Osler, Hoskin & Harcourt LLP
Box 50, 1 First Canadian Place
Toronto, Ontario, Canada M5X 1B8
416-362-2111 Main
416-862-6666 Facsimile
January 25, 2008
Via E-mail and Courier
Rachel Lavallée
Competition Bureau
Place du Portage 1
50 Victoria Street
Gatineau, Quebec K1A 0C9
Dear Ms Lavallée:
Osler, Hoskin & Harcourt LLP (“Osler”) welcomes the opportunity to comment on the revised Draft Predatory Pricing Enforcement Guidelines (the “Draft Guidelines”) issued by the Competition Bureau (the “Bureau”). We recognize that developing clear and comprehensive enforcement guidelines to distinguish between predatory pricing and aggressive pricing that is “pro-” rather than anti-competitive is challenging. We commend the Bureau for refining and clarifying its positions on such issues as its enforcement approach (criminal or civil), the appropriate measure of costs, and the role of pro-competitive business justifications, such as meeting the competition, in the analysis of below cost pricing.
In general, we are of the view that the Draft Guidelines represent a significant improvement over the draft Enforcement Guidelines for Illegal Trade Practices: Unreasonably Low Pricing Policies Under Paragraphs 50(1)(b) and 50(1)(c) of the Competition Act released in 2002. However, we also believe that there are some areas that would benefit from further refinement and clarification in order to reflect a more sound enforcement position, provide greater certainty to the bar and business community on the Bureau’s enforcement of the predatory pricing provisions of the Competition Act (the “Act”), and ensure consistency in the Bureau’s overall enforcement approach to the provisions of Act.
For the reasons identified in the Van Duzer Report1, the criminal law is not an effective or desirable framework within which to assess and, if necessary, sanction, pricing behaviour that may be predatory. These reasons were implicitly accepted by the drafters of Bill C-19, which proposed the repeal of the criminal predatory pricing provision in paragraph 50(1)(c) of the Act. Osler strongly endorses the repeal of paragraph 50(1)(c) as an appropriate and long overdue step that should be included in the next package of amendments to the Act. In the intervening period and solely as an interim approach, subject to our comments below, the broad approach described in the Draft Guidelines as to when conduct will be pursued under paragraph 50(1)(c) versus section 79 appears balanced and reasonable.
We firmly support reserving criminal enforcement of alleged predatory pricing for cases where conduct is egregious. However, we submit that it is very questionable, at the very least, whether some of the conduct that the Draft Guidelines state may attract criminal enforcement is “egregious”. We agree with the position that low pricing designed to enforce or invite participation in a cartel should be considered to be egregious. We also recognize that the Bureau should retain some discretion to pursue recidivist behaviour criminally.
However, we encourage the Bureau to clarify that past conduct may only weigh in favour of a decision to pursue criminal enforcement if such conduct was found to contravene the criminal predatory pricing provisions of the Act or was found to be predatory and was the subject of an order issued under the abuse of dominance provisions of the Act.
In addition to alleged predation relating to cartel activities and recidivist conduct, the Draft Guidelines in Section 5.2 state:
The Bureau will be more apt to pursue an investigation under the criminal predatory pricing provisions where there is evidence that the competitor was a new entrant or a vigorous competitor in the relevant market. Considerable weight will be given to evidence that the competitor was innovative, a “maverick”, or about to expand capacity to introduce new products, production processes or methods of distribution. The evidence will be all the more compelling where, historically, there have been market power concerns in the relevant market.
In our view, the conduct described in the passage quoted immediately above is qualitatively very different from the other conduct that the Draft Guidelines suggest may attract criminal enforcement. The circumstances described in the passage would potentially be applicable to a broad range of situations. In addition, there is nothing in the quoted passage relating to the mens rea of the alleged predator. The entire discussion relates to the new entrant or the vigorous competitor and whether there may have been “market power concerns in the relevant market”. We respectfully submit that conduct must have a repugnant and egregious moral element before it may attract the criminal enforcement apparatus. Accordingly, we suggest that the above passage of the Draft Guidelines be deleted or substantially narrowed and reworked. As it currently stands, Section 5.2 lacks the necessary clarity to permit legal advisors to properly counsel their clients on compliance with the relevant law.
In contrast to other conduct in respect of which the Act and the Bureau’s guidelines establish procedural safeguards relating to the choice of civil or criminal track2, equivalent procedural safeguards are not provided for enforcement of alleged predatory conduct. Accordingly, we submit that the Bureau should articulate similar safeguards in the final Guidelines. Most importantly, we submit that the Bureau should state that it will elect whether to pursue a case criminally or civilly within a set timeframe, and that it will not pursue a case criminally once a firm under investigation has been advised that the Bureau has elected to pursue a claim under the civil abuse of dominance provisions. Procedural due process is now well established under the Charter with the Supreme Court3 making it abundantly clear that when the predominant purpose of an inquiry is to determine penal liability, regulatory agencies must not utilize civil compulsory means to gather information and evidence. Thus, parties should have the right to be advised, at an early stage, and before they enter into voluntary cooperation or discussions with the Bureau, whether they will face exposure to potential criminal prosecution.
We suggest the Draft Guidelines clarify that (in addition to the existence of market power and the absence of a justifiable business rationale) the likely or actual elimination of a competitor is not a sufficient basis for the Bureau to conclude that below cost pricing is “unreasonable” and therefore in violation of the criminal predatory pricing provisions. Rather, consistent with our previous comments, we suggest that the final Guidelines make it clear that the Bureau will not pursue a matter under the “eliminating a competitor” branch of paragraph 50(1)(c) unless it is satisfied beyond a reasonable doubt that the alleged predatory conduct was accompanied by a degree of anti-competitive intent that is repugnant and egregious.
In the Preface and the Executive Summary, the Draft Guidelines state that the “Bureau considers predatory pricing to be a firm deliberately setting prices to incur losses for a sufficiently long period of time to eliminate a competitor, or otherwise inhibit competition in the expectation that the firm will subsequently be able to recoup its losses by charging prices above competitive levels or achieving another competitive objective”.
There are a number of troublesome aspects to this definition. First, the language “eliminate a competitor or otherwise inhibit competition” conflates the elimination of a competitor with inhibiting competition.4 As the Bureau has recognized on many occasions, inefficient competitors can be eliminated without any adverse impact on competition. Indeed, the same is true of efficient competitors, e.g., where other efficient competitors remain or where barriers to entry are low. We suggest that the words “or otherwise inhibit competition” be replaced with the words “or discipline a competitor”, and that the final Guidelines then describe the type of disciplining behaviour that could attract enforcement action under paragraph 50(1)(c) or section 79.
Second, use of the “competitive level” as the appropriate benchmark is very troublesome because, with the exception of commodity markets, it would be exceptionally difficult, if at all possible, to determine the “competitive level” of pricing in a market. Accordingly, we submit that the appropriate benchmark should be the “level that would have prevailed in the absence of the impugned conduct”. In contrast to mergers, where the well known “cellophane trap” could result in the failure to identify a merger to monopoly, it is difficult to imagine a situation in which the use of pre-predation price levels to define markets, apply the 35% market share safe harbour, apply an appropriate price-cost test and assess the likelihood of recoupment might lead the Bureau to fail to identify and effectively proceed against predatory behaviour.
Third, the words “achieving another anti-competitive objective” are unacceptably vague, as they provide no meaningful guidance or notice as to what conduct might be considered to be predatory. Accordingly, we strongly suggest that this addition to the Bureau’s longstanding policy be reconsidered. While we agree that a firm could theoretically recoup losses incurred over a low pricing period in a manner other than by charging higher prices (e.g. by being able to reduce service levels), this would only be the case if such actions translate into cost or other savings which directly impact the firm’s profitability. We submit that in order to demonstrate that a firm is able to recoup its losses, evidence of an ability to charge higher prices or otherwise concretely improve its profitability following the low pricing period should be required. The mere absence of a pro-competitive business rationale for below cost pricing cannot be sufficient to demonstrate recoupment.
The Draft Guidelines suggest that the Bureau will, before pursuing a complaint, examine whether a complainant is or will become unprofitable over the low pricing period5. We suggest that the Draft Guidelines clarify whether a finding that a competitor of the alleged predator is or would become unprofitable over the low pricing period is a requirement for making out a predatory pricing case (e.g. as a matter of legal analysis), or whether this test is solely a screening tool to assist the Bureau in deciding when it will choose to pursue a complaint.
On a related point, we suggest that the Bureau clarify whether it will pursue predatory pricing claims independently (i.e. in the absence of complaints from competitors) and if so, the circumstances in which the Bureau would be prepared to pursue such cases.
The Draft Guidelines state in Section 3.3 that the Bureau “generally” does not consider a firm with a market share of less than 35 percent to be capable of exercising market power. Footnote 27 then states that “With respect to predation, a firm that has a market share of less than 35 percent may have unilateral market power if it has a unique cost advantage or the ability to use strategic behaviour to build and entrench market power”. This later qualification significantly undermines the value of the 35 percent guideline. To the extent that there does not appear to be anything on the public record to suggest that the Bureau has initiated enforcement action in respect of alleged predatory pricing behaviour where the alleged predator had a market share of less than 35 percent, we suggest that there is no legitimate purpose to be served by including this qualification.
We submit that a firm that has a market share of less than 35 percent cannot exercise market power and that the Bureau should make clear that it will not pursue cases where a low pricing firm holds a share of the relevant market below 35 percent.
The Draft Guidelines state that “firms can recoup losses incurred in one market by exercising market power in another product or geographic market.” We suggest that the Bureau clarify this statement as well as its position on leveraging market power. In this regard, we suggest that the Bureau clarify that a firm cannot be said to have engaged in predatory pricing in any market unless the result is a substantial lessening or prevention of competition in the market in which the low pricing occurred. In addition, we suggest that the Bureau clarify that a predatory pricing case based on a theory of leveraging market power (i.e. where a firm dominant in market A uses its profits from market A to finance below cost pricing in market B where it is not dominant) can only be made out where market power is actually attained (i.e. in market B) as a result of the low pricing activity (consistent with the position of the Tribunal in Tele-Direct).
While we do not propose to provide specific comments on how avoidable costs should be calculated, we note that the Draft Guidelines state that where the appropriate market definition is a bundle of products, the Bureau will determine whether pricing is below cost by assessing whether the price for the bundle, as opposed to individual products within the bundle, are below average avoidable cost. It would be helpful if the Draft Guidelines clarify that this test will be used even where a single product firm alleges it has been harmed as a result of predatory pricing by a firm selling the competitive product as part of bundle of products.
We would be pleased to discuss any of the issues raised in this letter with you in further detail at your convenience. Again we appreciate being provided with the opportunity to comment on the Draft Guidelines.
Yours very truly,
Osler, Hoskin & Harcourt LLP
1A. Van Duzer and G. Pacquet, Anti-Competitive Pricing Practices and the Competition Act: Theory, Law and Practice, October 22, 1999, at 83-84.
2For example sections 52(7), 74.16, 45.1, 79(7) and 98.
3See, in this regard, R. v. Jarvis [2002] 3 S.C.R. 757.
4Likewise, the language at the very end of section 3.5 also conflates harm to a competitor and harm to competition.
5There is inconsistency between the “could be reduced” language used in the first two paragraphs of section 3.5 and language earlier in the document (e.g., at p. ii) which suggests that the test is whether the complainant is actually incurring or likely to incur a loss.