Commissioner of Competition
Abuse of Dominance Under the Competition Act
Federal Trade Commission/Department of Justice
Hearings on Single-firm Conduct
September 12, 2006
(Check against delivery)
Firstly, I would like to thank you for the opportunity to come here today and speak to you on Canada's competition policy and enforcement with respect to single-firm conduct. As the Commissioner of Competition, I am responsible for the administration and enforcement of the Competition Act. Under the Competition Act, single-firm anti-competitive behaviour is captured by the Abuse of Dominance provisions, found in sections 78 and 79 of the Act. I will outline the Competition Bureau's approach to enforcing the Abuse of Dominance provisions and the necessary elements for a successful application under the Act. I will also be discussing the last Abuse case that went before the Competition Tribunal, and finally, I will briefly touch on some of the challenges we face when enforcing section 79. But first, a bit of background about antitrust enforcement in Canada.
The foundation of competition policy in Canada dates back to 1889, when the Canadian Parliament passed the country's first legislation to protect competition. Ninety-seven years later, after significant legislative reform, competition policy took its current form with the Competition Act coming into effect. The most significant feature of this piece of legislation with respect to single-firm conduct was the decriminalization of monopoly. For all intents and purposes, the high burden of proof and the demonstration of public detriment required by the criminal monopolization provisions made the previous law unenforceable. The new Competition Act changed all that. Abuse of a dominant position was now a reviewable matter. A reviewable matter is one that is not inherently bad, but subject to review by the Competition Tribunal on the basis of whether, on balance, anti-competitive conduct has substantially lessened or prevented competition, or is likely to do so. Recognizing that economics is a foundation of the Competition Act, the Canadian Parliament considered it important that the body responsible for the interpretation and adjudication of the abuse of dominance provisions be well versed in this area. As a result, the Competition Tribunal was created; it is composed of both judges and lay members with backgrounds in economics, accounting or business. The unique structure of the Tribunal allows it to interpret the complex economic and legal issues brought before it-- from mergers to refusal to deal to abuse of dominance. And I believe that we have all seen that an abuse case is rarely anything but complex. The Tribunal's expertise aids in its interpretation of the fine line that often exists between competition on the merits and anti-competitive conduct.
The Competition Act came into force when Canada still had a relatively small protected economy, and it has continued to play a role as the Canadian economy evolves, often through consolidation of industries, such as forestry or mining, to meet the challenges of a global marketplace. It is only once a firm becomes dominant in its relevant market, that the firm's behaviour is open to examination under section 79. While the Act recognizes that big is not necessarily bad, the exploitation of that bigness to exclude, discipline or predate competitors is, where this exploitation results or is likely to result in anti-competitive effects.
The Act outlines three necessary elements, all of which must be met, to conclude that an abuse of dominant position has occurred. Firstly, it must be that one or more persons substantially or completely control, throughout Canada or any part thereof, a class or species of business. Secondly, that person or those persons have engaged in or are engaging in a practice of anti-competitive acts, and lastly that the practice has had, is having or is likely to have the effect of preventing or lessening competition substantially in a market. So, not only must a firm possess market power and be engaging in some type of anti-competitive behaviour, the Commissioner must clearly demonstrate that there has been or will likely be a significant impact on competition as a result of the behaviour. This test leads us to take a limited number of cases as we--consistent with the spirit of section 79--err on the side of non-intervention.
Having explained the background and some of the theory behind the abuse provisions, I would like to discuss how the Competition Bureau assesses allegations of abuse of dominance, and the Competition Tribunal's interpretation of the elements required under section 79. Most of the points that I will be addressing can be found in the Bureau's published "Enforcement Guidelines on the Abuse of Dominance Provisions", which provide a transparent outline of the Bureau's approach to the enforcement of section 79.
The first step taken in determining whether or not an abuse of dominance has occurred is relatively simple if you think about it -- there are no chicken and egg comparisons required here. Without dominance, no abuse, as defined by the Act, can be at issue. As previously indicated, we must first establish that a firm is, in fact, dominant.
Our analysis begins with the definition of a relevant product market. The Bureau considers a number of factors in its determination, most importantly, close substitutability. The geographic market must also be defined. Here the Bureau will consider factors such as evidence of foreign competition, imports and transportation costs.
Once the product and geographic markets have been defined, the law requires a determination of market power. This requirement is fundamental to the success of an application under section 79. The Tribunal has clarified that high market share, together with barriers to entry, will typically be sufficient to support a finding of market power. A prima facie conclusion of market power may be made on the basis of a high market share alone, but factors such as barriers to entry, excess capacity, and countervailing powers also bear weight in the Bureau's assessment. To date, the cases brought before the Tribunal have all included respondents which possessed very high market shares in the market(s) defined by the Tribunal, in excess of 80% in all examples. In the Abuse Guidelines, the Bureau states that a market share of less than 35% will normally not give rise to concerns of market power, while the Tribunal has indicated that a market share of less than 50% cannot be considered a prima facie conclusion of market power. Whether a firm with a market share falling below 50% would be found to exhibit market power remains to be tested.
Once dominance has been established, the Bureau must determine whether the practice engaged in by the dominant firm is anti-competitive. Simply being dominant in a market is not sufficient to attract liability under the Act. A business must engage in more than an isolated act to constitute a practice; this could mean engaging in several acts of the same (or similar) nature or several acts of a different nature. Assessing when behaviour is anti-competitive is complex--some acts could have a pro-competitive business purpose and not an anti-competitive purpose, for example, the introduction of a new brand.
Section 78 provides a non-exhaustive list of anti-competitive acts. The section references acts such as the pre-emption of scarce facilities or resources required by a competitor for the operation of a business, margin squeezing or requiring a supplier to sell to only certain customers. The Tribunal has found other acts not listed in the section, such as the use of long-term exclusive contracts, to be anti-competitive when engaged in by a dominant firm. In order to be found anti-competitive, the behaviour engaged in must have a predatory, exclusionary or disciplinary purpose vis-à-vis one or more competitors. The Tribunal does not require the Commissioner to prove subjective intent to meet this test, but rather that the "overall character" of the act(s) in question reveals a predatory, exclusionary or disciplinary purpose; this is determined by considering factors such as the reasonably foreseeable or expected effects of the act(s), any business justification and any evidence of subjective intent. Often the circumstances of the case lead the Tribunal to conclude that the respondent should be deemed to intend the effects of its actions.
For example, in a case called Laidlaw, the Tribunal found that the acts engaged in by Laidlaw could only be interpreted as being targeted towards its competitors. The respondent in that case had acquired competitors and imposed onerous no-compete clauses in the purchase agreements; utilized long-term contracts with highly restrictive clauses; and intimidated both customers and competitors through threats of litigation. In assessing all of the facts in this case, the Tribunal had no difficulty concluding that Laidlaw had engaged in a practice of anti-competitive acts in the relevant markets.
In each potential abuse case, once dominance and a practice of anti-competitive acts have been established, the Commissioner must still convince the Tribunal that "the practice has had, is having or is likely to have the effect of preventing or lessening competition substantially in a market". This requirement ensures that the Bureau examine the effect on competition as a whole, not just taking into account the repercussions of the practice on a specific competitor. In assessing the effect on competition, the Tribunal will examine the degree to which the anti-competitive acts preserve or enhance the dominant firm's market power, e.g. through the creation or enhancement of barriers to entry or expansion. While the issue of substantial lessening of competition has been considered by the Tribunal, it has not yet had the opportunity to comment on a substantial prevention of competition.
The Tribunal has noted in Tele-Direct, a case concerning telephone directory advertising, that where a firm has a high degree of market power in a market, even an act that has a small impact on the competitiveness of a given market may be considered substantial.
In assessing the impact of a practice on competition, the Bureau uses a "but for" test, namely, but for the anti-competitive practice in question, would there be significantly greater competition? This test has, recently, been endorsed by our Federal Court of Appeal, in the Canada Pipe decision, which I will discuss shortly. Under this standard, the question is not simply whether the relevant market would be competitive in the absence of the impugned practice, nor whether the level of competitiveness observed in the presence of the impugned practice is acceptable. Rather, the question is whether, absent the anti-competitive acts, the market would be characterized by, for example, materially lower prices, greater choice or better service.
Requiring a linkage between an act and an anti-competitive effect also requires that the Bureau consider all potential reasons for the maintenance or enhancement of market power and isolate the effects of the anti-competitive act in question. Thus, section 79(4) compels the Tribunal to consider, for example, whether the practice is a result of superior competitive performance. This is not the same as the efficiencies defence, included in the Competition Act's merger provisions. The Bureau, as stated in the Abuse Guidelines, takes the position that superior competitive performance is only one factor to be assessed in determining the cause of the substantial lessening or prevention of competition. It is not a justifiable goal for engaging in an anti-competitive act.
I shall now speak about some of the remedies available in Canada, where an abuse of dominance has occurred. Before litigating an abuse of dominance case, the Bureau will approach the dominant firm whose conduct is being investigated and seek a voluntary change in its behaviour to address the Commissioner's concerns about the act(s) in question. Where possible, alternative case resolutions are preferable to extensive litigation before the Competition Tribunal.
Assuming litigation is pursued, once the Tribunal has found that an abuse of dominance has occurred, it may make an order prohibiting the respondent from further engaging in the impugned practice. Where it feels that competition will not be restored through the issuance of a prohibition order, the Act also empowers the Tribunal to, in addition to or in lieu of the prohibition order, direct any respondent to the abuse application to undertake any action, including the divestiture of assets or shares, as are reasonably necessary to overcome the effects of the practice in the marketplace. As can be seen, the only remedies available to the Tribunal are injunctive (with the exception of AMPs against the conduct of domestic airlines). We are on record, supported by others such as the OECD, that a lack of financial consequences for dominant firms found to have abused their position is a significant shortcoming of the current legislation. This shortcoming is all the more acute in light of the fact that only the Commissioner is able to apply to the Competition Tribunal under section 79 and civil actions for damages are unavailable to injured parties.
Clearly, there is a fine line between anti-competitive and pro-competitive behaviour and the effects-based analysis in our abuse of dominance provisions recognizes that reality. Considerably more often than not the Bureau, after conducting an examination, decides to not take action because it is not convinced that there has been substantial lessening or prevention of competition resulting from the behaviour concerned. For example, in 2003, the Bureau discontinued an inquiry it had launched into the Quebec beer industry in 2000. Several Quebec microbreweries complained that they had suffered significant losses in product sales due to the business practices of two major breweries. Specifically, the microbreweries criticized the major brewers for exclusive dealing and abusing their dominant position by engaging in various anti-competitive acts. Following our investigation, the Bureau determined that these major brewers accounted for nearly 90 percent of sales of beer in the province and clients were bound by potentially anti-competitive contract clauses, however, there was insufficient evidence to confirm that the breweries= practices at issue substantially lessened competition. Seeing no clear negative effect on competition, we chose not to pursue this matter.
There has been limited case law on section 79 over the last 20 years. There have only been five contested cases to go before the Tribunal under section 79. Our latest contested case, the Canada Pipe case, brought some important developments with respect to our Abuse of Dominance provisions. Canada Pipe, a Canadian company, produces and sells cast iron drain, waste and vent (DWV) products through its Bibby Ste-Croix division. The practice at issue in this case was Canada Pipe's Stocking Distributor Program ("SDP"), fairly described as a loyalty rebate scheme. In contrast to a volume-based discount, under the SDP distributors of Canada Pipe's DWV products obtain quarterly and yearly rebates as well as significant point-of-purchase discounts, in return for stocking exclusively cast-iron DWV products supplied by Canada Pipe. Except for losing the yearly and quarterly rebates, there are no penalties attached to opting out of the SDP. It was alleged that the SDP enhanced and preserved, to a significant degree, Canada Pipe's market power in three relevant product markets. The Tribunal found that Canada Pipe was dominant in the relevant markets, however, the SDP, though a practice, was not anti-competitive, and, regardless, did not substantially lessen or prevent competition. Consequently, the Competition Tribunal dismissed the Commissioner's application under section 79, the first such dismissal in the five applications made by the Commissioner.
The Tribunal's decision was appealed to the Federal Court of Appeal and in June, our appeal was allowed and the case was remanded back to the Competition Tribunal for consideration. The Federal Count of Appeal's decision in Canada Pipe provided some much needed clarification on the application of section 79 and the appropriate tests required by the legislation.
As previously indicated, section 79 sets out three distinct elements that must be shown to exist before a finding of abuse of dominant position can be made. The Federal Court of Appeal clarified that the applicable test under the multi-element structure of section 79 consists of several discrete sub-tests, each corresponding to a different requisite element. The most significant statements made by the Federal Court of Appeal apply to the second and third elements of the three-part test under section 79.
With respect to the second element, an "anti-competitive act is identified by reference to its purpose" and the "requisite purpose of the act is an intended predatory, exclusionary or disciplinary negative effect on a competitor". As such, the inquiry under this section must focus upon the intended effects of the act on a competitor; not on the effects of those acts on the state of competition in the market or the general causes thereof. As a result, some types of effects on competition in the market might be irrelevant for the purposes of this sub-test, if these effects do not manifest through a negative effect on a competitor.
The Federal Court of Appeal noted that "proof of the intended nature of the negative effect on a competitor can thus be established directly through evidence of subjective intent, or indirectly by reference to the reasonably foreseeable consequences of the acts themselves and the circumstances surrounding their commission, or both."
It concluded that, even though evidence of subjective intent is neither required nor determinative, intention remains an important ingredient of the second element of the test under section 79. In particular, intention is relevant in the sense that while a respondent cannot disavow responsibility or the reasonably foreseeable consequences of its acts, a respondent might nevertheless be able to establish that such consequences should not, in the context of the section 79(1)(b) inquiry, be considered the intended "purpose" or "overall character" of the acts in question. In appropriate circumstances, proof of a valid business justification for the conduct in question can overcome the deemed intention arising from the actual or foreseeable effects of the conduct, by showing that such anti-competitive effects are not in fact the overriding purpose of the conduct in question. In essence, a valid business justification provides an alternative explanation as to why the impugned act was performed. To be relevant in this context, a business justification must be a credible efficiency or pro-competitive rationale for the conduct in question, attributable to the respondent, which relates to and counterbalances the anti-competitive effects and/or subjective intent of the acts.
The Court clarified that section 79(1)(b) relates to whether the impugned act exhibits the requisite anti-competitive purpose vis-à-vis competitors, while section 79(1)(c) concerns the broader state of competition, and whether the practice has the effect of substantially lessening or preventing competition in the relevant market.
The Court on appeal further clarified that the "but-for" test must be applied by the Tribunal in assessing the impact of a practice of anti-competitive acts on competition in the relevant market. The Federal Court of Appeal judgement clarified that the test mandated by section 79(1)(c) is not whether the relevant markets would or did attain a certain level of competitiveness in the absence of the impugned practice, or whether the level of competitiveness observed in the presence of the impugned practice is "high enough" or otherwise acceptable. These are absolute evaluations, while the statutory language of "effect of preventing or lessening...substantially" clearly demands a relative and comparative assessment. In order to achieve the inquiry dictated by the statutory language of section 79(1)(c), the Tribunal must compare the level of competitiveness in the presence of the impugned practice with that which would exist in the absence of the practice, and then determine whether the preventing or lessening of competition, if any, is "substantial". This comparison must be done with reference to actual effects in the past and present, as well as likely future effects. Only through such a comparative approach can the Tribunal determine, as the statutory provision requires, whether the impugned practice "has had, is having or is likely to have the effect of preventing or lessening competition substantially".
I'd like to touch now on some of the challenges that the Bureau has experienced with respect to abuse of dominant position. Some of these issues have been recently clarified by our Federal Court of Appeal and others remain to be clarified, most notably "joint dominance", the threshold for dominance, essential facilities, and the Regulated Conduct Doctrine or RCD.
Section 79 contemplates the possibility that "one or more persons" may be dominant in a market, however there have not been any contested cases involving joint dominance. The Bureau takes the position in cases of potential joint dominance that a combined market share equal to or exceeding 60 percent will generally prompt further investigation. In order for the Bureau to conclude that there has been a potential joint abuse of dominance there must be evidence to show coordinated behaviour albeit short of "conspiracy" covered by our criminal cartel provisions. The Bureau will consider the following questions: Is there evidence that the alleged coordinated behaviour is intended exclude, discipline or predate a competitor? Is there evidence of barriers to entry into the group or barriers to entrants into the relevant market? Is there evidence that members of the group have acted to inhibit intra-group rivalry?
To date, all cases previously before the Tribunal have included dominant firms with market shares in excess of 80% and significant barriers to entry, therefore the bar remains set very high. Would the Tribunal be willing to accept that a firm with a markedly lower market share possesses market power? I have said that market share is not the only factor taken into account when assessing market power, nevertheless, it remains a very significant one.
The issue of essential facilities is another area which has yet to be addressed in jurisprudence. Section 78 contemplates circumstances under which the withholding of facilities or resources essential to a competitor may be anti-competitive. The issue of essential facilities is especially relevant in network industries, such as telecommunications, that have been or will be deregulated. It remains to be seen under what market conditions, if any, the Tribunal would make an order that required a dominant firm to provide a competitor with, reasonable, access to its resource or facility. Sections 78 and 79, as written and as interpreted by the Tribunal, are broad enough, in my view, to tackle this difficult issue and our section 79 guidelines contemplate this.
Indeed, in a case involving "joint dominance" where a consent agreement was filed with the Tribunal, Interac provided a shared cash dispensing service whereby cards issued by one member of Interac can be used to obtain cash from an ABM owned by another Interac member and an electronic funds transfer service at the point of sale allowing consumers to make purchases at participating retail outlets. Individual financial institutions= proprietary networks, and small or regional shared electronic networks were, by comparison with Interac, inadequate substitutes. Financial institutions, and increasingly non-financial institutions, needed to connect to the Interac network to compete effectively in Canada in markets such as retail banking and credit cards. Part of the resolution set out in the consent agreement required Interac to open its network to potential participants on a non-discriminatory basis.
This brings me to my final point on the challenges of section 79, and it is a fairly significant one--the Regulated Conduct Doctrine or RCD, which is similar in some ways to the US implied immunity and state action doctrines. What happens where the conduct that contravenes the Competition Act is, or more importantly, could be regulated by another federal, provincial or municipal legislative regime? Regardless of whether the RCD or some other doctrine or defence immunizes an impugned conduct from a provision of the Act, the Bureau will always consider the regulatory context in which the conduct is engaged where it is relevant to the application of the provision of the Act in question. Our jurisprudence is minimal on the application of the RCD to reviewable matters, such as the abuse of a dominant position, however, we will not refrain from pursuing regulated conduct under the reviewable matters provisions simply because the provincial law may be interpreted as authorizing the conduct or is more specific than the Act given that the Bureau's mandate is to enforce the law as directed by Parliament not a provincial legislature or its delegate.
As can be seen, while the Federal Court of Appeal provided some much needed clarification on section 79, there still remains a number of frontiers left to be explored. The Bureau will continue to actively seek out cases which test the boundaries of section 79 and may provide valuable jurisprudence. We are also continuing to advocate the inclusion of financial consequences for companies found to have abused their dominant position. The Competition Act with its foundation in modern economics has served us well since 1986, and I am confident that it will remain relevant as Canada continues to adapt to a global marketplace. Thank you.