Check against delivery


The title of my talk today is “Section 46 – the great divide.” It is self evident; we can all see that s46 causes strong debates and emotion.

Today, my purpose is to show why this is the case, in my opinion, and as someone often forced into the middle of these debates.

I will also suggest that the Harper reforms offer an appropriate way forward.

I see three divides:

  • First, the divide between what the words of s46 mean to the wider public versus what they mean to competition insiders. It is bad public policy to have this situation.
  • Second, the divide between Australia and the rest of the world. Only Australia and New Zealand require proof of the nexus between market power and the conduct (the “take advantage);and few jurisdictions rely only on purpose. We are out of step with overseas law.
  • Third, the divide between the focus on “take advantage” and commercial and economic logic. This will be my most controversial point.

The insider / outsider divide

When I arrived at the ACCC I was immediately asked: “Do you think antitrust law should protect the competitive process or individual competitors?”  I was surprised that the Chair of the ACCC would be asked such a question. After all, everyone involved in competition policy knows it’s the former.

Now three and a half years into this position I understand better why this question is so often asked; it is driven by the precise wording of section 46, which I find is so often put to me. The text of the law basically says – “do not take advantage of your substantial market power for the purpose of (amongst other things) substantially damaging a competitor.”

On its face the wording of the section is directed at the impact of the conduct on individual competitors, rather than the impact of the conduct on the competitive process in the market.

Unsurprisingly, therefore, many small business people and organisations, many parliamentarians and people in the media, as well as many others, observe some behaviour by firms in a market, see that it takes away a large number of customers from a particular business (usually a small one), think ‘substantial damage to a competitor’, and therefore conclude that a ‘misuse of market power’ has occurred.

Yet I have also heard many times since arriving at the ACCC - “Well, we know the words say ‘substantially damage a competitor’ but we all know that the law really means ‘do not damage the competitive process.”

So there are these two very different views of s46. There is, on the one hand, an exclusive club, with members of the club knowing that section 46 means ‘avoid damage to the competitive process’.

On the other hand, those not in the club, the vast majority of the population, aren’t privy to this insight.

Insiders may well feel special, indeed clever, but this divide between common interpretation and true meaning is bad public policy.

This insider/outsider divide that has emerged contributes to a lot of unnecessary activity. The ACCC is often urged to take action under section 46 when harm has been done to an individual competitor, particularly if that competitor is a small business.

Supermarkets provide an illustration of this point. I am often told, including by a number of Parliamentarians, that when a supermarket opens in a new geographic area, the existing shops will be substantially damaged. I am then asked: what is the ACCC going to do about it?

The answer I give is that, on its face, such behaviour is not against the law, and nor should it be because more competition is being introduced.

As insiders, we all work on the basis that section 46 is about the competitive process. We also know that the ‘take advantage’ element is meant to provide the filter for distinguishing pro-competitive from anti-competitive conduct. Indeed, the courts have generally taken us in this direction.

But this is confusing to outsiders. Rather than appearing to use sensible logic, such as seeing whether competition is being enhanced or limited, the ‘take advantage’ reasoning can sound like a strange legal interpretation that stops the law working ‘as intended’.

The result is that there are many misguided views on what section 46 is meant to achieve, and the ACCC is very often urged to stamp down on conduct that section 46 should not prohibit.

It is for these reasons that the ACCC sees much merit in the recommendation of the Harper Review to reframe section 46 towards the purpose or effect of substantially lessening competition. This would directly align section 46 with the intent, which we all share, of protecting the competitive process rather than individual firms.

Yet as we know, this recommendation has met with resistance from certain quarters, and some claim that the proposed law change could inadvertently capture pro-competitive conduct by powerful firms.

The supermarket sector in particular has dominated the section 46 debate and, amazingly, the two major supermarkets have argued that, were the proposed changes implemented, they risk breaching the law if they were to open a supermarket in a market where they do not currently operate.

I find it curious that the main opponents of the Harper s46 change are suggesting that larger companies would be stopped from opening a new store in a new market. How can such a move be said to substantially lessen competition?

To most people, as I have already argued, the current drafting would seem aimed at stopping this activity, not the new Harper drafting.

It is my hope, and also my expectation, that the Harper Review’s recommendation will in time dissipate this ‘great divide’ between the insiders and outsiders. Indeed, I am encouraged that the Panel’s clear statement of what the law in this area should do may have already started such a process.

The Australia / international divide

There is much sensible general discussion about harmonising Australia’s laws with those applying overseas. I suggest section 46 is a great place to start.

Our section 46 could be brought more in line with the equivalent provision overseas, under three headings.


A minimalist approach to drafting, combined with a greater reliance on case law for interpretation, is apparent in some international jurisdictions.

Take the US approach for instance, where section 2 of the Sherman Act states succinctly:

Every person who shall monopolise, or attempt to monopolise, or combine or conspire with any other person or persons, to monopolise any part of the trade or commerce among the several States, or with foreign nations, shall be deemed guilty of a felony…

Similarly the EU takes a minimalistic approach to the framing of their legislative provision. Article 102 of the Treaty on the Functioning of the European Union says:

Any abuse by one or more undertakings of a dominant position within the internal market or in a substantial part of it shall be prohibited as incompatible with the internal market in so far as it may affect trade between Member States.

The provision then goes on to specify conduct which may constitute an abuse.

Case Law and guidelines then greatly assist interpretation. For example, these laws in essence focus on the competitive process, and actions by firms with a substantial degree of market power.

The Harper recommendations simplify s46 considerably.

No need for the disjointed nexus

There are very few jurisdictions (New Zealand is the exception) which approach unilateral conduct cases in the same manner as we do under section 46. Most jurisdictions do not require proof of the nexus between market power and the conduct itself (the ‘taking advantage’).

In other jurisdictions it is presumed that if a dominant firm engages in anti-competitive conduct (whether of a certain type and/or with a certain effect on competition), it will have abused its market power.

Re-focusing on effects on competition: the better lens

Delving deeper, it is widely recognised internationally that best practice should prohibit unilateral conduct by a dominant firm that has a deleterious effect on competition. Very few jurisdictions adopt Australia’s approach to solely focus on the purpose of the dominant firm to establish a contravention of unilateral conduct prohibitions.

More commonly, other jurisdictions adopt a hybrid approach where a contravention may be established if a dominant firm either:

  • engages in a type of conduct which is presumed to have an anti-competitive effect (e.g. there is no need to show the actual effect of the conduct) (form-based approach); and/or
  • engages in conduct which has an actual effect on competition, established through an analysis of the relevant market conditions (effect-based approach).

As recognised by the International Competition Network (ICN)

“…In reality, most jurisdictions apply a hybrid approach that combines a formalistic approach with varying degrees of analysis of effects, usually using rebuttable legal presumptions.”[1]

The approach taken in the EU is an interesting example.  Many have criticised the European approach over the years for using more of a form-based approach than an effects-based one.  This debate was addressed in a discussion paper and extensive public consultation which prompted the European Commission to publish guidance in 2008 setting out its enforcement priorities for abuse of dominance cases.

That guidance sets out an economic and effects-based approach to such cases and the European Commission made clear that it would prioritise those cases where the conduct of a dominant undertaking is liable to have harmful effects on consumers. 

In the US, an effects-based approach is taken to monopolisation, in that once the plaintiff establishes anti-competitive effects, most courts shift the burden to defendants to provide a legitimate business justification for the conduct. In most cases, courts will then determine whether the anti-competitive harm outweighs the pro-competitive benefits.

In essence, therefore, the proposed Harper recommendation on section 46 would move Australian law closer to international best practice on many fronts.

In the aftermath of the Competition Policy Review’s Final Report, Professor Harper has posed a question to those detractors of his Panel’s section 46 recommendation who are also multinational firms. He asks: How can they reconcile operating within other jurisdictions that do not require any ‘take advantage’, and that focus directly on the effect on competition, while saying they cannot do so here?

It seems to me a perfectly sensible question.

The logic divide

Monopolisation provisions, such as section 46, must provide a ‘test’ that when applied to the facts effectively distinguishes between:

  • firms with substantial market power acting in a manner to prevent, restrict or damage competition; and
  • firms with or without substantial market power competing vigorously through making superior offers, even though this will hurt their competitors.

On the one hand a test that fails to effectively make this distinction risks discouraging large firms from engaging in competitive conduct.

On the other hand, however, and often missed in the debate, a poor test risks discouraging smaller rivals and new entrants from innovating and ‘taking-on’ large firms. Ultimately consumers are the losers.

The test must also be clear and the results predictable. Firms with market power must be able to reliably predict if their business strategies or conduct are likely to fall within the provision.

Currently, the ‘take advantage’ element of section 46 attempts to provide a filter for sorting between pro-competitive conduct and anti-competitive conduct by a firm with substantial market power. As a filter, it is neither logical, effective or predictable.

A common approach to assessing whether a firm with substantial market power is ‘taking advantage’ of that power is to ask: would, or could, a firm behave in the same manner in the absence of a substantial degree of market power?

If it would, or could, the behaviour is not considered to involve a taking advantage of that power.

The failure of this counterfactual test to effectively and predictably filter pro-competitive conduct from anti-competitive conduct is at the heart of the reform of section 46 proposed by the Harper Panel.

There are three key problems with the ‘take advantage’ element of section 46 and the way it is implemented.

First, we need to predict the behaviour of a firm in a hypothetical world that can be difficult to imagine and agree on. Contemplating a market where all is the same, except the source of the firm’s market power is removed, is often fanciful.

It is not sufficient to argue there are cases, such as Melway, where the counterfactual test seemed to, eventually, get the right answer. It must get the right answer, if properly applied, in all matters, especially the vast majority of cases that do not make it to Court.

The challenges faced in applying the counterfactual test in the 0867 litigation in New Zealand strongly attests to its flaws.

Second, and fundamentally, it fails to address the key issue; the likely effect of the conduct on competition. Is the conduct likely to restrict, limit or hinder other firms from competing on their merits? This question is best addressed by focusing on the likely market outcomes from the conduct. The likely actions of a hypothetical firm in a hypothetical market would seem largely irrelevant.

Third, and most important, relying on the counterfactual test fails to recognise that the effects on competition of conduct by a firm with a substantial degree of market power can differ to the effects of the same conduct in a competitive market.

Unilateral conduct by a firm with a substantial degree of market power is much more likely to distort the competitive process than the same conduct by a firm without market power.

Indeed, and please let me stress, this would seem to be the point of monopolisation provisions.

Just because a firm without market power may find it commercially possible to engage in particular conduct does not remove the possibility that the same conduct, when engaged in by a firm with a substantial degree of market power, will foreclose competition or have some other anti-competitive purpose or effect.

In the Rural Press case, the majority of the High Court found that conduct involving the making of conditional threats by Rural Press, that unless Waikerie Printing withdrew from the Mannum areas (where Rural Press had substantial market power) Rural Press would introduce a rival newspaper to the Riverland market, did not involve a taking advantage of market power, even though they accepted the conduct involved a substantial lessening of competition.

The majority of the High Court stated:

To reason that Rural Press and Bridge took advantage of market power because they would have been unlikely to have engaged in the conduct without the "commercial rationale" - the purpose - of protecting their market power is to confound purpose and taking advantage. If a firm with market power has a purpose of protecting it, and a choice of methods by which to do so, one of which involves power distinct from the market power and one of which does not, choice of the method distinct from the market power will prevent a contravention of s 46(1) from occurring even if choice of the other method will entail it.

The Commission failed to show that the conduct of Rural Press and Bridge was materially facilitated by the market power in giving the threats a significance they would not have had without it. What gave those threats significance was something distinct from market power, namely their material and organisational assets. As the Full Federal Court said, Rural Press and Bridge were in the same position as if they had been new entrants to the Murray Bridge market, lacking market power in it but possessing under-utilised facilities and expertise.[2]

My point is that inquiring whether or not a new entrant would have been able to make the threat, as the High Court found the Act required them to do, does not assist in assessing the likely effect of the conduct on competition. The threat was credible, and hence the effect on competition in the Murray Bridge market substantial, because it was made by the monopoly provider in that market.

Inquiring what a firm without substantial market power has the capacity to do in the same circumstance, while being required by the current law, would seem irrelevant to the underlying economics and commercial reality of the situation.

Removing the ‘take advantage’ element, of course, necessitates the replacement of the current proscribed purposes in section 46 with a substantial lessening of competition test. Otherwise the provision would risk capturing pro-competitive conduct by firms with substantial market power that eliminates or substantially damages a competitor.

Some argue that the proposed Harper Panel change will create uncertainty and “chill” competition. I disagree.

First, removing the ’take advantage’ element eliminates a substantial source of existing uncertainty. Applying the counterfactual test to a set of facts can involve a range of guess work.  What will the market look like if you remove the source of the firm’s market power? How will firms behave in such a hypothetical market?  Will firms in such a market find it commercial to engage in the conduct in question?

Second, as these changes bring section 46 into line with other provisions of Australian competition law, they should, in my view, reduce uncertainty.

Courts and practitioners have extensive experience in assessing whether conduct is likely to substantially lessen competition through the application of other sections of Part IV of the Competition and Consumer Act 2010.

Firms large and small are subject to section 45 of the Act. This section prohibits a corporation from making a contract, arrangement or understanding that has the purpose or likely effect of substantially lessening competition. This section can potentially capture conduct that is largely unilateral in nature.

For example, contracts entered into by a firm to acquire all of an input necessary for the establishment of a major new competitor may have the purpose or effect of substantially lessening competition. Despite the contractual element of the conduct this conduct is unilateral in nature and could be captured by section 45.

The conduct in the Cement Australia case would seem to be such an example. The ACCC’s concerns in relation to shopper dockets are a different example where apparently unilateral conduct was likely captured by section 45.

The point is that firms, large and small, have for a long time been operating in an environment where they must assess whether their conduct is likely to substantially lessen competition. There is little or no evidence that I am aware of that operating within this law is deterring firms, large or small, from competing aggressively.

This point is often missed in our current debate.

Some may ponder why do we need section 46 if sections 45 or 47 can do the work. Courts have often found breaches of section 45 or 47 but no breaches of section 46. Cement Australia and Rural Press are two cases in point, as was Universal Music.

There are at least two reasons why we need a monopolisation provision.

First, there are circumstances where a firm with a substantial degree of market power may engage in conduct that is likely to substantially lessen competition that does not involve a contract, arrangement or understanding. For example, a firm engaging in predatory pricing to limit new entry and protect its monopoly position in a market may breach a monopolisation provision, but not section 45.

Second, it should be made clear that different rules govern behaviour of firms with a substantial degree of market power. As noted by the OECD:

“Unilateral acts by a firm with a high degree of market power are much more likely to distort the competitive process and ultimately harm consumer welfare than conduct by a firm that has no or little market power”.[3]

Put simply, a firm with a substantial degree of market power has greater ability and more to gain from conduct that restricts competition. Effective enforcement of laws targeting anti-competitive conduct by firms with substantial market power are a necessary foundation for the efficient functioning of our market economy.


I believe the Harper Panel’s proposed SLC test for Section 46 will reduce unfortunate debates, align our law more with overseas law, and make our law logical and in line with the rest of the CCA.

These are benefits worth having.

Thank you for your time today.


[1] International Competition Network (Unilateral Conduct Working Group), ‘Unilateral Conduct Workbook Chapter 1: The Objectives and Principles of Unilateral Conduct Laws’, p11 (Presented at the 11th Annual ICN conference Rio de Janeiro, Brazil, April 2012),

[3] OECD Policy Roundtables, Evidentiary Issues in Proving Dominance, 2006, page 7.