4. Concerns with the Current Thresholds
1. As discussed in the previous section, the role of the thresholds is to identify for scrutiny those changes in market structures and behaviours that are likely to result in significant economic harm through the exercise of market power. This section examines whether the thresholds for mergers and acquisitions, and for the abuse of a dominant position are effective in achieving this role. Being effective requires that they meet the criteria outlined in Section 2 and that they are capable of being applied to all sources of market power.
The Threshold for Mergers and Acquisitions
2. New Zealand has a very narrow threshold for mergers in comparison with other OECD economies. It is likely that this threshold excludes from full scrutiny mergers that fall short of creating single firm dominance, but may nevertheless have efficiency reducing outcomes. Indeed, the threshold of single firm dominance has lead the industrial economist Greer to conclude that:
...(the Act) assumes a competitive market structure in the absence of pure or near monopoly. Even mergers creating monopoly are allowed if it can be shown that the barriers to entry are low, or the merger's detriments will likely be outweighed by efficiencies.18
3. This focus on single-firm dominance ignores any potential harm that may result from joint dominance in a market. Consequently it prevents mergers being scrutinized in terms of whether the resultant market structure will be conducive to collusion, whether explicit, or tacit. Indeed, Hay (1996) argues that in terms of the Commission's "safe harbour" of 60% with at least one rival with 15%, such a firm would have more market power because it would anticipate some tacit collusion from its larger rivals.
4. An example of the limited ability of the merger provisions to take into account the potential harm that can result from joint dominance, is the Commerce Commission's decision on 12 February 1999 that gave TransAlta a clearance for its proposed acquisition of 40% of Contact Energy.
5. The Commission, bound by the threshold in the Act, noted in its decision that: "the incentives for collusive behaviour may be strengthened by the proposed acquisition". However in paragraph 117 the Commission states:
...this exhibition of enhanced market power by a group of oligopolists would not be evidence of a dominant position having been acquired or strengthened...dominance under the Act refers to single firm dominance, and a dominant position in a market refers to a firm being able to exercise a "high degree of market control".
As well, the Commission, in paragraph 121-122 states:
Clearly, a merger between Contact and TransAlta would be very likely to trigger antitrust concern in the United States. However, the threshold for anti-merger policy in that country under the Clayton Act is a "substantially lessening of competition", which is a lower threshold than the dominance threshold under the Commerce Act...The Commission is bound by the Commerce Act and by court precedents on dominance set under that Act.
The Problem of Tacit Collusion in Oligopolistic Industries
6. Tacit collusion can arise in some oligopolistic industries. As the word "tacit" implies, even in the absence of direct communication, firms in some highly oligopolistic industries can misuse market power by moving their industry in the direction of greater coordination.19 In such industries firms can co-ordinate their behaviour by doing little more than observing and anticipating the moves of their rivals. The effect of such oligopolistic coordination is parallel behaviour, such as parallel price movements, that approaches the results associated with explicit agreement to set prices, output levels, or other conditions of trade.
7. Co-ordination can occur in oligopolistic industries, as the decisions of firms are interdependent. Each firm recognises that the sale of its product depends on the prices that its rivals charge, and that its rivals will react to any change in price, or output, that it makes. Given this interdependence, firms recognise that their profits will be higher when co-ordinated policies are pursued than when each firm looks after its narrow self-interest.
8. This is because even though an individual firm could increase its profits in the short term by reducing its price, this price reduction may cause rivals to respond by similarly reducing their prices. This results in lower profits for all firms. Thus a preferable strategy might be for all firms to increase price where the drop in revenue from the lost sales, is less than the total rise in revenue achieved through the price increase.
9. In this way, even in the absence of any formal collusion among firms, tight oligopolistic industries can be expected to exhibit a tendency toward the maximisation of collective profits. Such behaviour can approach the pricing and output outcomes associated with pure monopoly,20 and thus impose significant efficiency losses throughout the economy.
Collusion Is Not Inevitable in Oligopolistic Industries
10. Concentrated industries may be more conducive to co-ordination, but this does not mean that oligopolies will always behave anti-competitively. Aside from competition from rivals, competition can come from a variety of other sources including imports, substitute products, countervailing power, and potential competitors.21
11. What determines whether tacit collusion will be prevalent in an oligopolistic industry or not, depends on the product and its market. It has been known for a long time that under certain structural conditions co-ordinating firms have incentives to deviate.22 Often a firm can increase the quantity it sells dramatically by undercutting the cooperative price slightly. Even though the firm must accept a lower price to achieve this result, it may find the temptation irresistible when the lower price remains in excess of cost.
12. In terms of what some of these structural conditions are, generally it has been found that coordination will be less likely to be successful the greater the degree that:
- the product supplied is heterogeneous, complex or changing;
- the industry is dependent on large and infrequent orders;
- there are depressed business conditions;23
- firms can increase sales by winning several contracts with large buyers, or by signing long-term contracts with buyers;
- captive production, or non price competition (e.g. advertising and servicing), allows firms to increase sales secretly;24 and
- an anti-competitive price rise would be defeated by new competition, or deterred by the prospect of new competition.25
How Prevalent Is Tacit Collusion in the Economy?
13. In the absence of large numbers of industry studies it is difficult to know how prevalent tacit collusion is in the New Zealand economy, and the magnitude of the efficiency loss imposed. However, there is some industry specific research that tends to suggest that tacit collusion may occur in some markets, at least in ones for homogenous products.
Retail Petrol Market
14. The International Energy Agency (IEA), in its May 1997, report suggested that there was evidence of tacit collusion in the retail petrol market. This report raised concerns about the degree of market concentration, and price levels relative to other countries which highlighted significant concerns about the strength of competition in that market. The IEA concluded that there were: "...grounds for suspecting oligopolistic pricing behaviour by the four oil companies that dominate the industry".26
15. The NZIER (1996)27 provided earlier evidence that supported the conclusions of the IEA. From their study the NZIER concluded that the market for petrol in New Zealand was not a competitive one resulting in harm to both consumer welfare and industry efficiency. Comparisons of New Zealand retail petrol prices and the landed price of imported petrol showed an increase of about 1 cent per litre per year between 1990 and mid-1996 (since then the trend has been falling). According to the NZIER, for every 1-cent per litre increase in the price of petrol, the oil companies' combined revenue increased by approximately $27 million per annum.
16. With the entry of an independent retailer (Challenge) into this market, in April 1998, and the announced intentions of other independents to enter the market, the retail price of petrol decreased significantly in the areas where new entry occurred.
17. The indication that a fifth participant was essential to securing reductions in the retail price of petrol, illustrates the desirability of having a threshold for mergers and acquisitions broader than dominance. Around the time that the petroleum wholesaling industry was deregulated two companies were formed to compete in the retail market. In 1986 the Top Group was formed and acquired 21 retail stations in the North Island, including 8 in Auckland and 7 in Wellington. At about the same time Solo, a group of 13 retail outlets under single ownership also entered into a long supply contract with BP. Both Top Group and Solo provided some prospect of increased rivalry possibly by the import of refined products.28 However on 20 August 1988 Top Group was sold to BP.
Overseas Research on the Prevalence of Tacit Collusion
18. Research from other economies illustrates the prevalence and harm of coordinated behaviour in oligopolistic industries. In the United States during the 1920s and 1930s, the three tobacco companies selling 90% of industry output, maintained virtually identical list prices and discounts for 20 years. During the same period there were many instances in which the defendants' prices had featured lockstep increases even during the Great Depression when the companies' costs had fallen. Courts in the United States found no economic justification for the suspect pattern of behaviour, and thus inferred an agreement from the pattern of behaviour.29
19. Again from the United States, between 1962 and 1970 the three largest breakfast cereal manufacturers (who controlled around 76% of total sales) increased their wholesale prices per pound by 14 cents - much more than the differential any single company could sustain through individual pricing action. Price leadership facilitated the joint movement to higher prices.30 Industrial economist FM Scherer in commenting on this case said:
My analysis revealed that Kellogg's upward price leadership was usually followed by major rivals General Mills and General Foods. One of the case's many frustrations was the failure to have accepted into evidence documents revealing that Kellogg's price analysts made their recommendations assuming that rivals would follow Kellogg's price lead and therefore that market shares would be preserved.31
20. Greer (1989), cites a number of studies that illustrate the occurrence of coordinated behaviour, including:
- Moore (1986) estimated that when other variables are held constant, airline fares in highly concentrated city-pair markets (with fewer than five carriers) are 20% to 40% higher than fares in unconcentrated markets (with five or more carriers);
- Fraas and Greer (1977) from an analysis of 606 cases of illegal price fixing in the U.S., found that the highest frequencies of cartel activity occur in markets with approximately four to 10 firms. Below four firms, cartel activity is relatively low because tacit collusion or single firm dominance is controlling; and
- Gribbon and Utton (1986) found that over the period 1960-1981 in the United Kingdom, 18 out of 21 "concentrated oligopoly" industries (i.e. where the two leading firms shared 50% or more of a market) priced non-competitively.
The Australian Experience
21. The possibility that the dominance threshold is resulting in too little scrutiny of mergers is borne out by the Australian experience. When Australia was considering moving from a dominance test, to a substantially lessening of competition test, the then Trade Practices Commission (TPC)32 pointed to a number of mergers, that although permitted by the dominance threshold, raised efficiency concerns. In their view these mergers did not receive proper scrutiny. They included the following mergers.
- Coles/Myer - this merger resulted in a substantial increase in concentration in the market for retailing, and caused the removal of a significant competitor from that market and possibly prevented entry in the shorter term by another competitor.
- Ansett/East-West - this resulted in a reduction in competition particularly on major eastern trunk routes and price competition in NSW largely disappeared.
- ICI/Berger-British Paints - following the merger prices of architectural paint went up approximately 35% and prices of automotive paint rose substantially.
- Amcor/VisyBoard/Smorgons - Amcor and Visy Board acquired the Smorgons fibreboard container operation. Since the merger, fibreboard container prices increased substantially.
- Ampol/Solo - Solo was an aggressive price competitor in petrol retailing and had a major impact on the Australian retail market. The merger was followed by a sharp rise in petrol prices.
22. The TPC believed that each of these mergers should have gone down the authorisation track, so that all the relevant efficiency considerations could have been properly examined.
Section 27 Addresses Explicit Collusion Only
23. The potential for tacit collusion in the economy is of concern from a competition law perspective, because the Act has no mechanism to respond to it. The threshold for mergers does not allow consideration of joint dominance, and section 27 responds to instances of explicit collusion only.
24. In terms of section 27, case law to date suggests that for any liability to arise from section 27, communication between parties, or at least a conscious mutual commitment, is a necessary condition. This test is derived from the leading United Kingdom case - British Basic Slag Limited where Cross J said:
All that is required to constitute an arrangement not enforceable in law is that the parties to it shall have communicated with each other in some way, and that as a result of the communication each has intentionally aroused in the other an expectation that he will act in a certain way.33
The need for communication of the terms of the arrangement between the parties was endorsed by the UK Court of Appeal. The same standard has been held to apply in New Zealand as well as in Australia.34
25. However, in Auckland Regional Authority v Mutual Rental Cars (Auckland Airport) Ltd (1988), Barker J said:
An arrangement or understanding comes into existence as a result of some communication between the parties; the communication can however occur by written or spoken word the one to the other or by one observing and interpreting the other's behaviour.35
Although this reference to communication by "observing and interpreting" suggests that section 27 could be used to remedy tacit collusion, it has never been applied by the New Zealand courts to such conduct.36 The few New Zealand and Australian cases37 that have considered the meaning of an "arrangement or understanding" in an oligopoly context, have concluded that collusion is not an inevitable result of an oligopolistic market structure, and should not be inferred merely from parallel conduct which produces apparently anti-competitive outcomes.38 In other words outcomes which in economic terms can only be explained by collusion, would probably be held to be insufficient to satisfy the evidential requirements of an "arrangement or understanding".39
Should Section 27 Be Amended to Address Tacit Collusion?
26. The competition jurisdiction in the European Union allows behavioural remedies to be used to address tacit collusion. The competition jurisdiction in the United States also allows this but to a more limited extent. The question for those with a stake in the New Zealand jurisdiction, is whether section 27 should address tacit collusion as well as explicit collusion.
27. The major risk of making such an amendment is the negative impact that this would have on commercial activity and thus economic growth. Co-ordinated behaviour is not inevitable in oligopolistic markets. It is only likely to occur in the presence of certain factors, such as markets for homogeneous products. Where it is observed, uniform conduct may reflect a set of identical business responses, by a group of similarly situated competitors, to the same economic conditions. Indeed, price uniformity may be the outcome of fierce competition between few sellers. Oil wholesalers argued this was the case in the petroleum industry. In practice it will be difficult to determine whether price uniformity has resulted from competition, or from tacit collusion. Conscious parallel behaviour should not, therefore, be treated as illegal in itself. At most it should be treated as circumstantial evidence of anti-competitive behaviour.
Implications for Merger Policy
28. The concern then is not that oligopolistic markets will inevitably result in anti-competitive outcomes, it is that these markets may facilitate collusion both explicit and tacit. The Act has a behavioural mechanism that allows instances of explicit collusion to be addressed. However, no such mechanism exists for tacit collusion. Yet this type of collusion can result in significantly inefficient outcomes. Another option for addressing tacit collusion, rather than amending section 27, would be to broaden the range of acquisitions that are potentially subject to the merger/acquisition prohibition. This may allow for merger scrutiny in terms of whether or not members of a resultant highly concentrated market are likely to compete or coordinate.
29. Such scrutiny would be consistent with what occurs in most other competition jurisdictions. The European Commission, at least since the Nestle decision in 1992, has shown a willingness to control oligopolies by interpreting its merger regulations to include situations of joint dominance.40 In Nestle the Commission concluded that many of the classic conditions facilitating oligopolistic coordination were present, and that:
The maintenance or development of whatever competition there remains in that market therefore requires particular protection. Any structural operation restricting even more the scope for competition in such a situation has to be judged severely.41
Does a Small Economy Need a Very Permissive Merger Threshold?
30. In a small economy, high domestic market concentration will be a pervasive feature. Concentration levels in New Zealand are very high by international standards. This is due mainly to the need for industries to achieve minimum viable scale, as well as to achieve economies of scale. In contrast, in a large economy a firm with a much smaller market share can attain economies of scale.
31. Given the size of New Zealand's economy, and the high merger threshold, it is not surprising that relatively few mergers and acquisitions have been prohibited. Of the 211 applications received for business acquisitions, over the period January 1991 to December 1996, only 15 were declined.
32. However, simply being a small economy does not necessarily imply the need for a very permissive threshold for mergers. For example, scale economies can be achieved through exporting.
33. Khemani (1991) states that a broader threshold is desirable for a small economy. Khemani points out that an erroneous assessment of the economic effects of a merger by the competition authority is likely to have a relatively greater impact in a small, rather than in a large, economy. This is because in a small economy an overly permissive policy may entrench monopolistic and oligopolistic elements in the market. In such markets, although the merger itself may generate efficiencies, these efficiencies may be eroded over time through the incentives to reduce output and increase price, and/or through organisational inefficiencies.
34. Khemani also points to the Canadian Bureau of Competition Policy's experience where the long time periods required to realise efficiencies from mergers in highly concentrated markets, impose significant costs on that economy. In Canada most of the efficiency claims advanced by merging parties are of the genre of increased plant or product-specific economies of scale. To realise such economies proposals typically involve the rationalisation of product lines, plant closures, and increased specialisation. However, in monopolistic or oligopolistic markets where competitive rivalry is lacking, the time periods required to realise the "claimed" potential efficiencies can be long, and the foregone costs of the potential efficiencies, high. Although merging parties advance efficiency claims, these efficiencies are not weighed, in and of themselves, against the substantial lessening of competition arising from the merger.
The Section 36 Threshold
35. Consistent with a light handed regulatory regime, the Commerce Act places primacy on the use of its behavioural remedies, rather than its structural ones, to promote competition in markets in New Zealand. This balance is reflected in, by OECD standards, our permissive merger regime. However, because market conduct can be heavily dependent on market structure, a permissive merger regime demands the safety net of robust behavioural prohibitions. A concern is that section 36 is not effective in prohibiting anti-competitive behaviour.
The Scheme of Section 36
36. Section 36 of the Commerce Act prohibits a person in a dominant position from using that position for an anti-competitive purpose. What constitutes a purpose of an anti-competitive nature is specified in section 36(1)(a), (b) and (c). Section 36(1) states:
| "No person who has a dominant position in a market shall use that position for the purpose of - |
| (1) | Restricting the entry of any person into that or any other market; or |
| (2) | Preventing or deterring any person from engaging in competitive conduct in that or in any other market; or |
| (3) | Eliminating any person from that or any other market." |
37. Unlike most other restrictive trade practices prohibited under Part II of the Commerce Act, section 36 is largely interpreted as a single firm only threshold. The policy rationale behind section 36 is to prohibit the misuse of market power by the unilateral anti-competitive conduct of a dominant firm. In order to contravene section 36, three key components must be proven. These components are:
- Dominant position in a market;
- Use of that dominant position; and
- A purpose of restricting, preventing, deterring, or eliminating competition.
In order to identify the potential for improving the effectiveness of section 36, each of the fundamental components to the section is considered in turn.
Dominant Position in a Market
38. Section 3(8) of the Act specifies factors to be considered when determining whether a person42 has a "dominant position" in a market. To have a dominant position in a market, persons must be in a position to exercise a "dominant influence" over the production, acquisition, supply or price of goods or services in a market. Therefore, a pre-requisite to being dominant is that a person needs to be a supplier or acquirer of goods, or services. Section 3(8) of the Act specifies the necessary considerations to determine 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. These include:
- 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;
- the extent to which that person is constrained by the conduct of competitors or potential competitors in that market;
- the extent to which that person is constrained by the conduct of suppliers or acquirers of goods or services in that market.
39. In essence, these components reflect considerations of the barriers to entry and market share, elasticity of supply and elasticity of demand. Taken individually, these "tests" ask valuable questions in terms of the current market environment; the possibility, sustainability and timeliness of new entry to the market; and the extent to which suppliers and consumers individually and collectively can act as viable constraints on the behaviour of a dominant player. The Courts in a number of ways have interpreted these components of "dominant influence" under section 36.
Case Law on Dominance
40. As discussed in Section 2 of this document, the case law on "dominance" has developed in two phases. In the first phase an economic approach to the concept of dominance was taken. In the second phase a "dictionary definition" approach was adopted. To summarise the economic approach tended define "dominance" in circumstances where a person can act to a large extent independently of its customers, with an ability to give effect to "an appreciable change in the price and other aspects of supply".43
41. The second phase occurred when the Court of Appeal in Telecom Corporation of New Zealand Limited v Commerce Commission44 signalled a shift away from an economic approach to a more dictionary definition based methodology. As is mentioned in Section 2, the current state of the dominance test was considered in both the High Court and Court of Appeal in the Port Nelson litigation.45 In this case, the High Court accepted, in light of the Court of Appeal's decision in Telecom - "that the test for dominance was no longer a matter of prevailing economic theory" (McGechan J).
42. While the Court of Appeal appears to support the view that a non-technical dictionary definition approach is not the established test for dominance, there is some ambiguity in this area. It would appear from a specific comment from the judiciary, that if there is ambiguity this should be addressed by Parliament. Gault J in the same judgement stated:
The decision in Telecom v CC has been criticised as setting too high a level of market power so as to limit the value of the section as an effective regulator of abusive anti-competitive conduct. In this respect it must be noted that the Commerce Act was amended in 1994 without any change being made to ss3 (8) or 36; compare the change to the corresponding Australian s 46 of the Trade Practices Act in 1986 from "in a position substantially to control a market" to "a substantial degree of power in a market".
Policy Implications
43. The case law in this area appears to have set a high threshold for dominance under section 36. This is largely interpreted as a single firm only threshold, despite the fact that section 3(8) arguably supports a lower threshold.46 The addition of a plain English definition along side the economic definition appearing in the Act both raises the threshold for triggering section 36 attention and, arguably, creates uncertainty about the respective weighting the courts will apply from case to case.
Use of a Dominant Position
44. Under section 36, it is evidentially necessary to consider whether there has been use of a dominant position for one of the purposes specified in section 36(1)(a) to (c). Given that no case in New Zealand to date has turned solely on the issue of use of a dominant position, little jurisprudence exists on this concept. To test this component however, the New Zealand courts have predominantly adopted a view consistent with the Australian High Court's decision in Queensland Wire Industries Pty Ltd v BHP Co Ltd.47 In this case, the Court decided that the Australian equivalent of "taking advantage of" equated to "use", and that this component was to be viewed as a causal link between dominance and purpose, where purpose was the important issue for consideration.
45. The Court in Natwest Australia Bank Ltd v Boral Gerrard Strapping Systems Pt Ltd supported this perception of "use" as a link between dominance and purpose.48 Here the Court expressed "use" as a type of causal connection between the dominant position and the conduct at issue. In this case, the Court went on to state that:
"the connection may be demonstrated by showing a reliance by the contravener upon its market power to insulate it from the sanctions that competition would ordinarily visit upon its conduct."
46. However, in the case of Telecom Corp of NZ Ltdv Clear Communications Ltd,49 the Privy Council took an opposite view of "use". In this case, the Privy Council indicated that the concept of "use" was of greater importance than purpose, and decided that it is possible to infer purpose from the use of a dominant position, but not the converse. The Privy Council developed a "use test" whereby a firm does not use its market dominance if it acts in a way that a non-dominant firm in a contestable market would act. The Privy Council stated:
In Their Lordships' view 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.
47. This change in focus has however, not been adopted in subsequent decisions by New Zealand courts in the Port Nelson cases. Here, the courts reverted back to the Queensland Wire rationale of "use" constituting a neutral causal connection between dominance and the purpose of the conduct in question. In the Ministry's view, this interpretation is the original intent of Parliament.
Purpose of Restricting, Deterring, Preventing or Eliminating Competition
48. The final component in showing that a dominant firm has breached section 36 is to prove purpose. The current definition applied to purpose under the Act (section 2(5)(b)) requires that the anti-competitive purpose or reason was the "substantial" purpose of the dominant firm. A substantial purpose is considered to be material in nature. In addition, the courts have also favoured defining "purpose" in terms of an element of "object or aim" by the dominant party as opposed to mere intention.50 Therefore, intention to do an act which is known will have anti-competitive consequences is not sufficient to show purpose under the Act.
49. It is also clear from current case law that the "effect" of the action of the dominant party is not relevant to this section.51 While the courts have not required direct evidence of "purpose", in the case of section 27 actions the court considers objective aspects such as the wording of the provisions, the terms of the arrangement and the circumstances leading to the arrangement to infer "purpose". However, in the case of section 36 actions, the only behaviour from which the court can infer "purpose" is the conduct of the dominant party itself. This usually involves an inference that a party intends the natural and probable consequences of acts committed by them. However, this inference is usually rebuttable on ascertaining the subjective intention of the person concerned. Accordingly, imputing an anti-competitive "purpose" to a dominant party under section 36 requires a relatively high threshold of evidence.
50. Much of the jurisprudence on "purpose" revolves around whether a subjective or objective test should be applied when showing "purpose" under section 36. There has been no real consensus on this issue. In Union Shipping52 the court stated
Proof of purpose, in the nature of these cases will often turn upon inferences drawn from actions and circumstances, with a sprinkling of internal memoranda and correspondence. Protestations of inner thoughts which do not reconcile with objective likelihoods are unlikely to carry much weight. In many cases, (...) both objective and subjective standards are met.
51. Further, in Port Nelson v the Commerce Commission (1996),53 the Court of Appeal cast doubt on the need to decide whether an objective or subjective approach is intended. Gault J stated
...we heard argument as to whether the proscribed purposes are to be ascertained subjectively or objectively (...) Much has been written on this distinction which is generally unimportant in practice.
52. Accordingly, while the courts have tended to devote much time to the issue of whether "purpose" under section 36 connotes a subjective or objective test, the most recent case law suggests this debate is of little value.
"Purpose Test" Versus "Effects Test"
53. In order to lessen the high threshold attached to "purpose" under section 36, some commentators have suggested the inclusion of an "effects test". An "effects test" would assess whether the conduct alleged to be in breach will have the effect, or likely effect, of restricting, deterring, preventing or limiting competition. There are two key factors that caution against inclusion in section 36 of an "effects test". First, such a test would substantially expand the scope of conduct that would fall within the reach of section 36. For example, a dominant firm that sought to scale up capacity to achieve economies of scale and scope would come within the reach of section 36 should an effects test be implemented.54 This would increase uncertainty and tend to deter efficient commercial activity.
54. Second, such a test could increase the risk of judicial error because of the difficulties associated in determining whether or not conduct is in breach of the Act. A ruling would require knowledge of:
- What the defendant would, or could, have done if it were in a hypothetical competitive market;
- What the plaintiff would look like if it were a competitor in that hypothetical market; and
- The damage done to the plaintiff by the defendant in the hypothetical market.55
55. All these factors require assumptions and estimations about firms' structures and behaviour that would be difficult to predict with reasonable levels of certainty.
56. Conversely, it is argued that the "purpose test" inadequately distinguishes between anti-competitive conduct and strong competitive conduct, and that the adoption of a rigorous "effects test" would mitigate this problem. On balance however, the Ministry considers that an "effects test" would not assist in promoting competition in our markets.
Definition of Purpose
57. Section 46(7) of the Australian Trade Practices Act 1974 provides a clear statutory test for purpose. It provides that a court may determine that a firm has a relevant "purpose" even if the existence of that "purpose" has been ascertained only by inference from that firm's conduct, or the conduct of any other person, or from other relevant circumstances. No commensurate provision currently exists in New Zealand's Commerce Act.
58. Despite the fact that our Act does not contain an express provision of this nature, the courts have indicated that they will infer "purpose" from the conduct of the dominant person and the particular circumstances surrounding it. This approach of inferring purpose from the dominant parties conduct is illustrated in Union Shipping:
Proof of purpose, in the nature of these cases will often turn upon inferences drawn from actions and circumstances, with a sprinkling of internal memoranda and correspondence.
59. By adopting a similar statutory test for "purpose" as that embraced in Australia, it is possible that we would simply be codifying an existing test currently used by the courts in this area. Greater clarity and the potential ability to utilise Australian precedents on the issue could also result.
Judicial Interpretation Has Raised the Threshold
60. The courts' interpretation of these concepts since the Act's enactment in 1986 has led to a rather narrow and restrictive interpretation and application of the section than when the Act came into effect. In particular the use of a plain English definition of "dominance" implies a higher threshold than originally intended. There may be a need to redefine "dominance" in a way that focuses directly on market power and the factors that influence the extent of a firm's market power. Such factors may include market share, elasticity of supply and elasticity of demand.
A Widened Section 27
61. It is generally accepted that Parliament intended the focus of section 27 and section 36 to be on different types of behaviour. Section 27 was designed to focus on collective anti-competitive behaviour and section 36 on unilateral behaviour. However, since Port Nelson, the scope of section 27 may have expanded into the domain of section 36.
62. Although judicial interpretation has narrowed the scope of section 36, the Court of Appeal may have deliberately widened the scope of section 27 as a means of redress. It is generally accepted that the decision in Port Nelson widened the scope of section 27 allowing it to be used to challenge behaviour of dominant firms. This section typically applies to agreements between competitors for the purpose of substantially lessening competition. However, even though this was not the situation in the Port Nelson case, the Court of Appeal held that it was sufficient if only one of the parties to the contract had an anti-competitive purpose. This means that section 27 can now be used to challenge unilateral conduct - previously the domain of section 36.
Policy Implication
63. Although the judiciary may have found a solution to the problem of a weakened section 36, at least in the case of Port Nelson, it is unclear how successful that remedy would be over the full range of anti-competitive behaviour that can be engaged in by a dominant firm. Some section 36 conduct may not be caught by section 27 as there may be no contract, arrangement, or understanding.
64. In any case it is a second best approach to resolving the problem. This is because until the precedent is well entrenched, use of section 27 to resolve "section 36 type problems" will raise the level of uncertainty for market participants. The cost to participants of managing that uncertainty may be substantial. As well, if use of section 27 proves to be less comprehensive in its coverage than section 36 in the form intended by Parliament, the economy as a whole will suffer through the foregone benefits that a greater level of competitive rivalry would have secured.
65. Moreover, it could be argued that Parliament implicitly decided to place greater reliance on the Act's behavioural prohibitions than on its structural ones in Part III. This balance only works when the behaviour prohibitions are robust. When the regime is "out of balance", as it appears to be now, we run the risk of allowing the creation of substantial market power without equipping market participants with the full range of tools to respond to abuses of that power.
Should Section 36 Prohibit Monopoly Pricing?
66. Some commentators have suggested that monopoly pricing should be considered to be an abuse of a dominant position in addition to what is provided by Part IV of the Act. Currently, section 36 does not prohibit the charging of a "monopoly" price. This is because section 36 prohibits dominant firms from using a dominant position for the purpose of restricting competition; it does not prohibit such firms from using market power for purposes that are not anti-competitive.
67. Where monopoly power is "transitory", there are a number of factors that suggest that prohibiting monopoly pricing would be detrimental to competition and efficiency, these are that:
- monopoly pricing in itself does not impair the opportunities of rivals, rather high prices encourage the entry and expansion of rivals. That is, the existence of monopoly prices creates the incentives for other firms to enter the market. Entrants typically bring innovative products and processes and stimulate the former monopoly incumbent to improve the efficiency of its operations;
- prohibiting monopoly pricing denies monopoly profits to firms whose monopoly power was acquired through having superior skill, product and business acumen. Thus restricting monopoly pricing would eliminate the primary incentive for firms to exhibit such behaviour in the market. For instance, it would deprive shareholders of the returns commensurate with the level of risk taken by the firm, and the firm's management and employees would forego the higher earnings and salaries that would have reflected the success of their endeavours;
- over time it would discourage the firm's on-going innovative and cost-reducing efforts by eliminating the reward.
68. However, where an industry has natural monopoly characteristics, or otherwise has non-transitory high market power, monopoly pricing will not attract rivalry. Rather, in such industries monopoly pricing can be used to achieve anti-competitive purposes. One example of this potential is the interconnection price that can be charged by a dominant firm for competitors to gain access to an essential facility.
69. In 1994 the Privy Council sanctioned the use of the Baumol-Willig rule for pricing interconnection services in telecommunications markets. This rule states that monopolists are entitled to provide services to competitors at the same price they implicitly charge themselves, including any monopoly profits. In response to the Privy Council decision, the Government issued a media release in 1996 that stated:
....developments since the Privy Council decision demonstrate that the major telecommunications industry players do not support the Baumol-Willig rule as a satisfactory basis for interconnection pricing. The Government considers that the ...rule has the potential to lessen competition, thereby limiting the rate of introduction of new products and services and lessening the benefits to users.
70. This begs the question as to whether section 36 and the courts should be used to regulate interconnection prices where they contain elements of monopoly pricing. There are several factors that suggest that this would be an ill-advised strategy.
71. First, requiring the courts to rule on access price assumes that the courts are able to determine what a "reasonable" non-monopoly price is. This is a difficult, if not impossible task to impose on the courts. To set such a price the courts would require knowledge of all cost, demand, and technological variables and all of these are subject to constant change.
72. Second, even if courts could determine what constitutes a reasonable price, enforcing such a price would involve the courts in ongoing supervision. Our current judicial system does not contemplate the courts taking such a role in commercial affairs.
73. As well, apart from enforcement, there is an issue that a reasonable price fixed today may through economic conditions become the unreasonable price tomorrow. As the President of the Court of Appeal commented in the litigation between Clear and Telecom:
It may be regrettable that the Court cannot resolve the matter, perhaps painting with a broad brush, but the Act rightly does not contemplate this. We are not a price fixing authority.
74. The Government's general approach in response to this issue is through, some or all of, the following:
- the use of meaningful information disclosure;
- the use of price control and the threat of price control;
- ensuring the Commerce Act is consistently able to facilitate competition in markets across the economy; and
- ensuring that the wider body of regulation is consistent with achieving an open economy which facilitates a high level of innovation and technological change.
Back to Top