Canada: What Have We Learnt In 20 Years?

This year is the 20th anniversary of Canada's Competition Act. The act represented a dramatic change in competition law enforcement in Canada. Most significantly, the previously ineffective offences governing monopolies and mergers were decriminalised and replaced with civil, administrative provisions. The commissioner of competition, head of Canada's Competition Bureau, became responsible for bringing merger and abuse of dominance cases before the newly created, quasi-judicial Competition TribunaL. The aim was to make merger law and abuse law especially more relevant. They were to be infused with economically-oriented market analysis. The remedial focus was to move toward correcting market imperfections, rather than punishing ilegal conduct. This article looks at a few of the key trends affecting the act's merger and abuse of dominance provisions at the 20-year mark.

The act overhauled Canada's moribund criminal merger law and ushered in economically- motivated merger review. Increased enforcement of mergers has resulted, with a more rigorous assessment of market effects and recognition of the importance of efficiencies. There are now two noteworthy aspects of merger review in Canada: the process for reaching merger settlements and the role of efficiencies in merger review.

Merger remedies .in Canada, as in many other jurisdictions, mainly come from negotiated settlement, not contested litigation. As such, providing for an efficient and effective consent settlement process is an important objective of Canada's merger control system. The consent order settlement process originally provided for in the 1986 act was replaced in 2002 with a more streamlined system of consent agreement registration. This shift to a 'rubber-stamp' system contrasts with the 2004 Tunney Act amendments in the US, which aim to provide a greater oversight role for the courts. The Canadian experience offers an example of one attempt at balancing certainty, timeliness and fairness considerations in merger settlements.

In 1986, a system was set up to settle mergers by tribunal consent orders. The proposed consent order settlement was filed with the tribunal, along with certain documents, including a 'competitive impact statement' and often an 'agreed statement of facts'. These were intended to explain the nature of the competition problem to define markets, assess barriers and so on and why the proposed consent order was an appropriate remedy. Interested parties could comment and potentially intervene, and the tribunal would decide whether to issue the order. But the process was plagued by three concerns.

First, uncertainty surrounded whether the tribunal itself would approve a settlement. The very first consent order application under the 1986 Act was rejected by the tribunal, citing concerns that the proposed order was highly complex, vaguely defined and would require perpetual monitoring. Although the tribunal issued several other consent orders in subsequent years, the possibility of rejection lingered. In 2000, the tribunal again refused to grant a proposed consent order over concerns about its enforceability.

Second, the scope for comment and intervention by interested third parties created the prospect of delays and uncertainties. Most notably, the Imperial Oiirrexaco consent order application proved to be a lengthy, drawn out affair, involving extensive interventions and hearings. The merger was originally announced in January 1989 and, following the internal review by the bureau and the application before the tribunal, was finally resolved more than a year later in February 1990 when the tribunal approved a revised consent order. Subsequent consent orders and procedural rule changes saw the scope for such third-party scrutiny scaled back; nonetheless, the 'Imperial Oil' case continued to cast doubts about the timeliness of the process.

Third, because the commissioner and the merging parties had concerns about exposing their negotiated settlements to the potential vagaries of the tribunal consent order process, the commissioner would in practice accept undertakings from the merging parties to resolve competition concerns, without invoking the tribunal process. Although the commissioner enjoyed broad discretion to determine whether competition concerns had been addressed, including comfort received through undertakings, the act did not explicitly provide for enforcement of undertakings as it did for tribunal orders, and there continued to be questions about their enforceability.

In June 2002, the merger settlement process in Canada changed significantly, as the act was amended to replace the consent order process with a more streamlined consent agreement registration system. This was supposed to address the concerns identified above. A consent agreement is now filed before the tribunal for 'immediate registration'. The tribunal no longer looks at whether the proposed settlement would probably eliminate the substantial lessening or prevention of competition presumed to arise from the merger. Instead, the consent agreement is simply registered with the tribunal, and then has the same force and effect as a tribunal order. The only scope for scrutiny is that a 'directly affected' third part may, within 60 days, seek to object if it can establish that the terms of the consent agreement could not be 'the subject of an order of the tribunal'.

The merging parties and the commissioner now have a higher level of confidence that their negotiated settlement will remain intact. In light of this streamlined system, the commissioner appears to have largely eschewed the practice of merger undertakings, in favour of registering commitments as consent agreements before the tribunaL. The result has been a dramatic increase in the use of consent agreements, compared with the former consent order process. In less than four years, there have been as many agreements registered with the tribunal as there were consent orders issued in the preceding 16 years of the act's existence.

But new concerns have surfaced that the pendulum has swung too far towards rubber stamping, without sufficient tribunal or interested party oversight. Should there be a subsequent dispute on the meaning of the consent agreement or a request to vary it, the lack of initial oversight has raised the prospect that the tribunal could be called upon to interpret and enforce terms that it might otherwise have viewed as unenforceable. In that regard, there have been several cases recently in which the tribunal has been required to consider merger remedies after the fact, having had no initial say on the merger, as opposed to assessing in advance the merger's likely anti-competitive effects and the appropriateness of particular remedies.

Closely related to the concern about enforceability of terms, the tribunal is faced with no 'evidentiary record', such as a competitive impact statement or agreed statement of facts, which could shed light on the intent of the consent order. Including such documents could no doubt provide useful information in many cases, although it may be unrealistic to expect, particularly given the timing constraints of many mergers, that there wil always be agreement regarding the exact nature of the competitive problem and appropriateness of the remedy - matters over which the merging parties and the bureau often agree to disagree, but are prepared to reach a practical compromise. Additionally, it may be noted that the one contested application regarding the interpretation of a consent order - 'Gemini II' - involved extensive litigation from 1992 to 1994 at the tribunal and Federal Court of Appeal, although the original consent order filing included a detailed competitive impact statement and agreed statement of facts, as well as other supporting documents.

Also, consent agreement registration has narrowed the scope for scrutiny by interested third parties. The tribunal recently interpreted, for the first time, the meaning of 'directly affected' and signalled that thirdparty participation in the consent agreement process wil be confined to parties that are directly affected in terms of competition. Whether there should be broader rights for third parties will continue to be debated. At a minimum, as Canada's parliament has decided against private actions for mergers, it is doubtful that a third party, commenting on a consent agreement process, would be able to pursue a stronger consent agreement and 'challenge' something the commissioner elected not to oppose.

A distinguishing feature of the act, relative to other jurisdictions, is the explicit inclusion of an efficiency defence in merger review. Section 96 of the act provides that the tribunal shall not issue an order against a merger that is likely to produce merger-specific "gains in efficiency that wil be greater than, and will offset" the likely anti-competitive effects of the merger.

The original Merger Enforcement Guidelines of 1991 indicated that the bureau interpreted section 96 to mandate a total welfare trade-off analysis, which essentially involved setting the loss in allocative efficiency (deadweight loss) against the gain in productive efficiency (merger cost savings) to determine whether overall welfare was adversely affected. Under this approach, the 'wealth transfer' from consumers to the merged entity was viewed as neutral and not factored into the calculation.

Despite the seemingly permissive nature of the total welfare approach, in no case has the bureau publicly declined to challenge an otherwise anti-competitive merger on the basis of total welfare considerations.

The issue of how the bureau applied section 96 in practice crystallsed in the 'Superior Propane' litigation from 1998 to 2002, in which the tribunal allowed a merger on the basis of effciency gains, despite fiding that it would likely lessen competition substantially and lead to material price increases. In contrast to the total welfare approach set out in the 1991 guidelines, the bureau advocated that the wealth transfer from consumers to producers should be counted as an anti-competitive effect. The tribunal ultimately opted for a more flexible 'balancing weights' approach, with the importance of wealth transfer to be addressed case by case. On the evidence, the tribunal held that the efficiency gains were sufficient to allow the merger even though significant price increases were expected.

Following 'Superior Propane', concerns that section 96 does not adequately protect consumer interests, and perhaps those of small and mid-sized enterprises, have led to various consultations and amendment proposals regarding that section. No consensus has emerged on whether amendment is required.

The cost saving side of the trade-off equation, virtually overshadowed by the wealthtransfer debate, has remained comparatively straightforward throughout section 96's history. Fixed cost savings are of particular significance. For example, in 'Superior Propane', fixed as well as variable cost savings, were included in the trade-off analysis provided that they were real and merger-specific. Similarly, Canada's 2004 Merger Enforcement Guidelines reiterate that: "(bJoth variable and fixed cost savings are relevant to the analysis because both generate producer surplus (even though it is recognised that generally only variable (ie, marginal) cost savings lead to price reductions)." The same comment is echoed in the subsequent consultation paper on the treatment of efficiencies.

The importance attached to fixed cost savings has major practical implications. Merging parties can include all merger-specific cost savings in the efficiencies analysis without having to determine whether they are fixed or variable - often a difficult task. This contrasts with the siniation in most other jurisdictions, such as the US, where efficiencies must usually be shown to benefit consumers in some sense. In Canada, merging parties can fully claim any merger-specific cost savings, including redundant fixed cost overhead savings, which often represent the most significant savings from mergers among direct competitors.

Although Canadian merger law has traditionally concentrated most on single firm market power, there has been increased attention in recent years on the potential anti-competitive coordinated effects of mergers. Yet, the role of efficiencies in the coordinated effects context remains largely underdeveloped. This no doubt reflects that efficiencies have historically been viewed in Canada through the lens of welfare tradeoff analysis, which has focused attention on merger to monopoly or near-monopoly situations. It remains to be seen, as coordinated effects analysis evolves in Canada, whether the role of efficiencies in merger review will take on new significance, such as consideration of how a merged entity's improved cost structure may disrupt incentives to collude.

Along with merger review, the other main pilar of Canadian competition law to be affected by the 1986 act was abuse of dominant position. Two particular problems faced the pre-1986 law against monopolies. It was criminal, requiring a criminal standard of proof and focusing retroactively on punishing conduct rather than on remedying economic effects. Also, there was a vague 'public detriment' standard. And so, from 1910 until 1986, only one contested monopolisation case resulted in a conviction.

The 1986 amendments replaced the criminal monopolies provision with section 79, an administratively reviewable, civil provision governing abuse of dominant position. The vague public detriment standard was supplanted with an economic approach requiring three elements: a single or joint dominant position; a practice of anti-competitive acts; and a substantial lessening or prevention of competition. In addition, the remedial focus was put on restoring competition going forward. The goal was to create more relevant and economically sophisticated review of dominant firm conduct. That relevance was felt from the outset, with many of the commissioner's early successes before the tribunal regarding abuse of dominance applications.

The shift from the public-detriment test to a purely economic standard of review is evident in several features of the abuse of dominance provision. There must be harmful economic effects. Being dominant, even where there is anti-competitive conduct, may not suffice - there must likely be a substantial lessening of competition. In deciding if a practice is likely to lead to that, consideration must be given to whether the practice is the result of 'superior economic performance'. The remedial focus is purely prospective and economic - altering conduct or structure to correct the market imperfections caused by an abuse of dominance. There are no fines to punish past behaviour. Finally, the abuse provisions are not directed at goals that are not purely economic, such as market integration in the EU.

Section 79 in Canada is substantively narrower than section 2 of the US Sherman Act or article 82 of the EC Treaty. Section 2 in the US provides for attempted monopolisation, whereas section 79 is confined to situations where a dominant position already exists. Article 82 in Europe includes as an 'abuse' conduct that exploits consumers, such as high prices. By contrast, Canada's 'abuses' are confined to conduct that harms the competitive process because of its predatory, exclusionary or disciplinary- effects, not that which exploits the fruits of successful monopolisation on the merits.

Because the market shares involved in dominance cases in Canada have been high - typically over 80 per cent - the exact market share threshold for dominance is largely untested (although the bureau's position is that the threshold may be as low as 35 per cent). Much of the analysis has dealt with whether conduct by admittedly large firms actually produces effects harmful to competition, not just competitors. For example, the most recent tribunal decision on abuse of dominance, 'Canada Pipe', upheld a loyalty incentive programme by a firm it found to be dominant, on the basis that anti-competitive effects were not demonstrated. This overriding emphasis on economic effects under section 79 has more in common with section 2 of the Sherman Act than article 82 in Europe.

Evolving economic thinking on dominant firm conduct will continue to influence how section 79 traces the boundary over the next several years between aggressive, permissible conduct and abusive, anticompetitive practices by large firms in the Canadian marketplace. The comparative experiences in the US, Europe and beyond, as well as the growing attention throughout the global antitrust community, such as at the ICN, will be of increasing interest.

Beyond this inevitable substantive evolution of the law, one of the most likely questions facing section 79 in the coming years concerns whether the scope of enforcement be broadened, in the form of either fines or private access to the tribunal? Perhaps the most obvious distinguishing feature of Canadian abuse of dominance law to date, particularly compared with that of the EU or the US, is that there are no fines, as in Europe, or private action, as in the United States. Instead, remedies under section 79 are confined to forward-looking measures to promote the commissioner's view of the public interest in competition, not the private interest of competitors or other parties.

Nevertheless, the possibility of monetary penalties or private access in respect of section 79 has gained greater attention in recent years. In November 2004, a bill was introduced in parliament to allow the tribunal to impose an 'administrative monetary penalty' of up to C$10 million for a first infringement of section 79 and up to C$15 million for subsequent violations. The bill died with the dissolution of parliament but the debate is likely to resurface. Similarly, limited rights of private access to the tribunal have been granted in recent years for sections of the act that could be considered subsets of the abuse of dominance provision - such as refusal to deal, exclusive dealing and tied selling. The relief available is confined to remedial orders, not financial relief. There have been suggestions that such private access should be extended to abuse of dominance generally and expanded to include damages.

Many of the challenges in 1986 remain today - most notably, how to get the economics substantively right while keeping the process relevant and effective. The one word that seems to pervade all of these issues in Canada is efficiency - from properly weighing efficiencies in merger analysis, to allowing large firms to behave efficiently, to making the processes for merger settlements as efficient and effective as possible. How these various efficiency considerations play out in the coming years wil have a major impact in shaping the act for the next 20 years and beyond.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.