University of New South Wales Law Journal
|
|
Home
| Databases
| WorldLII
| Search
| Feedback
University of New South Wales Law Journal |
|
R IAN McEWIN[*]
The Dawson Committee[1] (‘the Committee’) was asked to
review some key provisions of the Trade Practices Act … in view of the significant structural and regulatory changes that are occurring in Australia that impact on the competitiveness of Australian businesses, economic development and affect consumer interests.[2]
However, the Dawson Review (‘the Review’) did not look at the structural and regulatory changes occurring nor even whether the Trade Practices Act 1974 (Cth) (‘Act’) has ever been appropriate to Australia’s circumstances as a small and open economy (‘SOE’). Instead, the Review simply describes the history of the Act and proposes some minor changes at the margins, principally, it seems, because ‘the Committee considers that the competition provisions have served Australians well’.[3]
But where is the evidence to support this statement? The Committee’s conclusions were drawn solely from (self-interested) submissions. The Secretariat came from a Department that has long had responsibility for the Act – hardly a body likely to find fault with past stewardship. Relying on a paucity of evidence about the Act’s impact is not surprising given the composition of the Committee. While all members of the Committee are distinguished in their field, no independent economist was appointed to the Committee. This is despite a general acceptance in the United States (‘US’) and Europe not only that the main goal of competition law is economic, but also that its administration involves increasingly sophisticated economic tools. And any assessment of the Act’s impact must necessarily involve sophisticated empirical investigation.
We can only speculate on the reasons why the Committee was set up in this way. One possibility is that the Committee was only meant to maintain the status quo and to tinker at the (particularly procedural) edges. Another possibility is that the government wanted to appoint a committee that acted like a ‘lightning rod’ for the complaints of various lobby groups – mainly small business. The Australian Competition and Consumer Commission (‘ACCC’) has performed this function admirably for the government in recent times.
Competition law is economic regulation that affects both the structure of markets and the conduct of market participants. The essential ‘mischief’ being corrected is the use, either by a single firm or firms acting jointly, of ‘monopoly power’ – often defined as the ability ‘to control prices or exclude competition’.[4]
But, as with any form of economic regulation, the effectiveness of competition law needs to be examined in terms of the extent to which its goals are achieved, after regulatory costs are considered. To determine effectiveness it is necessary to look at likely outcomes with and without regulatory intervention. Rarely is this done, if at all, in competition law. It was not done by the Committee.[5]
The Committee’s Terms of Reference suggest that the purpose of the Review was more concerned with promoting rivalry and opportunity (ie, competition) rather than economic efficiency. Economic efficiency focuses on the impact or efect of conduct on price, output and product quality (ie, consumer welfare). Rivalry is more concerned with the maintenance of effective or workable competition and as such is primarily concerned with market structure, including the number of competitors. However, promoting rivalry is inconsistent with promoting economic efficiency because it runs the risk of protecting inefficient firms in the name of competition by prescribing a minimum number of competitors.
The Review does not adequately resolve the conflict between promoting rivalry and economic efficiency. The Review simply states that
[i]n most circumstances maximising competition will maximise economic efficiency. Thus a test designed to prevent the substantial lessening of competition will generally be a good test for economic efficiency.[6]
Further, it asserts that:
Concentrated markets can be highly competitive. It may be possible to object to the structure of such markets for reasons of policy (the disappearance of the corner store, for example), but not on the grounds of lack of competitiveness.[7]
In essence, the Committee views the world though a structuralist lens, with effective or workable competition at the forefront of its view. Workable competition is concerned with economic welfare, but the Committee was not asked to confine itself to economic goals. Rather, the Committee was asked to reconcile a number of populist goals including whether the Act:
• inappropriately impedes the ability of Australian industry to compete locally and internationally;
• provides an appropriate balance of power between competing businesses, in particular, between small and large businesses;
• promotes competitive trading; and
• is ‘responsive to the transitional needs of industries undergoing, or communities affected by, structural and/or regulatory change and to the requirements of rural and regional areas’.[8]
The Terms of Reference are difficult, if not impossible to reconcile. For example, do laws dealing with price-fixing and other collusive activities impede Australian industry? On one hand, prohibiting cartels improves domestic consumer welfare by lowering prices and increasing output. Increasing domestic competition, by ‘teaching’ firms how to compete, may even increase international competitiveness.[9] However, domestic collusion may facilitate export cartels in industries with international market power.
What is an ‘appropriate balance of power’ in a competitive market? To some, the term may suggest ‘equality of power’. On this view, inefficient firms will perceive the ability of other firms to lower prices or provide better products as ‘unfair’. These imbalances of market power necessarily result from differences in economic efficiency. Penalising efficiency because it leads to an inappropriate power balance will certainly limit increases in concentration; however, it leads to less efficiency and so to a lesser ability of Australian firms to compete internationally and against imports.
When efficient firms drive inefficient firms out of the market, thereby increasing concentration, they take part in a competitive process that rewards efficient conduct. The goal of promoting competitive trading may be seen as fair and consistent with protecting competitors (particularly small firms). The Committee did not address these trade-offs directly. Rather, it emphasised the importance of protecting competitive processes rather than competitors.[10] In line with mainstream economic thinking, the Review states that redistributive goals are better achieved via other economic instruments (such as taxation or subsidies) rather than competition regulation.[11] The Committee should be commended for this stance.
Australia, like New Zealand and Singapore, is an SOE. This raises the question of whether the competition law of an SOE should differ from that of larger countries like the US or those in Europe. Economic structures inevitably differ between small and large economies; hence, size may be important to the way competition law is drawn up and administered. Small countries have limited numbers of disputes and limited resources by comparison. Therefore, to what extent should small economies copy legislation and case law from larger countries, as Australia did in 1974? Further, should small economies adopt a stricter rules-based approach with lower administrative costs, or should they be more flexible, given the particular circumstances in small economies where markets are more concentrated?
Competition laws around the world are based on three main models:
• the US model, which has been followed by Australia, NZ and (to some extent) Japan;
• the Japanese and German models, which have been followed by Taiwan, Korea and Thailand (which used the Korean model); and
• the European Community model.[12]
However, none of these approaches ‘represent a carefully planned or clearly articulated set of coherent policies; rather, each reflect[s] a variety of compromises and political adjustments …’[13]
The Committee was asked to examine the appropriateness of competition law to Australia’s economic circumstances. This should have included an examination of the Act in terms of its institutional administrability and suitability. This was not done. Instead, in one of its few references to economic conditions, the Committee noted that ‘[i]n a relatively small economy like Australia, the misuse of market power can be particularly detrimental to competition’.[14] Nevertheless, it decided that despite ‘some differences between Australia and other jurisdictions with substantially larger domestic markets, the competition provisions in Part IV have served Australia well’.[15] As previously noted, this conclusion was reached without any empirical evidence of the impact of the Act. Moreover, there is no empirical evidence on the impact of the Act, much less whether the Act provides net social benefits to Australia.
The Committee’s approach was not unusual – substantive rules are often examined in isolation – the implicit assumption being a perfect enforcement regime. But enforcement is neither costless nor perfect.
Given Australia’s size one possibility could have been to ask whether Australia’s competition law comprises
simple rules that filter the category of probably-beneficial practices out of the legal system, leaving to assessment under the Rule of Reason only those with significant risks of competitive injury.[16]
Designing competition laws with administrability in mind means distinguishing between:
(a) rules that are clear to business, regulators and courts and leave little discretion to decision-makers, but which may yield the wrong outcome; and
(b) broad standards that give considerable discretion to decision-makers, that enable better decisions, given clear goals and adequate information, but which could involve ‘political’ judgments that redistribute income rather than promote better economic outcomes.
There are costs and benefits to both. Rules are clear but narrow (or ‘bright-line’) and as they must change frequently to keep pace with change, the costs of making them are high. Also, it is often difficult to obtain legislative consensus over details, but they still provide better guidance to the well-informed. Some examples of rules include those deemed per se such as price-fixing. Resources are not required to determine whether the conduct is actually bad or not. Another example would be specific guidelines set by the regulator, such as market share thresholds set to determine whether a merger should be examined.
Rules reduce uncertainty both for business and the courts. However, rules can also deter pro-competitive conduct. This is a ‘Type II error’, or a ‘false negative error’ where efficient conduct is wrongly condemned. For example, a ban on resale price maintenance can stop pro-competitive conduct designed to stop free-riding. A per se rule reduces the possibility of a ‘Type I error’, or a ‘false positive error’ where inefficient conduct is not found, but the possibility of finding a Type II error is increased. A trade-off must necessarily be made between the two errors.
In the alternative, broad standards, because of the difficulties and costs involved in agreeing on precise rules, often provide great discretion to regulators. Judges are given the job of working out details, and it is therefore more important to ensure that judges make right decisions. This means that regulators and judges need to be more informed about the issues rather than the rules. However, unclear standards reduce compliance (because people discount the likelihood of being caught by a vague standard) and increase regulator and judicial discretion.
A ‘rule of reason’ is a standard. It considers whether conduct makes consumers worse-off, or impairs economic efficiency. A detailed examination creates business uncertainty because the regulator and/or court must determine the impact of the conduct requiring business to second-guess the regulator and/or court decision. Further, regulators and/or courts rarely have sufficient information: estimates and forecasts are made.
Standards necessarily embody vague statutory language. For example, a provision may prohibit ‘a merger that reduces competition’ without defining ‘competition’. Regulators and/or courts must then determine whether competition refers to the number of competitors, the intensity of rivalry irrespective of numbers, or the extent to which prices differ from costs (the economic meaning of market power). Broad standards reduce the probability of a Type II error but increase the probability of a Type I error. Determining the proper balance is difficult in practice.
How should economic conditions determine a country’s competition law? According to Gal, an economy’s size may ‘necessitate small but important changes to existing doctrines, legal presumptions, or modes of enforcement adopted from large economies’.[17] Gal argues that SOEs should be more relaxed about mergers because ‘[market] concentration is a necessary evil in order to realize scale economies’.[18] She suggests that SOEs should promote ‘cooperative agreements [which] may enable a group of firms to carry on an activity on a more efficient scale; to reduce information or transaction costs; to engage in expensive innovative projects; or to eliminate free rider problems’.[19]
At the same time she states that SOEs should
adopt a strict anticollusion and anti-exclusionary conduct policy. The limited number of firms that can operate in a small market necessarily increases interdependent conduct … where such collusive conduct becomes tacit collusion, a strict anticollusive policy should be applied … Similarly, a strict policy should be adopted toward exclusionary practices with no offsetting benefits, when practiced by dominant firms. Such a policy should particularly focus on deterring the creation and maintenance of artificial barriers to entry … Predatory pricing, for example, whereby a firm lowers its prices with a view of raising them once its competitor(s) exit the market, should be strictly prohibited.[20]
However, as in Australia and New Zealand, where resources are limited and legislation extensive and complex and there is considerable discretion given to the regulator, the likelihood of policy error and the ability of regulators and courts to pursue non-economic goals is increased. With regulatory error and self-interest, Gal’s conclusions do not necessarily follow. Issues of administrability may be more important in determining optimal laws. This point is illustrated by the following discussion of s 46. The Dawson Committee argued not only against an effects test but concluded that s 46 had been sufficiently reviewed.
SOEs have more concentrated markets and companies therefore have potentially greater market power, in comparison to companies in larger economies. However, SOEs have limited resources and this means that regulatory focus should be aimed at obtaining the greatest net return. The cost-benefit calculus may change over time.
Judgments about competition law matters are made with inadequate information about current conditions and imperfect crystal balls. This means that there are legitimate disagreements about many matters. In addition, like in any decision-making body, regulators need to confront their own incentives. Stigler states:
An efficient enforcement system … requires intelligent guides to the regulators, telling them which things are good and also important, which things are good but unimportant, and which are positively harmful. All too often the system of values, or incentives, is badly skewed. This is a fault, not of the regulators, but of the legislatures and executives who establish the incentives.[21]
The legal system may also provide incentives for strategic gaming by firms. Therefore, instead of ‘serving as the bulwark of competition, these institutions … become the most powerful instrument in the hands of those who wish to subvert it’.[22]
Competition laws are generally believed to serve the public interest. As a result, debate mainly focuses on the legal rules and cases, and little attention is given to the institutions involved. There is considerable debate in the US about whether competition law actually promotes the public interest. For example, DiLorenzo has found many of the industries allegedly monopolised by trusts before the introduction of the Sherman Antitrust Act[23] (‘Sherman Act’) (such as coal, lead, leather, linseed oil, liquor, petroleum, salt, steel) had experienced falling prices and increased output in the previous decade, thereby threatening inefficient firms.[24] Faith, Leaven and Tollison have found evidence that the antitrust policy process is valuable to politicians who can ‘sell’ protection from antitrust intervention.[25] Coate, Higgins and McChesney also provide evidence of how the selection of mergers to challenge is influenced by political pressures.[26]
It is the administrability of predatory conduct that creates a major problem for SOEs like Australia. The costs of administering such rules are high and include, not only the direct regulatory and judicial costs, but also the indirect costs such as loss of productivity due to executive time spent on competition law matters, compliance costs and regulatory costs including the costs of wrong decisions. The latter includes the costs of failing to detect conduct or market structures that reduce economic efficiency or consumer welfare (a Type I error) and stopping conduct or market structures that increase economic efficiency or consumer welfare (a Type II error). As Easterbrook has pointed out, the latter conduct is self-correcting in the market, the former is not.[27] Hence there should be regulatory caution.
Conduct regulation has become more complicated since the advent of the ‘post-Chicago School’ of economics. This creates greater difficulties for regulators because
post-Chicago methodologies are necessarily more complex, reflecting the greater complexity of observed markets. The problem of administrability is a serious one for antitrust, which is not a justifiable legal enterprise unless it makes markets work better than they would work in the absence of intervention. A more accurate model whose increased complexity increases the rate of error may be a poorer choice than a less accurate model which is nevertheless simple and easy to choose.[28]
The ‘abuse of dominant power’ provision (s 46 of the Act) is influenced, at least in part, by case law dealing with monopolisation, such as § 2 of the Sherman Act. Section 46 requires:
(1) ‘substantial degree of power in a market’: this is a market power filter. The economic rationale is that a firm without market power cannot inflict damage on competitors unless it is more efficient;
(2) ‘shall not take advantage of that power’: this is a conduct test – that is, did the firm do something it could only do with market power?; and
(3) ‘for the purpose of’ eliminating or substantially damaging competitors, preventing the entry of competitors, deterring or preventing competitive conduct – this is an intent test.
The conduct test is more problematic and a source of continuing difficulty for the courts and regulators. It is also the source of considerable business uncertainty. Section 46 contains a conduct test that depends more on the source of the power rather than its impact. This is not the situation in the US. In Brunswick Corp v Pueblo Bowl-O-Mat, Inc,[29] the Supreme Court decided that a plaintiff must prove ‘antitrust injury, which is to say injury of the type the antitrust laws were intended to prevent and that flows from that which makes defendants’ acts unlawful’.[30] Lopatka and Page note that:
Antitrust injury gave lower courts a practical way to implement the Sherman Act’s ‘consumer welfare prescription within the bounds of established substantive rules … While the Court has resisted overruling its substantive precedents, antitrust injury requires a determination of the law’s purposes and an examination of the effects of a challenged practice … This process of redefining the substantive law through an examination of antitrust injury affected the law of merger, predatory pricing, maximum vertical price fixing, and even price discrimination under the Robinson Patman Act.[31]
The Dawson Committee rejected an effects test for s 46. The effects tests in other parts of Part IV (ss 45, 47 and 50) were distinguished on the basis that these sections are
concerned with conduct involving competitive relationships between two or more corporations, whereas section 46 is concerned with unilateral anti-competitive behaviour on the part of a corporation with a substantial degree of market power. It is the behaviour which gives rise to the prohibition rather than its effect although, of course, the ultimate object of the section is to protect and advance a competitive environment and the competitive process rather than to protect individual competitors.[32]
However, the real problem with focusing on conduct is that it is impossible to determine whether conduct is good or bad without looking at its impact. The presumption, under the current approach, is that if there is sufficient competition then it will lead to competitive outcomes. But why not just concentrate on effects directly?
Whether conduct is competitive or not depends on the competitive impact; this differs from case to case. Using a conduct test without looking at the effects necessarily involves having a model of desirable conduct. Interventionists like to use the model of ‘perfect competition’ as a benchmark. In this model all the terms of competition are given and any discretion involves using market power – firms have no discretion. Section 46 targets discretion. As Gleeson CJ and Callinan J stated in the High Court appeal of Boral Besser Masonry (now Boral Masonry Ltd) v Australian Competition and Consumer Commission[33] (‘Boral’):
The essence of power is absence of constraint. Market power in a supplier is absence of constraint from the conduct of competitors or customers. This is reflected in the terms of s 46(3). Matters of degree are involved, but when a question of the degree of market power enjoyed by a supplier arises, the statute directs attention to the extent to which the conduct of the firm is constrained by the conduct of its competitors or its customers.[34]
In the Federal Court,[35] Heerey J said:
If the impugned conduct has a business rationale, that is a factor pointing against any finding that conduct constitutes a taking advantage of market power. If a firm with no substantial degree of market power would engage in certain conduct as a matter of commercial judgment, it would ordinarily follow that a firm with market power which engages in the same conduct is not taking advantage of its power.[36]
But what is meant by ‘no substantial degree of market power’? If perfect competition cannot be the proper ‘conduct’ benchmark because firms must have some discretion in the real world, then another standard must be found such as ‘workable’ or ‘effective’ competition. In an effectively competitive world firms are constrained to the maximum extent possible, given actual market conditions.
However, if workable competition is used as a benchmark, then determining acceptable conduct involves making a judgment about how many firms should be in the market. and what these firms would do in a hypothetical ‘effectively competitive’ market. This would provide the optimal amount of constraint. But this is a highly speculative exercise, involving not only determining the number of firms, but also predicting how they will behave over time. The benchmark changes with market circumstances, therefore, specific conduct may be acceptable in one market but not another. This creates considerable uncertainty.
In an SOE, predation tests should only be concerned with looking at the effect with and without the conduct. Firms without market power cannot have an impact. Firms with market power will have some impact (the greater the degree of market power). In deciding whether to investigate or take action, a regulator should be forced to consider the likely impact of intervening compared to the costs of intervention (including the costs of making a mistake).
Lawyers often prefer intent tests to establish liability because they are usually more concerned with the morality of conduct rather than economic impact. There are two general standards of legal intent.[37] General intent implies that the person knows they are doing a particular act, that is, setting price below cost price or setting a minimum resale price. Specific intent requires proof that the firm actually intended to drive a competitor out of the market, prevent their entry into the market or generally harm competition. It is clear that ss 46(1)(a), (b) and (c) of the Act refer to specific intent.
How should specific intent be proved? Section 46(7) states that purpose may be established either by direct evidence or inference. But should it come from subjective or objective evidence? Subjective evidence is often not required in other areas of the law, for example tort law.
In Boral, Kirby J argues that subjective evidence should be taken into account:
Deriving conclusions about the ‘degree of market power in a market’ and on whether a corporation had ‘taken advantage of that power’ involved deciding substantially objective questions. But identifying the corporate ‘purpose’ of any such conduct, necessarily involved (to some degree at least) estimates of the subjective will of the officers of the impugned corporation who acted on its behalf in the context of an objective analysis of the state of the market and the level of competition within it.[38]
However, Gleeson CJ and Callinan J said:
The danger of confusing aggressive intent with anti-competitive behaviour, in the context of alleged predatory pricing behaviour, was pointed out by the United States Court of Appeals, Seventh Circuit, in AA Poultry Farms Inc v Rose Acre Farms Inc. The court said: ‘Firms “intend” to do all they business they can, to crush their rivals if they can … Almost all evidence bearing on “intent” tends to show both greed-driven desire to succeed and glee at a rival’s predicament … Monopolists set price by reference to their costs … competitors set price by reference to the market’.[39]
What role should intent play in an effects test? If subjective intent is irrelevant on the basis that all firms try to drive each other out of business, what about objective intent? Determining intent objectively means judging whether a reasonable person would have believed that the effect of the conduct would have driven the firm out of the market or or stopped their entry into it. A subjective intent to destroy a competitor is of no concern if a firm does not have market power. Why? Because unless a firm has market power its conduct can have no effect on competition or on other competitors. Given that both an objective standard and market power can only be assessed by looking at the effect of conduct, why not concentrate solely on the effect of the conduct?
Evidence of intent can serve as a useful guide to the likely effect and decision-makers are in the best position to know the likely effects of their conduct.
The outcome of the Dawson Review is very disappointing. The Committee did not do what they were asked to do. They relied upon assumptions unsupported by evidence. Their analysis is weak and unrelated to Australia’s economic circumstances and institutional context. The Committee rejected an effects test in s 46 on extremely flimsy grounds, with little understanding of what control of substantial market power should be trying to achieve in an SOE.
[*] Dr McEwin currently works for the Singaporean Ministry of Trade & Industry assisting with the introduction of a general competition law. The views expressed here are his own and do not necessarily reflect those of the Ministry of Trade & Industry. He is also an Adjunct Professor of Law at both Bond University and the National University of Singapore. He is grateful to the Editor for incisive comments.
[1] Committee of Inquiry for the Review of the Trade Practices Act, Parliament of Australia, Review of the Competition Provisions of the Trade Practices Act (2003) (the ‘Dawson Review’).
[2] Dawson Review, ‘Terms of Reference’ <http://www.tpareview.treasury.gov.au/content/termsofref.asp> at 16 July 2003.
[3] Dawson Review, above n 1, 6.
[4] United States v EI du Pont de Nemours & Co[1956] USSC 87; , 351 US 377, 391 (1956).
[5] The US Federal Trade Commission has started examining whether their merger decisions were correct, but this is only a recent development.
[6] Dawson Review, above n 1, 32.
[7] Ibid 36.
[8] ‘Terms of Reference’, above n 2.
[9] Michael E Porter, Competition and Antitrust (2001) Institute for Strategy and Competitiveness <http://www.isc.hbs.edu/CanadianCompetitionPolicy_062601.pdf> at 16 July 2003.
[10] Dawson Review, above n 1, 7.
[11] David A Weisbach, ‘Should Legal Rules Be Used to Redistribute Income?’ (2003) 70 University of Chicago Law Review 439.
[12] John O Haley, ‘Competition Law for the Asia-Pacific Economic Cooperation Community: Designing Shoes for Many Sizes’ (2002) 1 Washington University Global Studies Law Review 1, 6.
[13] Ibid.
[14] Dawson Review, above n 1, 29.
[15] Ibid 34.
[16] Frank H Easterbrook, ‘The Limits of Antitrust’ (1984) 63 University of Texas Law Review 1, 11.
[17] Michal S Gal, ‘Size Does Matter: The Effects of Market Size on Optimal Competition Policy’ (2001) 74 Southern California Law Review 1437, 1476.
[18] Ibid 1456.
[19] Ibid 1461.
[20] Ibid 1471.
[21] George Stigler, The Citizen and the State: Essays on Regulation (1975) 171.
[22] William Baumol and Janusz A Ordover, ‘Use of Antitrust to Subvert Competition’ (1985) 28 Journal of Law and Economics 247.
[23] 15 USC §§ 1 7 (1890).
[24] Thomas DiLorenzo, ‘The Origins of Antitrust: An Interest Group Perspective’ (1985) 5 International Review of Law and Economics 73.
[25] Roger L Faith, Donald R Leavens and Robert D Tollison, ‘Antitrust Pork Barrel’ (1982) 25 Journal of Law and Economics 32.
[26] Malcolm B Coate, Richard S Higgins and Fred S McChesney, ‘Bureaucracy and Politics in FTC Merger Challenges’ (1990) 33 Journal of Law and Economics 463.
[27] Easterbrook, above n 17, 17–23.
[28] Herbert Hovenkamp, ‘Post-Chicago Antitrust: A Review and Critique’ [2001] Columbia Business Law Review 257, 336–7.
[29] [1977] USSC 14; 429 US 477 (1977).
[30] Ibid 489.
[31] John E Lopatka and William H Page, ‘Antitrust Injury and the Evolution of Antitrust Law’ (2002) 17(1) Antitrust 20, 20–1.
[32] Dawson Review, above n 1, 79.
[33] [2003] HCA 5; (2003) 195 ALR 609.
[34] Ibid 632.
[35] Australian Competition and Consumer Commission v Boral Ltd [1999] FCA 1318; (1999) 166 ALR 410.
[36] Ibid 440.
[37] See, eg, Ronald A Cass and Keith N Hylton, ‘Antitrust Intent’ (2001) 74 Southern Cal ifornia Law Review 657.
[38] Boral [2003] HCA 5; (2003) 195 ALR 609, 683–4.
[39] Ibid 626.
AustLII:
Copyright Policy
|
Disclaimers
|
Privacy Policy
|
Feedback
URL: http://www.austlii.edu.au/au/journals/UNSWLawJl/2003/15.html