protocol in place on 1 April 1999 was not particularly effective. This much would appear to be plain. The industry itself moved to refine those arrangements so that, by the middle of 2000, improved switching protocols were introduced. If, however, the switching protocol did not work as introduced on 1 April 1999, this would support the conclusion that competition may not have been enabled to the extent anticipated by the Commission in Decision 333. That conclusion would also appear to be supported by the various witnesses who referred to "chaos" in the industry at the time. [b] BOPE was not, until some time later, itself a party to these voluntary industry arrangements. [c] Furthermore, the question of access to meters, as opposed to switching and other related matters, was never dealt with by the industry rules. [488] It therefore seems to us at least arguable that BOPE, and indeed other firms, may have retained substantial market power or dominance after 1 April 1999 by reason of the fact that the reforms intended to take that power away from them had not been fully effective. It was not, therefore, a question of BOPE's behaviour which maintained that dominance. Rather, BOPE's delaying tactics, and its conduct which exaggerated and increased switching costs, was use of the dominance, thesubstantial market power, that BOPE continued to enjoy, even if other firms chose not to "use" the similar power they also enjoyed. In this context, it may be observed that BOPE's policy of not leasing or selling deemed MARIA compliant meters did impose some costs on BOPE, namely the costs of – in effect – abandoning its investment in such meters. We have therefore also considered whether BOPE's willingness and ability to incur such costs were indicative of substantial market power or dominance. In this context, such conduct would impose costs on BOPE either in a market in which BOPE was dominant, or in a workably competitive market. We therefore do not see that behaviour, i.e. the no-leasing policy, as in and of itself being enabled by dominance or substantial market power. Rather, and if it were established that BOPE had continued to be dominant after 1 April 1999, there would have been no point to that behaviour unless it had provided the opportunity for BOPE's delaying tactics and its associated conduct. [489] Very clearly, however, the Commission did not put its case on that basis. [490] If the case had been put that way, then BOPE would have had the opportunity to answer it, and for its expert witnesses to express their views on the case as so put. In our view, that was not the course this hearing took, although we do recognise that considerable attention was paid by Dr Pleatsikas in particular to the costs associated with BOPE's conduct. [491] We also note, in this context, that the Commission's own witnesses acknowledged the "chaos" that existed in the electricity industry at the time and the fact that it was some time before the industry's voluntary, MARIA, contractual arrangements were implemented and refined so as to effectively provide for retail competition. Again, those are matters in our view which, if the case had been put in the way we have suggested it is possible to think of it, would also have been analysed in more depth to determine the relevance of those matters to the case as so alleged. [492] The result of the foregoing discussion is that we are not satisfied that the Commission established the pleaded local markets necessary to support its cause of action under s 36. Although we have expressed some tentative views about the existence of local markets for electricity metering services in the Horizon serviceterritory, no other market formulation was put to us in submissions, nor was there any other evidence supporting the existence of local markets of the kind argued for by the Commission. Further, BOPE did not expect to, and did not, respond to any other formulation of the applicable markets. The result is that we are unable to reach a concluded view as to the nature and dimensions of the applicable markets.BOPE's purposes[493] We also record here an assessment of BOPE's purposes in the context of s 36. We recognise that, in terms of the overall conclusions we have reached, such a comment is not, strictly speaking, called for. These matters are, however, also of some potential relevance under s 27. Furthermore, a considerable portion of the evidence we heard was related directly or indirectly to this topic. [494] Not surprisingly, BOPE and the Commission took opposing positions as regards BOPE's purpose in adopting the approach that it did to the question of access by competing retailers to its meters. [495] For BOPE, the evidence can be summarised by reference to Mr Tweedie's evidence in chief. This was supported by Mr Bahirathan and Mr Doidge. For Mr Tweedie, the position was relatively straightforward. [496] It simply was not in BOPE's own best interests to provide access by way of lease or similar arrangement to its meters. This would have required it to set up a meter leasing business, something that was not a priority for it, that it did not have the resources to do and that would have diverted it from more pressing issues. Furthermore, approaches for access to meters by people such as Meridian, Mighty River, Genesis and TrustPower were made to BOPE when it was struggling to get on top of its own business. It had taken a bad decision, to rely on KCE to provide most of the management support for BOPE, and that compounded its difficulties. Staff at BOPE and KCE were under significant pressure simply managing BOPE's own business, and its relationship with its customers. Responding to competitors' requests for access to its meters was not a priority, and was not something for which there were any specific regulatory frameworks or obligations.[497] BOPE's early behaviour also reflected, and was part of, the general uncertainty and chaos that existed in the industry in 1999 and into 2000. [498] Over time, however, and under pressure from retailers, BOPE realised it needed to have a fixed policy on the matter. That gave rise to the ultimate decision at the joint venture board meeting of 1 August. [499] There was, furthermore, a sound business rationale for that decision. BOPE itself was not in a position to expand its customer base in any material way. Therefore it did not require, or have an interest like the major retailers, in reciprocal meter leasing arrangements. Further, to establish the internal resources to support leasing meters to its competitors would have been a significant undertaking. For Mr Tweedie, to create a leasing business for BOPE was a complete no-go. [500] Mr Tweedie referred to a letter from BOPE to Energy Online on 1 June 2000:We believe that to attempt to set up a "meter rental" business as you request would be financially unsound for us, limiting our resource in both the support of our customers and Bay of Plenty Electricity's core business as an electricity generator and retailer. (CB 6/1542)[501] That was, from BOPE's perspective, the basic reason, and justification, for its actions. [502] For the Commission, BOPE's explanation was, in reality, at worst a smokescreen, and at best an ancillary reason not taking away from another substantial reason, namely that of impeding competition. [503] In reaching that conclusion, the Commission pointed to a number of what it regarded as key documents. [504] Most tellingly, from the Commission's perspective, in a paper entitled "Strategy Paper for Retail and Metering and Business Units" prepared for a BOPE joint venture meeting of 8 and 9 November 2000, the following comment was recorded:The policy of not allowing other retailers to rent our meters plus our aggressive defensive positioning, combined with stable, competitive andtransparent pricing has achieved our objective of minimal customer losses – approximately 1.5% compared with the national switch average of 14%. (CB 7/1797)[505] In other words, the policy of not allowing other retailers to rent meters had the objective of reducing customer losses. This, the Commission said, clearly showed BOPE's purpose. [506] This extract was put by Mr Brown to Mr Tweedie. Mr Tweedie dismissed it as reflecting neither his nor the firm's policy, but perhaps some misunderstanding by the unreliable Mr Bulley. This would, he said, have been either Mr Till or Mr Bulley and he had made clear what he thought of their overall performance. We do note, at this point, although the evidence was that Mr Till's employment had been discontinued, Mr Bahirathan in cross-examination explained that Mr Bulley had in fact retired as managing director of BOPE at the end of 2005. In our view, that evidence is not consistent with Mr Tweedie's assessment during trial of Mr Bulley. We have little doubt, from observing Mr Tweedie in the witness box, that if he had thought Mr Bulley was as bad as that, Mr Bulley would have been very unlikely to have remained in BOPE's employment until the time indicated by Mr Bahirathan. We therefore accept the Commissioner's submissions that this document is relevant when considering BOPE's purposes. [507] The Commission identified a range of other evidence which, it said, supported the conclusion that BOPE had the purpose of thwarting competition. [508] BOPE was, in the first instance, well aware of the significance of access to meters. [509] Mr Bulley had demonstrated that awareness in comments to Mr Tweedie in connection with BOPE's dealings with TrustPower. In the content of the proposal to lease the meters at the 20 Telecom sites to TrustPower, he made the following comment in September 1999:They are again making the point that they are paying HEDL [Horizon] for line charges and do not wish to pay twice. It still bears that they are only concerned about the Telecom sites (20) in our patch. To avoid having scrap meters should we consider selling the meters as suggested?The other issue for consideration is that a bi lateral arrangement as tabled allows for a notice period of 1 month and is site specific ie we can exclude any sites from the agreement – therefore leaving us (and them) in control if we were unhappy with any marketing push. As also discussed on Tuesday we need to be ready to move into TrustPowers region and without a metering arrangement for the mass market it will be difficult. (CB 3/610)[510] In other words, and with reference to the second paragraph, it would be difficult for BOPE to compete with TrustPower without the benefit of access to TrustPower's meters. [511] In November 2000 BOPE had acquired some business from the Rotorua City Council. BOPE, from Mr Bulley's perspective, wanted access to TrustPower meters, or at a minimum certain relays, to make that business worthwhile. TrustPower was insisting on a bilateral leasing arrangement. Mr Bulley commented:Options are to enter into an agreement with the risk of a move into our patch at a time when we will be introducing a metering change. The upside is that we get on with the Rotorua contract and start the agency campaign on the 2 December and grow the business. The downside being the potential loss of customers. (CB 6/1762-1763)[512] This comment shows BOPE's awareness that it would be easier for TrustPower to compete with BOPE if it had access to BOPE's meters. [513] The Commission also pointed to a number of documents which, it said, reflected BOPE's general approach to make it as difficult as possible for competing retailers to enter the market, by drawing out for as long as possible negotiations in respect of metering agreements so as to facilitate its "win back" campaigns. [514] As regards Mighty River, the Commission pointed to the following comments in the BOPE Joint Venture Management Committee meeting papers for 1 December 1999:Mighty River –agreement tabled. Mighty River are putting considerable pressure to put an agreement in place. They have stated that they need this agreement to supply the customers they obtained through the AnchorMart deal. This impasse is to be reviewed by Richard when all avenues to get customers back has been exhausted. (CB 4/1148)[515] In other words, the delay in responding was for a purpose – to enable BOPE to "win back" customers that had already agreed to switch to Mighty River.[516] As regards Genesis, the Commission referred to a number of documents that showed BOPE had used the drawn out negotiation period to deter competition. As noted above, in June 2000 Genesis was endeavouring to arrange access plans to metering for approximately 170 customers acquired as part of two national deals – Farmlands Coop and New Zealand Fire Service. Genesis met with Todd on 12 June, and followed up with a letter to BOPE on 15 June. The matter would appear to have been discussed at a BOPE Joint Venture meeting on 22 June, after which there was an exchange of emails. (a) On 28 June, Mr Bulley advised as follows:Taken a call from Mark Pickup of Genesis following up on his letter regarding meter access in the BoPE region. As we have got the requested transfers down to 13 ICP's we have done all we can as part of the Farmlands deal. My view is that to be consistent with the TrustPower arrangement and as a commercially justifiable stance we move forward and set up the metering agreement and transfer the customers. This will avoid inconsistency if we are challenged and avoids friction at the FreshStart end. (CB 6/1589)(b) Mr Tweedie, having noted that following the JV meeting on 22 June a decision to stall for more time with Genesis had been taken, then proposed – on various conditions – "the idea of saying to Genesis that we will allow them to use our meters for only these 13 customers ". Mr Tweedie concluded by stating "Please reprimand me if you believe I am going soft." (CB 6/1588). (c) The next day, Mr Bulley emailed Mr Tweedie, taking the Genesis proposal further. Included in that email was the following comment:We could offer to sell the 13 metering set to move the outstanding issue forward without opening the door in any way. (CB 6/1593)(d) Mr Tweedie replied that same day:Yeah that is a new angle selling the 13 meters to Genesis but have we ever got a chance of getting the meterscustomers back --- easier if we own the meters?? Good to see your team have a clear view and strategy on all of this -- -- maybe we sit tight a bit longer with the letter to Jackson that I suggested. (CB 6/1593)[517] As matters transpired, the "13 meter" offer was not made to Genesis at that time. Rather, and apparently in line with the original decision made after the JV meeting that a holding response be sent, on 6 July Mr Bulley wrote, with Mr Tweedie's approval, to Mr Jackson of Genesis advising that, for various reasons, a formal decision was to be made on 1 August at the next BOPE Joint Venture meeting. [518] Internal discussions continued, including by reference to the Commission's investigations. By 14 July BOPE had indicated its willingness to sell meters to Genesis. Genesis acknowledged that offer on 14 July, noting that by their count 170 ICPs were involved for "Federated Farmers and Farmlands customers, and a handful of fire stations". (CB 6/1624) [519] On 18 July Mr Tweedie raised with Mr Bulley the question of the position of the customers who had returned to BOPE, and the need to protect that decision. He went on to say:I said we would only cooperate once a meter sale deal had been agreed --- not before --- of course that was on the assumption of limited transfers i.e. 30ish, not 150 odd as they expect. If we find our assumptions are wrong then it may be back to square one. (CB 6/1634)[520] The email does not say what "square one" is; the clear inference is, however, that it would involve BOPE revisiting its willingness to provide access to its meters to Genesis. [521] As regards TrustPower, there is first the reference by Mr Bulley to the proposed, limited, leasing arrangement leaving BOPE (and, admittedly TrustPower) "in control" if either party is unhappy with any marketing "push". In like form, papers for a BOPE Joint Venture meeting of 5 October 1999 record the following comment:The national deals achieved by TrustPower for all Telecom and Fletcher Challenge sites has resulted in continued requests for a bi-lateral metering agreement. Following discussions with Richard Tweedie, we have agreed tomove forward with the 20 Telecom sites in our region. This arrangement will allow a metering rental system to be set up, avoid stranded assets and keep us in the metering 'driving seat' during the period when we are not aware of TrustPower's timing for a mass marketing campaign into our region. (CB 3/698)[522] In our view, there can be little doubt – with reference to this evidence – that BOPE was consciously exploiting its ability to delay and deter entry by potential competitors both to put them off its market, and to use the time gained to attempt to win back customers. [523] That BOPE had a purpose of deterring competitors was, in our view, also effectively acknowledged by Mr Tweedie in his evidence and by Mr Farmer in his submissions. [524] Mr Tweedie said quite explicitly that he did not think it was BOPE's obligation to assist competitors. Similarly, there was no particular reason why BOPE should respond to requests for meter leasing agreements promptly, and it was up to the competitors to press BOPE to respond and make it a priority. [525] Similarly, Mr Farmer in his closing submissions referring to the strategy paper quoted from at para [504] above, made the submission that the note was clear on its face that BOPE's objective was minimal customer losses. He went on to submit:However, such objective is the rational objective of every customer-based firm which relies on customers for survival. Further, a policy that a firm will not make entry easy for competitors by subsidising that entry is entirely defensible on any view of competition policy.[526] In our view, that acknowledges that, as a matter of fact, BOPE had a purpose of deterring competition. [527] In this context, the comment of the Board of the Privy Council, as regards Telecom's impugned behaviour in the Telecom v Clear Communications case is pertinent, as it acknowledges that, in and of itself, anti-competitive purpose may not be particularly significant: what is significant for s 36 purposes is the proper characterisation of the behaviour by which that purpose is pursued.[528] We also note that we consider BOPE considerably exaggerated the implications for it of providing meter leases to customers. It did so consistently by reference to the phrase "establishing a meter leasing business", and then asserting a range of difficulties and disadvantages associated with such a course of action. We think those difficulties and disadvantages were exaggerated. [529] We note, in particular, that in evidence Mr Bahirathan acknowledged that, relative to the information BOPE was required to manage in any event, the only additional information that would have been involved with "establishing a meter leasing business" would have been to track lessees of meters as a result of switches. In other words, rather than charging an energised retail customer for meter leasing, BOPE would have had to record and charge the lessee of the meter for the meter – absent any electricity charges – and have a system to keep track of the parties who leased that meter over time. This would not appear to us, with respect, to be a particularly onerous matter. Furthermore, we note that in answer to a question from the Court, Mr Doidge acknowledged that the phrase "meter leasing business" could as accurately be described as establishing a meter leasing function. We think in fact this is a more accurate description of what would have been involved, and gives a better flavour of the complexity of the matter. [530] We therefore record, that had it been necessary for our purposes, our conclusion would have been that BOPE, in engaging in the range of conduct impugned by the Commissioner, had the purpose of restricting entry and preventing and deterring persons from engaging in competitive conduct with it.Section 27Arrangement or understanding[531] As already noted, BOPE has at all times accepted that it had a policy of not leasing meters to potential competitors during the period 1999 to present. Mr Farmer acknowledged in his submissions that BOPE did not take issue with the Commission's characterisation of that policy as an arrangement or understanding between the BOPE joint venture parties, save as to the involvement of the namedindividuals, who were a party to the arrangement only in their capacity as officers of the joint venture parties. [532] BOPE also acknowledged that it gave effect to that policy during the pleaded period. [533] As to the scope of the pleaded arrangement or understanding, the Commission pleaded only that it contained a provision that "BOPE should pursue a general policy that it would not agree to lease its non time of use meters in the Eastern Bay of Plenty to other intending retailers of electricity in that area". Crucially, the Commission did not plead any of the other strategic conduct canvassed earlier in our judgment in relation to s 36. The assessment under s 27 therefore focuses exclusively on the refusal to lease. The question for the Court's determination is therefore whether that provision had the purpose of substantially lessening competition, or had the effect or the likely effect of lessening competition. [534] Mr Farmer's primary submission was that s 27, which prohibits cartel or collusive conduct, was not intended to apply to a situation such as the present, in which the joint venture parties entered into an arrangement as to how the joint venture business would be operated. Mr Farmer points out that joint venturers are obliged to so collude by their obligations of good faith. Accordingly, it is only by an accident of provenance that the refusal to lease policy is impugned. [535] Mr Farmer refers to US authority in Texaco v Dagher 547 US 1 (2006) in which the Supreme Court held that joint venture pricing did not offend the price fixing prohibition. The Court made the point that joint venturers are for those purposes to be regarded as a single entity operating within the market. By analogy to the present case, BOPE should not be penalised for cartel conduct when the joint venture presented itself as a single entity to the outside world. The provenance and implementation of the impugned provision was therefore entirely analogous to the actions of a parent company directing a subsidiary, or to shareholders agreeing on a course of conduct for their company. [536] Within the scheme of the Commerce Act, Mr Farmer points out that s 31 provides a clear exemption for joint venturers in relation to price fixing. It wouldseem illogical that the same principle would not apply to s 27. To conclude, Mr Farmer submits that the Commission's real complaint is with the policy itself, not how it came about. Therefore, if the Court is not minded to find the policy itself anti-competitive and in breach of s 36, there is no good reason to find it in breach of s 27 simply because it was decided upon by two parties due to the choice of ownership structure. [537] While we acknowledge the force of those submissions, in our view this is simply a question of the scheme of the Act. Exemptions to s 27 are provided in s 44 of the Act. Joint ventures are not among the exempted ownership structures. Conversely, joint ventures are expressly exempted from the price fixing provisions. Given the very specific terms of the available exemptions in the Act we do not think it is the place of this Court to hold that s 27 does not apply to arrangements between joint venture parties. [538] In terms of that finding, however, and the fact that from 15 August 2001 onwards BOPE was a wholly owned subsidiary of Todd, our conclusion is that on and from 15 August 2001 the exemption in ss 44(1A) did apply and that, accordingly, s 27 could only apply up to that date.In a market[539] Section 27(2) prohibits giving effect to a provision that has the purpose, effect, or likely effect, of substantially lessening competition "in a market". It is therefore necessary, before the Court can assess the degree to which competition is lessened (if at all), to reach a conclusion as to the relevant market. [540] The Commission's case under s 27 was that the relevant market in which competition was substantially lessened was the Eastern Bay Retail Electricity Market. That is to say, the geographic dimensions of the relevant market were delineated with reference to the Horizon Area. The Commission argued that BOPE's policy of not leasing meters had the effect, the likely effect, and the purpose of substantially lessening competition in that market.[541] Just as it did in relation to the Eastern Bay Electricity Metering Services Market, the Commission relied on Dr Pleatsikas' evidence in defining the bounds of the local retail electricity market. Dr Pleatsikas' view was that the same cost- imposing conduct which he saw as giving rise to a local metering services market also gave rise to a local market for retail electricity. We have held that the evidence failed to establish the existence of a local metering services market as pleaded by the Commission. It follows from that conclusion that the Commission equally failed to establish the existence of a local retail electricity market. We reiterate our comments at [485] to [490] to the effect that it might have been possible to argue for the continuation of a local market based on the limited effect of market reforms, rather than on BOPE's allegedly cost-imposing conduct. However, we are similarly unable to reach a concluded view on that matter as it was not the basis of the Commission's case. [542] The result is that an essential element of the s 27 cause of action – the relevant market in which competition is said to have been substantially lessened – is not established by the Commission. [543] Nevertheless, if we are wrong in that, we also go on to discuss the remaining elements of the cause of action on the assumption, for argument's sake, that the Commission did establish the pleaded local market. [544] Furthermore, we note that BOPE pleaded in its statement of defence that the applicable retail electricity market was a national market. Accordingly, we also go on to discuss the s 27 cause of action on the basis of the national market argued for by BOPE.Effect/likely effect – local market[545] Even had the Commission established the existence of a local market for retail electricity, we think it follows as a natural consequence from our discussion of the s 36 cause of action that the policy of refusing to lease meters in and of itself did not have the effect, and could not have had the effect, of substantially lessening competition within the market for retail electricity. The allegation that the policyhad the effect or likely effect of substantially lessening competition is based squarely on the Commission's central assertion that the alternatives to leasing meters were not economically equivalent to leasing and that BOPE's policy of not leasing meters therefore imposed significant additional costs on potential competitors. We have held that the cost of those various means of obtaining metering services would converge to the point of indifference (see [403] to [440] above). Put another way, BOPE would always price leasing up to the point of indifference with other options. It cannot therefore be said that BOPE's refusal to lease meters imposed an additional cost on potential competitors. It follows that the policy did not, and could not, substantially lessen competition in the market for retail electricity.Effect/likely effect – national market[546] The same reasoning applies a fortiori if the market is assumed to be one for retail electricity at the national level. At a national level, the cost implications of the no leasing policy in the Eastern Bay of Plenty region could not be said to have the effect or likely effect of substantially lessening competition in the national market.Purpose – local market[547] The question of purpose is rather different. As noted by all three judges inANZCO, 'purpose' is not simply to be equated with 'effect' or 'likely effect'. Accordingly, a defendant firm may have the purpose of substantially lessening competition even where they do not achieve that effect, or were not likely to achieve that effect. Anderson J expressly notes the possibility that a defendant firm might be found to have had the purpose of substantially lessening competition albeit it was mistaken as to the actual or likely effect of the provision to which it gave effect. [548] Further, the mixed objective/subjective nature of the test entails two different approaches to any given case. On the one hand, the Court might conclude that the objective impossibility of the provision achieving a substantial lessening of competition necessitates the conclusion that it did not have that purpose, notwithstanding some subjective evidence tending to support an anti-competitive purpose (Glazebrook J's approach). On the other hand, the Court might find thesubjective evidence of anti-competitive purpose sufficiently strong that the impossibility of the provisions actually achieving that purpose is of lesser significance (the possibility referred to by William Young J). As we have said earlier at [340], our view is that all three judges in ANZCO acknowledged the potential application of both approaches, and differed chiefly as to the degree of emphasis they would accord to either the objective or subjective evidence. [549] Based on our discussion at [493] to [530] we consider that the evidence of BOPE's subjective purpose is sufficiently strong that, had the local market been established, we would have held that BOPE breached s 27(2) on the basis that its no lease policy had the purpose of substantially lessening competition in the local retail electricity market.Purpose – national market[550] We take a different view if one assumes the existence of a national market for retail electricity. The possibility of achieving a substantial lessening of competition relative to the national electricity retail market by implementing a regional no- leasing policy is so remote, given BOPE's insignificant share of that market, that we do not think the evidence of subjective purpose is sufficient to overcome it.Conclusion[551] We conclude that, because the Commission failed to establish the existence of its pleaded local market, the causes of action under s 27(2) fail. Proceeding on the basis of the national market pleaded by BOPE, the no leasing policy did not have the effect, or the likely effect, of substantially lessening competition in the market. We also do not think it is possible to conclude that the policy had the purpose of substantially lessening competition in the market if one assumes a national market. [552] If we are wrong, and the Commission did in fact establish the existence of its pleaded local markets, then our view would be that the no leasing policy did not have the effect or likely effect of substantially lessening competition but did have the purpose of doing so.Judgment[553] We therefore find that the Commission has not established that BOPE breached either of s 36 or s 27 of the Act as claimed. There will be judgment for BOPE accordingly. [554] The question of costs was not the subject of any submissions at trial and is reserved. [555] In the first instance, and in the absence of agreement between the parties on the question of costs, counsel are requested to file a joint memorandum proposing how the issue of costs might, on a procedural as opposed to a substantive sense, be dealt with. The Court does not, at this point, see it necessary to set any timetable for dealing with the question of costs._________________________ Clifford Jfor the Court