Victoria University of Wellington Law Review
Drawing on recent developments in Australian, United Kingdom and United States jurisprudence, Professor Korah casts doubt on the approach recently taken by New Zealand courts in one of the most controversial areas of competition law: the access to its facilities that a corporation in a dominant position must give to its would-be competitors. She argues that before imposing such obligations courts ought to be more sophisticated in assessing the economic effects of such obligations and especially the need to preserve an incentive to make the considerable investment required to create such facilities.
Professor Korah was the 1999 Chapman Tripp Fellow. This article is an edited version of a paper presented at the offices of Chapman Tripp during the tenure of the Fellowship.
It is lovely to be invited for a third time to New Zealand, where so many interesting steps have been taken to make the economy more efficient, despite the limited local demand that leads to many markets being concentrated. It has been a great pleasure to be welcomed by new colleagues at Victoria University and to meet again Jack Hodder, who contributed so markedly to my LLM seminars in London in the year dot minus three. I am delighted and gratified to meet such an august and knowledgeable audience, although I am somewhat intimidated by the thought that I know so little of your law and of your markets.
There is concern in many jurisdictions that firms found to enjoy substantial market power may be under a special responsibility to grant to their competitors access to essential facilities. Where a firm enjoying market power has some asset which a new entrant to a concentrated market must have or use, it is tempting to say that the market would operate better if the newcomer were entitled to access. On a static analysis, this may be true – there will probably be one more firm in the market, and more will be produced and sold at lower prices. On a dynamic analysis, however, it is important that the incentive to create the original asset not be reduced. The reduction in the incentive to invest in specific facilities in other industries may be more important than the static improvement in a particular market.
I am concerned by the propensity of antitrust authorities to take the static view and require access to be granted at a price at which the incumbent would prefer not to deal, thereby reducing the incentive to make the original investment. The particular investment will already have been made, but if antitrust authorities frequently compel access for the benefit of free riders wishing to compete downstream, the incentive to invest in facilities essential to other activities will be reduced.
New Zealand competition law was inspired directly by that of Australia as part of the arrangements for Closer Economic Relations (CER) and, in its turn, the Australian trade practices legislation was largely inspired by the experience in the United States and European Community.
Under section 2 of the Sherman Act in the United States, some of the cases went very far in finding that a firm with substantial market power was under a duty to grant access to essential facilities for the benefit of those wishing to compete with it downstream. Professor Areeda's seminal article "Essential Facilities: an Epithet in Need of Limiting Principles" has helped to stem the tide. Even where more than one person can conveniently use the facility, there are objections to requiring access. An obligation to supply not only reduces the incentive to make the original investment, it also reduces the incentive to duplicate or improve it where this would otherwise be sensible. Moreover, somebody will have to settle the amount of compensation to be paid, often on a continuing basis.
Areeda concluded with 6 principles:
(1) compulsory access should be very exceptional;
(2) a facility held by a single firm is "essential" only when (a) it is critical to the plaintiff's competitive vitality and (b) the plaintiff is essential for competition in a market;
(3) access should be required only if it would increase competition by reducing price or increasing output or innovation. This is unlikely when (a) it would chill desirable activity, (b) the plaintiff is not a competitor, actual or potential, (c) the plaintiff merely substitutes for the monopolist or (d) the monopolist already had the usual privilege of charging the monopoly price for its resources;
(4) even when these conditions are satisfied, denial of access is never unlawful per se. A legitimate business purpose saves the defendant. It is for the defendant to come up with a justification, but for the plaintiff to establish that it is insufficient;
(5) the defendant's intention is rarely illuminating, as every firm that denies access to a rival does so to limit competition. Only exclusion by improper means should be restrained;
(6) no remedy should require access on terms which the court cannot adequately explain or supervise.
The United States courts now seldom require access.
The New Zealand cases under section 36 of the Commerce Act 1986 start from the Australian precedent of Queensland Wire Industries Pty Ltd v The Broken Hill Proprietary Co Ltd and Another, decided under section 46 of the Trade Practices Act 1974. BHP produced some 97 per cent of steel products and supplied 85 per cent of them. It was the only producer in Australia and almost the only supplier of Y-bar, from which it made star pickets for rural fencing. It refused to supply Y-bar to Queensland Wire, which competed downstream, except at a price alleged to be excessive. The issue was whether this amounted to taking advantage of a substantial degree of power in a market for one of the prohibited purposes.
In the Federal Court, Pincus J denied that there was a breach of section 46 on the ground that BHP's conduct was not reprehensible. The Full Federal Court held that there was no market for Y-bar, over which BHP held market power, since it had never been sold. Both of these views were rejected by the High Court.
It has since been held that there must be a causal connection between the market power and the conduct. In Natwest Australia Bank Ltd v Boral Gerrard Strapping Systems Pty Ltd, French J had struck out a statement of claim under section 46 because there was no causal link between a refusal to supply and the alleged market power. Mason CJ and Wilson J stated:
In effectively refusing to supply Y-bar to the appellant, BHP is taking advantage of its substantial market power. It is only by virtue of its control of the market and the absence of other suppliers that BHP can afford, in a commercial sense, to withhold Y-bar from the appellant. If BHP lacked that market power – in other words, if it were operating in a competitive market – it is highly unlikely that it would stand by, without any effort to compete, and allow the appellant to secure its supply of Y-Bar from a competitor.
The High Court rejected the moral interpretation of the Act adopted by Pincus J in favour of one based on competition for the benefit of consumers. However, Dawson J went further and tried to distinguish competition on the merits from other exclusionary conduct:
The difficulty in determining what conduct constitutes taking advantage of market power and what does not, stems inevitably from the need to distinguish between monopolistic practices, which are prohibited, and vigorous competition, which is not. Both here and in the United States the search continues for a satisfactory basis upon which to make the distinction. For the most part, all that emerges are synonyms, which are not particularly helpful. Words such as "normal methods of industrial development", "honestly industrial", "anti-competitive", "predatory" or "exclusionary conduct" merely beg the question.
The best way of ousting a competitor is to provide better bargains for consumers than it does. Those better bargains are the objective of competition law. Such conduct may be exclusionary, but it would be ironical to condemn it. In all three systems – in the United States, Australia and New Zealand - there are many judicial statements that competition law is intended to protect competition or consumers, rather than particular competitors. Efficiencies by a dominant firm benefit consumers and the economy as a whole, even if they harm competitors. Their interest must not be ignored. Conduct that has no purpose or likely effect other than to restrict output and raise price has no compensating benefits to justify it.
For the last 20 years, the distinction between competition on the merits and other exclusionary practices has been very important in the United States. In Berkey Photo Inc v Eastman Kodak Co for instance, the second circuit Court of Appeal treated as lawful taking advantage of the efficiencies provided by vertical integration. Frances Hanks and Philip Williams have argued that the test of "taking advantage" used in Queensland Wire places efficiencies at the heart of section 46. On no other basis am I able to draw the distinction between competing on the merits and other exclusionary conduct. Even a firm without market power would refuse supply where breaking the chain of production or distribution increases cost or reduces quality.
However, I am concerned that the judgment, if the words of Mason CJ and Wilson J quoted above are followed literally, might reduce the incentive to make the initial investment. The issue did not arise in Queensland Wire. The incentive of being the only producer of star pickets can hardly have been an element in BHP's decision to build a steel mill. It was only one of nearly 60 products to be produced. The incentive to make a monopoly profit downstream, however, might well be important in other cases. Philip Areeda stated in his suggested principles that access should be required only if it would increase competition by reducing price or increasing output or innovation, which is unlikely when it would chill desirable activity – here in making the original investment.
As Hanks and Williams observed, where the efficiency of refusing access is based on avoiding disruption of the industrial process, it is clear that a firm without market power would refuse supply. So would a firm that wanted to avoid free riding, which might chill the incentive for dealers to invest in promoting its product. The same economic arguments apply when a duty to supply by a firm with market power might deter other firms from investing in other markets. Although in that case the supplier might well deal even if the market were more competitive.
Both Hanks and Williams and Areeda stressed the economic argument. Can the interpretation of "taking advantage" in section 46 be limited to include consideration of reducing incentives to the original investment?
In Queensland Wire, the High Court accepted that there may be legitimate reasons for a monopolist to refuse supply. It may, however, have been thinking of commercial matters such as the credit worthiness of the buyer, inadequate services that might be expected of a dealer and so forth rather than of incentives to the original investment.
Leave has been given to appeal to the High Court against the majority judgment of the Full Federal Court in Melway Publishing Pty Ltd v Robert Hicks Pty Ltd. Melway Publications publishes a volume of street directories of Melbourne each year. It now enjoys a substantial degree of power in the market for street directories for the Melbourne area and has a market share of about 85 per cent. From the beginning, some 30 years ago, it adopted the policy of distributing only through wholesalers which each dealt with a particular segment of the retail market – service stations, office stationers, authorised car dealers and automotive parts dealers. Notice of termination was given to one dealer, which then requested supply of 30,000–50,000 copies, which it would have sold mainly at the expense of the existing retailers. This was refused, and the dealer alleged that Melway Publications had taken advantage of its power in the market for the purpose of "deterring or preventing a person from engaging in competitive conduct in that or any other market".
The Federal Court held that this infringed section 46. No appeal was brought against the finding that the relevant market was street directories for Melbourne and that Melway Publications enjoyed a substantial degree of market power over it. The appeal to the Full Federal Court was limited to the application of the words "take advantage for" the prohibited purpose.
In discussing the requirement of "taking advantage" Heerey J, dissenting, pointed to the need to prove a causal link between the market power and the refusal to supply. Melway directories had been sold the same way 30 years ago, when Melway Publications must have had a small share of the market. Another publication of the same firm, Sydway, was also being marketed in a similar way, although its Sydney directories enjoyed a market share of only 10 per cent. Consequently, Melway Publications was doing only what it did when or where it had no market power. There was no causal link between any market power and the refusal to supply. Melway Publications was not taking advantage of its power in the market. Heerey J also pointed out that the sales that were foregone would be at the expense of its other distributors. The case was different from Queensland Wire since any hypothetical sale by BHP's competitors would have been at the expense of BHP.
The Judge went on to make good use of the article by Hanks and Williams and their view that efficiencies lie at the heart of the phrase "taking advantage". Where breaking a production chain would be costly, even a supplier with no market power would refuse supplies. As Hanks and Williams also observed, there are many kinds of efficiencies, including protecting distributors who invest in promotion from those who would otherwise take a free ride on that promotion and are able to sell at lower prices. Heerey J took this small step beyond Queensland Wire by accepting that an efficiency by encouraging a wholesaler or retailer to invest in promotion was relevant. The step was small because a non-dominant firm might have refused supply to protect its distribution system.
However, the majority of the Full Federal Court disagreed with Heerey J, and applied section 46 to protect a competitor rather than competition. The two majority Judges ignored the point that the sales requested might not have added to Melway Publications' total sales and that, consequently, it might have refused supply in the absence of market power. Unlike BHP, Melway held the copyright in its directories, so Hicks could obtain them directly or indirectly only from Melway. From the facts reported, it sounds as if the large market share of Melway Publications was due to efficiently producing and supplying what customers wanted – the very result the Act should promote.
I would like the courts to take an important further step and go beyond the words of Queensland Wire. I would like courts to accept that a refusal to supply may be justified on the ground that an obligation to supply would reduce the incentive to make the original investment, even if in a more competitive market supply would have been available. Each decision requiring supply makes the particular market more competitive but, on a dynamic analysis, chills the incentive to incur sunk costs in other emerging markets where the first mover may acquire market power. It will not be necessary for the High Court to decide on this further step in its judgment in Melway Publications, but it could.
The old problems of admitting and testing economic evidence, which would be helpful in deciding how much protection is justified by the need to induce risky investment have come to an end with the new rules of court for the Federal Court derived from the experience of the Trade Practices Tribunal (now renamed "the Competition Tribunal").
The Trade Practices Tribunal in cases based on section 45 has emphasised the importance of not reducing incentives to invest when considering whether to authorise a long-term contract on the basis of which considerable sunk investment had already been made. Authorisation is not possible when section 46 is at issue, but the judgment of Professor Brunt, Mr Aldridge, and Lockhart J is helpful in stressing the importance of incentives to significant, risky investment.
Re AGL Cooper Basin Natural Gas Supply Arrangements is the best articulated judgment of the Tribunal about exclusive provisions in long-term contracts. The contract provided for the exploitation of the Cooper Basin and the supply of the gas to Sydney for 30 years. The most important restrictions were:
(1) a take or pay provision: the buyer was to take at least 80 per cent of the maximum quantity of gas the seller was required to supply, or to pay for it;
(2) it agreed not to acquire gas elsewhere until it had taken the maximum quantity; and
(3) the buyer agreed to give the producers the chance to sell more than the agreed quantities, provided their terms were no less favourable than any alternative offers.
After noting that major gas exploration contracts where capital expenditure is amortized over 20 years or more were common for sensible commercial reasons, the Tribunal accepted that the earlier authorisation should not be withdrawn even though it might have preferred detailed changes to its terms.
It added some general comments at the end. It accepted that it was required to look prospectively and that there was no likelihood of the particular investment ceasing to be useful if the protection granted by the long-term contract was withdrawn. It said:
1. A distinction can be drawn between those long-term contracts that are necessary to sustain substantial, long lived, sunk investments, as in this matter, and those long-term contracts that create no such social utility, but are, rather, an instrument of foreclosure.
It referred to Re Tooth & Co Ltd; Re Tooheys Ltd as an example of a long-term sales contract providing no benefit to the public:
2. We have no doubt that, speaking generally, there is social value in the preservation of contractual commitments, and that society proceeds on the basis that, in the broad, contractual commitments will be met. The institution of contract goes hand in hand with the institution of property rights which, in turn, gives rise to appropriate incentives for behaviour that will be both efficient and fair. It is part of the web of interconnectedness that characterizes society as a whole. Accordingly, we reject the ACCC (Australian Competition and Consumer Commission) view that, on the future with and without approach, contractual commitments are "sunk" and can be disregarded.
3. We affirm, nevertheless, the appropriateness of the future with and without test. Choices today with respect to the use of resources relate to today and tomorrow, not yesterday. Our treatment of contractual commitments will affect the investment decisions of the future and the reliance that may be placed on long-term commitments.
4. It was a long-term contract, the Letter of Agreement, that in the beginning made possible a project that has given rise to benefit of the public. This does not mean that all the detailed terms were necessary, or that there is not the possibility of some anti-competitive terms having been introduced. Hence, we have to examine whether a less restrictive contract would, in Dr King's phrase, have "done the job". If a less restrictive contract would suffice, both in respect of past obligations and future obligations, the consequent detriment must be subtracted from the benefit created.
5. It is consistent with our methodology to revisit the past and ask the rationale for the various provisions included at that time. A clause that now seems anti-competitive might in earlier times have contributed to the viability of the project. On the other hand, the rationale for a clause may have been anti-competitive at the beginning.
6. We do not rule out the possibility that circumstances may have changed so as to diminish the importance of certain contractual protections to the viability of the investments. There is evidence of less need, with current and future developments in the gas and transmission markets, for contractual protection against opportunism and risk. As regards opportunism, commercial alternatives have emerged for both the buyer and seller, were either side not to fulfil its side of the bargain. As regards risk, there are beginnings of developments in trade in financial instruments and future commitments that can offer some protection. . . .
The Tribunal, concluded that circumstances had changed since the authorisation by the Commission in 1986 and it was necessary to decide whether the authorisation should continue. It concluded, however, that it should not remake the contract, but warned that in future parties should try to provide only such protection as could be justified as necessary to make the transaction viable.
I would also argue that since the access regime was introduced in Part IIIA of the Australian Trade Practices Act 1974 there is less need to impose an obligation to supply under section 46. The weakness of that argument is that the interpretation of the legislation is not clear. It is clear that access to goods, as opposed to services is not covered, and the distinction between goods and services is unclear. Moreover, the process of obtaining a right to access and the determination of the terms of dealing are complex and involve several institutions. Moreover, many conditions limit the power of the Minister – for example the service must be of national importance, there must be sufficient capacity to supply the provider, himself or herself and third parties.
Refusals to supply are elements under other sections of the Act, which will not be considered here. Section 47 forbids exclusive dealing when it has the purpose or, on balance, has or is likely to have the effect of substantially lessening competition. Resale price maintenance is forbidden per se by section 48. Sections 96 and 96A define resale price maintenance to include a refusal to supply a price cutter.
A duty to supply under New Zealand law might have arisen in two ways. However, the common law doctrine of prime necessity, was held by the Court of Appeal to have been supplanted by section 36 of the Commerce Act 1986 in so far as it applies to electricity. The problems of requiring access to facilities now are far more difficult or arbitrary than those that arose in the medieval cases. In so far as it holds that the doctrine of prime necessity no longer applies to electricity, I welcome the judgment. I would have preferred it to have followed the High Court, which held that the doctrine had been replaced in relation to all products. The appropriate charge for using the only wharf in a port could be easily determined by reference to prices in more competitive ports nearby. The appropriate charge for access to a national electricity grid or a local telephone loop is far more difficult to set and supervise.
The other possible source of a duty to supply is section 36. More than one New Zealand judgment has stressed the value of construing the Commerce Act 1986 in line with Australian Law. So, the Australian developments are important also in New Zealand. Moreover, the New Zealand tradition is to construe legislation in accordance with its purpose to a greater extent than has been customary in Australia. So it would be easier for judges to permit justification of a refusal to supply on the ground that to require supply would lessen the incentive to make the original investment and create problems in fixing and regulating the appropriate compensation.
Either party to a case to be decided by the High Court under the competition provisions of the Commerce Act 1986 may request that a lay member of the Court be appointed to sit with the judges. Most of the lay members are economists very familiar with problems of competition policy. So the economic considerations are often more clearly expressed by the High Court than in the three cases in the Court of Appeal, which are of varied quality.
The passage about competition on the merits from Dawson J's judgment in Queensland Wire set out above was quoted with approbation by Tipping J in the New Zealand case, New Zealand Magic Millions Ltd v Wrightson Bloodstock Ltd. He added:
I would venture the following proposition. It is not a breach of section 36 if a person, albeit with a dominant position, simply acts in a competitive manner. It would be an irony if such conduct could be attacked because it is competition which the Act is designed to promote. However if someone with a dominant position takes some action for a purpose prescribed by section 36 then clearly they are using their dominant position in a way which section 36 prohibits. The line may well be a fine one in certain cases but it will be a matter of fact and degree and ultimately of judgment in the individual case whether the line between what one might call legitimate competition and illegitimate competition has been crossed.
I agree with the first two sentences, but find the last sentence unhelpful. Judges may refer to the need for judgment when they cannot articulate the appropriate criteria. Competition on the merits is normally for the purposes listed in section 36. All competition is exclusionary. What should differentiate competition on the merits from other exclusionary conduct is that the efficiencies and incentives to efficient investment outweigh the exclusionary effects.
One of the earliest judgments delivered under the 1986 Act was Auckland Regional Authority v Mutual Rental Cars (Auckland Airport) Ltd. The ARA gave concessions to Avis and Hertz to rent out their cars from the terminal building on the basis that no third company would be granted a similar concession. Budget claimed that ARA was using a dominant position in the market of granting concessions for the purpose of restricting entry of other persons, such as Budget, into the market of renting out cars from the terminal building.
ARA was clearly dominant. Mr Gault, as he then was, cited some of the American cases requiring access to a bottleneck facility. The High Court accepted that a gateway facility is likely to beget a separate and identifiable geographic market. It held that ARA's motive in granting semi-exclusive franchises might have been to maximise its income from granting concessions, but it also had the purpose of deterring entry by Budget.
The Court concluded that the agreement to limit the concessions to two infringed section 36. It did not consider the important middle concept: the "use of" market power although, clearly, the grant of a semi-exclusive franchise might come within it. The High Court made it clear that the ARA still had discretion to admit or exclude Budget, but no shortage of space or other justification for excluding it had been alleged.
Where a body with governmental power has the power to grant franchises, it is particularly important that other entrants should not be excluded without good reason. I doubt whether the incentive to build the airport at Auckland was affected by the need to consider Budget's application on its merits. Indeed the need for a government licence may lead to super-dominance, which in Europe, may be a reason for imposing a more onerous duty to supply.
In Clear v Telecom Corporation of New Zealand, the question arose as to the charge that the incumbent privatised operator of telecommunications services might make to Clear, the new entrant, for connection to the local loop. The incumbent, Telecom Corporation, admitted that it was dominant over the local loop and had a duty to provide interconnection for a new entrant. The only issue was the criteria for charging. The High Court started its construction of section 36 by quoting from Posner J in Olympia Equipment Leasing Co v Western Union Telegraph Co:
Opinion about the offence of monopolisation has undergone an evolution. Forty years ago it was thought that even a firm with a lawful monopoly . . . could not be allowed to defend its monopoly against would-be competitors by tactics otherwise legitimate; it had to exercise special restraint – perhaps, indeed, had to hold its prices high, to encourage new entry. . . . Later, as the emphasis of antitrust policy shifted from the protection of competition as a process of rivalry to the protection of competition as a means of promoting economic efficiency ... it became recognised that the lawful monopolist should be free to compete like everyone else; otherwise the antitrust laws would be holding an umbrella over inefficient competitors. A monopolist, no less than any other competitor, is permitted and indeed encouraged to compete aggressively on the merits. . .Today it is clear that a firm with lawful monopoly power has no general duty to help its competitors, whether by holding a price umbrella over their heads or by otherwise pulling its competitive punches. . . .
The High Court went on to cite McGechan J and Mr Blunt in the first Port Nelson case, Tipping J in Magic Millions, as well as Queensland Wire for the propositions that sections 36 and 46 are intended to benefit consumers rather than competitors. In Queensland Wire the judges envisaged that even a monopolist may refuse to supply for a "legitimate reason".
The High Court in Clear v Telecom continued to quote Posner J in Olympia on the United States essential facilities doctrine:
The monopolistic-refusal-to-deal cases qualify rather than refute the no-duty to help-competitors cases. If a competitor is also a customer his relationship to the monopolist is not only a competitive one. The monopoly supplier who retaliates against customers who have the temerity to compete with him, by cutting such customers off, is severing a collateral relationship in order to discourage competition.
The High Court went on to quote the ratio decidendi of Queensland Wire and the link required between "use of a dominant position" and the prohibited purpose. Eventually it decided that Telecom was entitled to make a high charge to Clear for inter-connection, equal to the opportunity cost – what Telecom lost through not providing the call itself - that is the price it would have charged for the call less the cost saved by Clear performing part of the service.
The Court of Appeal quashed the decision on the ground that it enabled Telecom to charge a monopoly price, but their decision was, in turn, quashed by the Privy Council. The Board referred to efficiency, justifications, and the need to protect consumers. It also observed that it would be hard to establish that Telecom did not have an anti-competitive purpose when dealing with its hated rival, Clear; the focus must be on "use of a dominant position".
The New Zealand High Court was required in Shell (Petroleum Mining) Limited v Kapuni Gas Contracts Limited to decide whether a rather similar agreement to that for the Cooper Basin was contrary to sections 27 or 36. It recognised that:
there is now a recognisable trend for efficiencies to be considered in terms of their pro-competitive effect. . . . We were also persuaded that if the [sellers] retain part or all of the Kapuni field as a result of the court's decision they will have a greater incentive to prove up more reserves which, on the evidence, may well exist.
However, on the facts, it was not convinced that the exploration and efficiency arguments sufficed to overcome the foreclosure of competition which arises from the buyer's unconstrained market power. It said that had it been asked to consider the matter earlier in the life of the contract, it might have authorised the agreement for five or 10 years.
It is clear that the Court accepted that the incentive to make the original investment was relevant in deciding whether an agreement infringed section 27. The Court balanced the degree of foreclosure against the efficiencies made possible by reducing the risk attached to the investment.
In virtually all the cases since Hoffmann-La Roche & Co AG v Commission, the Court of Justice of the European Communities, often referred to as the ECJ, has used the same definition of "abuse" provided in article 82 of the European Communities Treaty (ex article 86):
The concept of abuse is an objective concept relating to the behaviour of an undertaking in a dominant position which is such as to influence the structure of the market where, as a result of the very presence of the undertaking in question, the degree of competition is weakened and which through recourse to methods different from those which condition [normal commercial operations] has the effect of hindering the maintenance of the degree of competition still existing in the market or the growth of that competition.
However, the ECJ has never attempted to define what is meant by competition on the merits. Early judgments, such as that in Commercial Solvents, condemned refusals to deal without considering whether the requirement to supply increased competition or discouraged desirable investments. In those days, most people thought that article 82 was intended to protect competitors, especially smaller firms. Only in the last few years has the ECJ stated that article 82 is intended to protect consumers rather than particular competitors.
The Commission of the European Communities, whose task is to administer the competition rules, has gone very far in applying an essential facilities doctrine, requiring non-discriminatory access to sea and airports even when there were other routes that might have served.
Magill marked the high-water mark of obligations to supply. Magill published one issue of a weekly comprehensive guide to the three television stations that were received in Ireland and Northern Ireland, but was successfully sued for copyright infringement by the television stations which each published an individual guide of its own programmes. The Court of First Instance (CFI) and the ECJ upheld a decision of the Commission against the television stations for abusing their dominant position over the information in the listings and preventing the production of a new product for which there were no substitutes and for which there was demand. Both courts said that the circumstances were exceptional.
The judgment caused a furore and has been narrowly construed by the Commission's legal service and by the Community Courts. In Oscar Bronner v Mediaprint Zeitung- und Zeitschriftenverlag GmbH and others, Advocate General Jacobs wrote an analytical opinion that was followed by the ECJ, although less theoretical reasoning was expressed in the judgment.
The Advocate General started from the proposition that most European legal systems recognise the right to trade freely and choose one's own customers. He added that:
. . . the primary purpose of article 82 is to prevent distortion of competition – and in particular to safeguard the interests of consumers – rather than to protect the position of competitors.
This view, which has seldom been expressed so clearly in Europe, has frequently been expressed in New Zealand and Australian judgments. From it Mr Jacobs went on to say that refusals to supply should not be illegal unless supply was needed to constrain market power in a connected market.
He warned against reducing the inducement to make a significant investment and explained the judgment in Magill partly on the ground that it was hard to justify copyright protection for programme lists in terms of incentives for investment. At the end of his opinion he also stressed that compensation must be payable when access is granted and that the amount is difficult to determine and may require continuing regulation.
He advocated careful balancing of the relative advantages and disadvantages of introducing competition into the neighbouring market against the reduction of the incentive to make the initial investment and the problems of continuing to regulate the compensation payable. Balancing these competing benefits has much to be said for it, rather than the all or nothing test of "using" or "taking advantage of" a dominant position, which does not consider the weight of each consideration.
Like Advocate General Fennelly in Compagnie Maritime Belge, Mr Jacobs suggested that, where a dominant position is particularly strong, there may be wider duties to help competitors. The Courts and Commission in Europe may be developing a concept of super-dominance. Many of the cases have involved a duty to supply by a firm enjoying exclusive rights under national law. Where a facility has been paid for by the state, it is particularly hard for a newcomer profitably to duplicate it. Moreover, the need to protect incentives to invest may be less important for investment by the government. Buyers from the government through privatisation must expect to be under a duty to give access.
Despite early judgments alleging a wide duty by a firm enjoying a dominant position to supply those competing downstream, the European Community authorities are coming to recognise that the function of article 82 is to protect competition and consumers rather than competitors. It is coming to be accepted, without much help from the wording of article 82 of the European Community Treaty, that a duty to supply should be narrowly defined on economic grounds.
In the United States and in Europe the obligation for a dominant firm to supply products or give access to a facility has been narrowed as it became clear that competition law is intended to help consumers rather than particular competitors. The passage trying to distinguish between competition on the merits and other exclusionary conduct that I have quoted from Dawson J in Queensland Wire and which was accepted by Tipping J in Magic Millions is important. I hope that it may be the start of retrenchment of the duty of a monopolist to supply in Australia and New Zealand.
It is clear from those cases that where interrupting an industrial process would be inefficient, a dominant firm is not "using" or "taking advantage of" its market power if it refuses supply. The other economic objections to imposing a duty to supply are:
(1) reduction in the incentive to the original investment;
(2) reduction in the incentive for exclusive dealers or retailers to invest in promotion;
(3) reduction of the incentive to duplicate the facility where this is practicable; and
(4) the problems of continuous regulation to determine the terms of dealing.
Heerey J has recognised the second objection as a justification for not supplying. My hope is that this will be accepted by the High Court in Melway Publications and then in New Zealand.
The light-handed regulation in New Zealand has left the parties to negotiate the terms of dealing, subject to the control of the courts under section 36. As recognised by the Privy Council in Telecom Corporation of New Zealand Ltd v Clear Communications Ltd:
In the absence of guidance as to the principles applicable, the parties were, as the High Court said, "negotiating in a fog". It is a regrettable fact that the decision of this appeal will only decide whether, in the past, Telecom has abused its dominant market position. It will not decide whether Clear's past stance in negotiations was reasonable, let alone fix the terms for interconnection.
Some argue that what may be needed is a regulator that can decide the terms of dealing in advance. The parties would not then be left to fix them without guidance as to the criteria. They would be able to talk to the regulator, which would be able to issue a decision binding on the firms negotiating access. Moreover, ex ante, a regulator can devise a price cap that leaves some incentive to cost pruning by the dominant firm. There is no way that a court deciding, ex post, whether a firm has infringed article 36 can do so. The drawback might be regulatory capture. The regulator might become too sympathetic to the views of the regulatees.
P Williams and Joshua S Gans have devised a formula for assessing compensation that would leave sufficient incentive to the original investment, which they suggest should be adopted and publicised by the National Competition Council under Part III of the Australian Act.
The European courts and Commission are beginning to consider that the extent of a duty to supply should depend on how dominant a firm may be. Advocate General Jacobs has also required the balancing of the static advantage of introducing competition to a market where there is a stranglehold monopoly against the reduction in the incentive to invest in other markets and the problems of continuing to regulate the compensation payable.
There seem to be grounds for the hope that efficiently producing what consumers want to buy is now seen to be the objective of section 36 of the Commerce Act 1986, and that the courts may eventually be sufficiently convinced of the need for appropriate incentives to limit the duty to supply either by requiring that the monopolist has a real stranglehold on an important related market, or by imposing terms of dealing that do not greatly reduce the incentive to the original investment.
Se livrant à l'étude des règles applicables en droit néo-zélandais, anglais et américains, le professeur Korah, se livre à la critique de l'approche retenue par des tribunaux néo-zélandais (notamment dans l'affaire Clear contre Telecom) dans l'un des domaines des plus controversés du droit de la concurrence : Quelles mesures concrètes doivent être imposées à une entreprise en situation de position dominante sur un marché donné pour que puisse toujours être assuré un accès au dit marché par ses concurrents potentiels?
Le professeur Korah, considère qu'il appartient aux tribunaux de faire preuve de circonspection et parfois d'invention lorsqu'il s'agira d'appréhender et de quantifier l'impact économique des obligations qu'ils peuvent imposer à l'entreprise en position dominante surtout lorsqu'elles induisent un coût économique et financier important.
[*] Chapman Tripp Fellow November and December 1999 at Victoria University of Wellington and Professor Emeritus of Competition Law at University College London.
I am indebted to Chapman Tripp and Victoria University of Wellington for encouraging me to write up in the form of an article the presentation I gave, and for constructive discussions and help in finding material. I am also grateful to the Queensland University of Technology for making their library available to me over the New Year and to many people for conversations about the importance of not reducing incentives to invest. My greatest debt is to Dr Maureen Brunt who suggested several cases at which I should look and made extremely constructive comments about an early draft of this article. None of these friends have seen the final version and cannot be responsible for it.
 See the excerpt quoted from the first Port Nelson case, below n 15.
 Philip Areeda "Essential Facilities: an Epithet in Need of Limiting Principles" (1990) 58 Antitrust Law Journal 841 ["Essential Facilities"]. See also Phillip Areeda and Herbert Hovenkamp Antitrust Law: An Analysis of Antitrust Principles and their Application (Aspen Law & Business, New York, 1996) Volume IIIA para 771-774.
 See the views of the Privy Council in Telecom v Clear below n 41.
 See below the long quotation from Posner J in Olympia cited by the High Court in Clear v Telecom below n 48. The recent United States cases were discussed by Donald Baker in "US: International Cooperation on Competition Law Policy" (1997) 25 International Business Lawyer 473; Robert Schuman Centre European Competition Law Annual 1998 (Hart Publishing, Oxford, 1998).
 Queensland Wire Industries Pty Ltd v The Broken Hill Proprietary Co Ltd and Another  HCA 6; (1988) 83 ALR 577 [Queensland Wire]. I have simplified the facts: BHP supplied the Y-bar to a wholly owned subsidiary that made the star pickets.
 As amended, section 46 provides:
(1) A corporation that has a substantial degree of power in a market shall not take advantage of that power for the purpose of –(a) eliminating or substantially damaging a competitor of the corporation . . . in that or any other market;
(b) preventing the entry of a person into that or any other market; or
(c) deterring or preventing a person from engaging in competitive conduct in that or any other market.
 Queensland Wire above n 5, 584, per Mason CJ and Wilson J.
 This view is hard to reconcile with Re Queensland Co-operative Milling Association Ltd – Proposed Merger Re Defiance Holdings Ltd – Proposed Mergers 1 and 2 (1976) 8 ALR 481, 517, where the Trade Practices Tribunal described a market as a place where actual or potential transactions may take place.
 Natwest Australia Bank Ltd v Boral Gerrard Strapping Systems Pty Ltd  FCA 511; (1992) 111 ALR 631 (FCA).
 Queensland Wire above n 5, 585.
 Queensland Wire above n 5, 585, per Mason CJ and Wilson J; 587, per Deane J; 600 and 602, per Toohey J.
 Queensland Wire above n 5, 593.
 See also Queensland Wire above n 5, 601, where Toohey J quotes Donald and Heydon Trade Practices Law: Restrictive Trade Practices, Deceptive Conduct and Consumer Protection (Law Book Company, Sydney, 1989) 221:
There is a stock distinction in discussing monopolisation legislation between predation and competition; between unfair rivalry and fair rivalry; between relying on one's market control and relying on one's efficiency; . . . and between deliberateness and the unexpected.
Efficiently producing what consumers want to buy may however be deliberate. In my view, the helpful reference is to efficiency.
See also Queensland Wire above n 5, 602 per Toohey J.
 Queensland Wire above n 5, 586 per Mason CJ and Wilson J.
 In Tru Tone Ltd v Festival Records Retail Marketing Ltd  NZCA 179;  2 NZLR 352, 358 (CA) [Tru Tone] Richardson J stated:
In terms of the Long Title the Commerce Act is an Act to promote competition in markets in New Zealand. It is based on the premise that society's resources are best allocated in a competitive market where rivalry between firms ensures maximum efficiency in the use of resources.
These two interrelated elements of competition in markets are crucial considerations under both section 36 and section 27.
The first paragraph of this important statement of the attitude of the Court of Appeal was quoted by McGechan J and Mr Gair Blunt in Union Shipping NZ Ltd v Port Nelson Ltd  NZHC 61;  2 NZLR 662, 699 (HC). They went on to say:
It is the permission of competition which the Court is directed to foster. Parliament, as a matter of policy, has decided benefits will flow from that course. Whether such is a correct economic or social analysis is not a matter for the Court. Within that objective, the particular objectives of ss 27 and 36 are clear. . . .Section 36, following in the footsteps of a tradition at least as old as the Sherman Act (USC 15 ss 1 – 7) recognises that even in competitive markets dominant positions do arise which in the end can generate anti-competitive activity. Accordingly it is intended to prohibit the use of such dominant position within a market for serious anti-competitive purposes. Such provisions are directed at the protection of the concept of competition as such. They are not directed at the protection of individual competitors, except in so far as the latter may promote the former. In the trade practices area, the 1986 Act clearly follows in a general way a number of approaches adopted in Australia under the Trade Practices Act 1974, which in turn in some areas pick up principles developed under United States antitrust legislation. Developments and approaches in those jurisdictions can be kept in mind accordingly. The legislation to no small extent breaks new ground, reflecting the increasing complexity and maturity of the commercial environment in this country in recent decades, and the demands which that has imposed. It is legislation of a type where the Court should not hesitate to adopt necessary purposive approaches, in line with Northern Milk Vendors Association Inc v Northern Milk Ltd  1 NZLR 530 paying due respect to legislative policy.
On the importance of protecting consumers rather than competitors, see also Ellis J and Professor M Brunt Clear Communications Ltd v Telecom Corporation of New Zealand Ltd (1992) 5 TCLR 166, 194 (HC); and in Telecom Corporation of New Zealand Ltd v Clear Communications Ltd  1 NZLR 385, 407 (PC).
 Berkey Photo Inc v Eastman Kodak Co  USCA2 548; (1979) 603 F 2d 263 (2nd Cir).
 In their outstanding article, Frances Hanks and Philip Williams "Implications of the Decision of the High Court in Queensland Wire"  MelbULawRw 4; (1990) 17 Melbourne ULR 437, 446 ["Implications of the Decision"]. For later developments, see also Philip Williams "The Exercise of Market Power: its Treatment Under the Australian and New Zealand Statutes" in David K Round (ed) The Australian Trade Practices Act 1974 – Proscriptions and Prescriptions for a more Competitive Economy (Kluwer Academic Publishers, Boston, 1994) 607, 610. His article discusses the other Australian cases on "taking advantage of market power".
 See the text from Queensland Wire above n 10.
 See principle 3(a) in "Essential Facilities" above n 2.
 See also Areeda's fifth principle in text from "Essential Facilties" above n 2.
 "Implications of the Decision" above n 17, 449-450.
 Queensland Wire above n 5, 586 per Mason CJ and Wilson J.
 Melway Publishing Ltd v Robert Hicks Pty Ltd (1999) ATPR 41-668 [Melway Publishing].
 Melway Publishing above n 23, para 51.
 Melway Publishing above n 23, para 17.
 See the passage cited in text to above n 10, or Dawson J at 593.
 See Maureen Brunt  Fordham Corporate Law Institute 357, 363 - 366.
 Re AGL Cooper Basin Natural Gas Supply Arrangements (1997) 19 ATPR 44 158, 44 216 [Re AGL Cooper]. See comment by A I Tonking "Long-term Contracts: When are they Anti-competitive?" (1998) 6 Competition and Consumer LLJ 13, especially 19-23 ["Long-term Contracts"]. Tonking sets out the particular terms more fully and criticises the Tribunal in several cases for failure to adopt consistent conclusions. The other judgments also paid attention to the hold up problems and the need for long-term protection of significant investments, especially when demand was not highly predictable.
 The others are Re A C Hatrick Chemicals Pty Ltd (No 2) (1978) 18 ALR 129 (Trade Practices Tribunal) (1978) 1 ATPR 17578, and Broken Hill Pty Co Ltd; Review of Notice re Koppers Australia Pty Ltd (1981) 3 ATPR 42,807 (Trade Practices Tribunal). On all three judgments of the Tribunal and on Shell (Petroleum Mining) Company Ltd v Kapuni Gas Contracts Ltd (1997) 7 TCLR 463 see the interesting article by A I Tonking above n 28 "Long-term Contracts". Tonking concludes that although the judgments consider relevant factors, the results are hard to predict. He would prefer a more systematic analysis of the exact terms of the agreement and its potential anti-competitive effects on the basis of the four cases he considers.
 Re AGL Cooper above n 28, 44, 216.
 Re Tooth & Co Ltd; Re Tooheys Ltd (1979) 2 ATPR 18 174 (Trade Practices Tribunal).
 The "with and without" approach is a comparison of what has happened after the transaction with what would have happened without it.
 Re AGL Cooper above n 28, 44,220.
 Enacted as part of the Competition Policy Reform Act 1995 (Cth).
 A third party requiring access to a service, which it cannot negotiate, may request the National Competition Council to make a recommendation that the service be declared an essential facility. If the Council decides to make such a recommendation to the Minister, the Minister may so declare the service with the consequence that third parties are entitled to access. If the parties cannot negotiate acceptable terms, the ACCC may act as arbitrator and settle the terms. From each of these steps there is an appeal to the Competition Tribunal and thence, on points of law, to the Federal Court. There are two other paths to access: an access undertaking given by the provider of a service and accepted by the ACCC and certification of State and Territory access regimes.
 Which means a restriction on the acquirer of goods obtaining similar goods elsewhere. A refusal to supply on the ground that the other party has or has agreed not to apply similar goods elsewhere counts as a refusal to supply.
 Vector Limited v Transpower New Zealand Ltd  NZCA 167;  3 NZLR 646 (CA).
 Mercury Energy Ltd v Transpower New Zealand (1998) 8 TCLR 554, 583 (HC) per Williams J and Dr M Brunt:
Whilst the Court is not unaccustomed to dealing with matters of complex economic theory, pricing methodologies and the like, and whilst it has mechanisms available to assist it in deciding these questions, it was common ground that the setting of pricing principles would be a protracted matter of considerable complexity, ill-suited to the adversarial process and one which would require to be regularly if not continuously repeated. It may not be overstating the position to say that if the Court became involved in such a process, it could become permanently involved. Given that the doctrine has never been invoked in this area in this way over the past dozen years or so, it would be a startling innovation for this Court now to assume and devote a considerable amount of resources to an ongoing supervisory if not regulatory role of major complexity in this sector of the electricity industry . . . . Had Parliament intended the Court to fulfil such a role, in our view it would have said so unmistakably clearly.
We therefore take the view that the Acts under discussion "conceal for the area [they] cover the texture underneath" and eclipse the doctrine of prime necessity.
 Section 36 prohibits a person who has a dominant position in a market from using that position for the purpose of restricting the entry of any person into any market, deterring or preventing any person from engaging in competitive conduct in any market or eliminating any person from any market.
 New Zealand Apple and Pear Marketing Board v Apple fields Ltd  NZCA 169;  3 NZLR 158, 164 (CA) per Cooke J. At 173, Richardson J spent some time analysing the Australian case law, as did the Privy Council:  NZLR 257, 262. See also Telecom Corp of New Zealand Ltd v Commerce Commission & Others  NZBLR 99,239, 102,358 and Auckland Regional Authority v Mutual Rental Cars (Auckland Airport) Ltd  NZHC 213;  2 NZLR 647, 669 (HC).
 In Telecom Corporation of New Zealand Ltd v Clear Communications Ltd  1 NZLR 385, 403 (PC) [Telecom v Clear], the Privy Council adopted as the law of New Zealand the ratio of Queensland Wire and said:
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.
 Although the law goes back much further, the desirabilty of adopting a purposive approach to complex new legislation was clearly stated in Northland Milk Vendors Association Inc v Northern Milk Ltd  1 NZLR 530, 537 (CA). Cooke P started his judgment by saying:
This is one of a growing number of recent cases partly in a category of their own. They are cases where, in the preparation of new legislation making sweeping changes in a particular field, a very real problem has certainly not been expressly provided for and possibly not even foreseen. The responsibility falling on the courts as a result is to work out a practical interpretation appearing to accord best with the general intention of Parliament as embodied in the Act – that is to say, the spirit of the Act. In doing so we have to bear in mind that freedoms such as that of the owner of a business to conduct the business as he sees fit are not to be restricted unless it clearly appears that this must have been the intention of the legislature. Obviously therefore a great deal turns on the need for the courts to appreciate and give weight to the underlying ideas and scheme of the Act. . .
As will be seen, the present case . . . is one in which we have found the long title is of some help, Whether or not the legislature has provided those aids, the Court must try to make the Act work while taking care not themselves to usurp the policy making function, which rightly belongs to Parliament.
 See text regarding Queensland Wire above nn 12 and 13.
 New Zealand Magic Millions Ltd v Wrightson Bloodstock Ltd  NZHC 887;  1 NZLR 731, 761 (HC).
 Auckland Regional Authority v Mutual Rental Cars (Auckland Airport) Ltd  NZHC 213;  2 NZLR 647, 678 (HC).
 Clear v Telecom Corporation of New Zealand  5 TCLR 166, 194 (HC) [Clear v Telecom].
 Clear v Telecom above n 46, 194.
 Olympia Equipment Leasing Co v Western Union Telegraph Co  USCA7 794; (1986) 797 F 2d 370, 375 (7th Cir) [Olympia].
 Union Shipping NZ Ltd v Port Nelson Ltd  NZHC 61;  2 NZLR 662, 766.
 Queensland Wire above n 5, 586.
 Clear v Telecom above n 46, 196.
 The Commerce Act does not make excessive pricing illegal. It prohibits directly only the use of market power to exclude competitors.
 Telecom v Clear above n 41.
 Telecom v Clear above n 41, 402, quoting from Olympia.
 Telecom v Clear above n 41, 403. The Board confirmed that "it cannot be said that a person in a dominant market position 'uses' that position for the purposes of section 36 unless he acts in a way which a person not in a dominant position but otherwise in the same circumstances would have acted".
 Telecom v Clear above n 41, 407.
 Telecom v Clear above n 41, 403.
 Shell (Petroleum Mining) Company Ltd v Kapuni Gas Contracts Ltd (1997) 7 TCLR 463 , but discussed in "Long-term Contracts" above n 28.
 Earlier the High Court had held that where an agreement does not substantially lessen competition, there is no need for it to be authorised. Indeed in Fisher & Paykell v Commerce Commission  NZHC 307;  2 NZLR 731 (HC), where the public benefit was insufficient to warrant authorisation, it held that the exclusive agreement did not infringe section 27.
In the Cooper Basin case the Australian Tribunal was not asked to consider whether the exclusive terms infringed s 45, only whether the Commission was right to withdraw the authorisation.
 Hoffmann-La Roche & Co AG v Commission  2 ECR 461.
 The numbers of nearly all the articles of the European Communities Treaty were altered by the Treaty of Amsterdam, which came into force on 1 May 1999.
Article 82 prohibits, as incompatible with the common market, the abuse by one or more undertakings of a dominant position within the common market or a substantial part of it in so far as it may affect trade between member states.
 This is a poor translation from the judgment originally drafted in French and translated into the authentic language of the case, which was German, meaning competition on the merits.
 Advocate General Fennelly did attempt to do so in Compagnie Maritime Belge Transports SA v EC Commission  4 CMLR 273 [Compagnie Maritime Belge].
 Istituto Chemioterapico Italiano Spa and Commercial Solvents Corporation v Commission of the European Communities  3 ECR 223,  13 CMLR 309,  CMR 8209 [Commercial Solvents].
 Radio Telefis Eireann and Independent Television Publications Limited v E C Commission  4 CMLR (Antitrust) 718 EUECJ C-241/91P; ,  ECR I-743,  1 CEC 400 [Magill].
 PMI/DSV OJ 1995, L221/34,  5 CMLR 220
 For example Tiercé Ladbroke v Commission T-504/93,  ECR II-923,  5 CMLR 309 and European Night Services v Commission cases T-374, 375, 384 and 388/94,  ECR II-3141,  5 CMLR 718,  CEC 955. Comment, Valentine Korah "The Ladbroke Saga" (1998) 19 ECLR 169.
 Oscar Bronner v Mediaprint Zeitung- und Zeitschriftenverlag GmbH and others Case C-7-97,  ECR I-7817,  4 CMLR 112 [Oscar Bronner].
 Advocates General are full members of the ECJ of equal status with the judges. Often their opinions provide fuller and more cogent reasoning, as they do not have to compromise. Independent judges, who never dissent or write individual opinions, may not be able to agree on the substance of their collegiate judgment but have to agree on its words, so much theory gets out, and even the minimum necessary to arrive at a result may be fudged. The judgment in Oscar Bronner does not bear the signs of a compromise. It follows the result of the Advocate General and may be treated as endorsing it.
 Oscar Bronner above n 68, para 58.
 See above Queensland Wire above n 5; Tru Tone above n 15.
 Oscar Bronner above n 68, para 61.
 Oscar Bronner above n 68, para 62.
 Oscar Bronner above n 68, para 63.
 Oscar Bronner above n 68, para 69.
 Compagnie Maritime Belge above n 63.
 In Portuguese Airports, OJ 1999, L69/31,  CEC 2168, the Commission stressed the exclusive right conferred on ANA to manage all the airports in Portugal and imposed particularly strict obligations not to discriminate in favour of the Portuguese airline.
 Oscar Bronner above n 68, para 65.
 Oscar Bronner above n 68, para 66.
 Although the High Court can avoid the issue.
 Telecom v Clear above n 41, 390.
 P Williams and Joshua S Gans "Efficient Investment Pricing Rules and Access Regulation". It and other papers can be found on <http://www.mbs.unimelb.edu.au/gans/research.html> .