Executive Summary The International Center for Law & Economics (ICLE)[1] submits these comments on the Competition Bureau’s draft Anti-Competitive Conduct and Agreements Enforcement Guidelines, which consolidate enforcement approaches following substantial...
Executive Summary
The International Center for Law & Economics (ICLE)[1] submits these comments on the Competition Bureau’s draft Anti-Competitive Conduct and Agreements Enforcement Guidelines, which consolidate enforcement approaches following substantial amendments enacted through Bill C-56 (2023) and Bill C-59 (2024). These reforms significantly expand the Bureau’s authority by introducing “excessive and unfair selling prices” as an anti-competitive act, extending civil review under section 90.1 to agreements between non-competitors, broadening private access to the Competition Tribunal, and sharply increasing potential penalties.
Although Parliament expanded the statutory toolkit, the Bureau retains considerable discretion in interpretation and enforcement priorities. These comments reflect three overarching concerns. First, the draft Guidelines rely on flexible and open-ended standards that risk reducing predictability for firms seeking to comply. Second, several provisions risk condemning conduct—such as vertical restraints, self-preferencing, and certain pricing practices—that economic evidence shows often enhances efficiency and consumer welfare. Third, expanded private enforcement combined with higher penalties and potential monetary awards makes legal uncertainty far more costly than under the prior regime.
Competition enforcement necessarily operates under uncertainty. An error-cost framework provides a useful guide. False positives—mistaken condemnation of pro-competitive conduct—impose broad and durable harms by chilling pricing, contracting, investment, and innovation across the economy. Once conduct is deemed unlawful, market forces cannot correct the error. False negatives, by contrast, tend to be more temporary: where firms genuinely exercise market power without efficiency justification, elevated profits attract entry and expansion. These asymmetries counsel restraint in close or ambiguous cases.
The 2022–2024 amendments heighten the importance of this framework. Private parties now face lower thresholds to seek Tribunal review, and beginning in 2025 the Tribunal may now order monetary payments to private applicants. Private incentives do not necessarily align with consumer welfare, and Canadian law lacks U.S. doctrines such as “antitrust injury” that screen out strategic litigation. Clear limiting principles in the Guidelines are therefore essential.
Several substantive areas warrant particular care. The excessive-pricing provision presents acute interpretive challenges. High prices often perform pro-competitive functions by allocating scarce goods and signalling opportunities for entry. Determining whether a price is “excessive” requires cost allocation, valuation of intangible capital, and risk adjustment—tasks that authorities cannot perform reliably, especially in multi-product firms and with respect to multi-sided platforms. Economic research shows that cost-based benchmarks are particularly ill-suited in platform and high fixed-cost industries. The Guidelines should therefore state unequivocally that high prices alone do not constitute anti-competitive conduct and should assess pricing only where it forms part of exclusionary conduct that prevents entry or expansion, such as margin squeezes.
The extension of section 90.1 to vertical agreements under a “significant purpose” standard similarly risks overreach. A substantial literature demonstrates that vertical restraints typically enhance efficiency and consumer welfare. Francine Lafontaine and Margaret Slade’s survey of the theoretical and empirical evidence shows that such restraints often solve coordination problems and increase joint manufacturer–consumer welfare. Because the Act does not define “significant purpose,” the Guidelines should avoid interpretations that would sweep in ordinary, efficiency-motivated distribution arrangements. Safe harbours and presumptions of legality are particularly important in light of expanded private enforcement.
The draft Guidelines appropriately recognise that self-preferencing can have benefits and that concerns arise primarily where firms control access to markets. This approach should be strengthened by treating self-preferencing as a form of vertical integration that is generally pro-competitive, with enforcement limited to cases involving durable market power, meaningful barriers to multi-homing or switching, and demonstrable consumer harm.
Algorithmic pricing likewise does not alter the Act’s core requirement that sections 45 and 90.1 require proof of an agreement—a meeting of the minds. The Guidelines correctly distinguish agreements from lawful conscious parallelism. Enforcement should focus on clear evidence of coordination, such as exchanges of competitively sensitive information or agreements on pricing parameters, rather than parallel outcomes produced by independent algorithmic optimisation.
Although firms can exercise monopsony power in labour markets, the empirical evidence supporting broad labour-antitrust enforcement remains mixed. Estimates of labour-supply elasticity vary widely, and concentration measures often poorly proxy for labour-market power due to worker mobility and non-competitive frictions. Targeted labour regulation—such as minimum-wage laws, collective bargaining frameworks, limits on non-compete clauses, and licensing reform—offers more direct and reliable tools to address worker welfare.
To reduce uncertainty and error costs, the Guidelines should articulate clear market-power screens and safe harbours. A market share below 30% should generally preclude a finding of market power absent extraordinary circumstances. For vertical agreements under section 90.1, presumptive legality should apply where shares and foreclosure remain below 30% or where restraints serve documented efficiency objectives. The Guidelines should also state clearly that the Bureau will prioritise cases supported by clear evidence of substantial competitive harm and exercise restraint where effects are ambiguous, contested, or likely offset by efficiencies.
Taken together, these recommendations would make the Guidelines more predictable, administrable, and economically grounded. They would focus enforcement on conduct most likely to harm competition and consumers, while reducing the risk that expanded statutory authority chills efficient and innovative business conduct.
I. Introduction and Overview
The International Center for Law & Economics (ICLE) submits these comments on the Competition Bureau’s draft Anti-Competitive Conduct and Agreements Enforcement Guidelines (ACCA Guidelines).[2] ICLE is a non-profit research centre that applies economic analysis to legal and regulatory questions, with a particular focus on competition policy and market institutions. ICLE scholars have previously submitted comments to the Bureau on the Future of Competition Policy in Canada (Appendix A)[3] and on algorithmic pricing and competition (Appendix B).[4]
The draft Guidelines seek to consolidate enforcement approaches across several provisions of the Competition Act following significant amendments enacted through Bill C-56 (2023) and Bill C-59 (2024). These amendments materially expanded the Bureau’s authority. They added “excessive and unfair selling prices” as an anti-competitive act under section 78,[5] extended section 90.1’s civil agreements provisions to non-competitors,[6] and increased maximum administrative monetary penalties to the greater of $10 million for individuals or $25 million for corporations, or up to three times the benefit derived from the conduct.[7]
These changes present the Bureau with an important interpretive choice. Although Parliament broadened the statutory language, the Bureau retains discretion in setting enforcement priorities and interpretive standards. These comments focus on how the Bureau can exercise that discretion to protect competition without chilling conduct that benefits consumers.
Three concerns motivate these comments. First, the draft Guidelines rely on an overly flexible analytical framework that reduces predictability regarding which business practices may attract enforcement. Second, several provisions—particularly those addressing excessive pricing and vertical agreements—risk condemning conduct that economic research generally associates with enhanced consumer welfare. Third, the combination of expanded private access to the Competition Tribunal and substantially higher penalties significantly increases the costs of legal uncertainty.
These comments proceed as follows. Section II examines the error-cost framework that should guide the Bureau’s exercise of enforcement discretion. Section III identifies specific provisions where the Guidelines would benefit from greater clarity and economic rigour. Section IV offers concrete recommendations for revision.
II. Managing Enforcement Error Through Effects-Based Analysis
Competition enforcement requires judgments under uncertainty, and those judgments carry a risk of error. The costs of those errors are not symmetric. False positives, in particular, can impose broader and more durable harm than false negatives. Drawing on Frank Easterbrook’s error-cost framework, this section explains how mistaken intervention can chill pro-competitive conduct and innovation across the economy. These considerations are especially important in Canada’s current enforcement environment, where recent statutory changes increase the need for clear limiting principles and interpretive restraint in the Bureau’s Guidelines.
A. The Asymmetric Costs of Enforcement Error
Agencies and courts rarely have complete information about costs, demand conditions, or innovation paths. As a result, enforcement decisions risk two types of error. Type I errors (false positives) arise when authorities condemn conduct that benefits consumers or has no anti-competitive effect.[8] Type II errors (false negatives) arise when authorities fail to intervene against genuinely harmful conduct.
Although both error types can impose costs, their consequences are not symmetric. Judge Frank Easterbrook of the 7th U.S. Circuit Court of Appeals articulated this asymmetry in his 1984 article “The Limits of Antitrust.”[9] False positives tend to generate lasting harm because they prevent markets from discovering the benefits of the prohibited conduct. When an authority condemns a pricing strategy or distribution arrangement, firms across the economy observe the signal and avoid similar practices, even when those practices would benefit consumers. Legal prohibitions do not self-correct.
False negatives, by contrast, often prove temporary. When a firm restricts output and raises prices without an efficiency justification, the resulting profits attract entry. As Easterbrook observed, a monopolistic practice wrongly excused “will eventually yield to competition, though, as the monopolist’s higher prices attract rivalry.”[10] False negatives also tend to create little precedent and therefore send weak economy-wide signals. Their effects, while potentially harmful in the short term, usually remain confined to the firms and markets at issue. Market forces can correct under-enforcement in ways that they cannot correct over-enforcement, which can entrench itself through precedent and chill pro-competitive or benign conduct.
Empirical evidence supports this self-correction mechanism. Brian Albrecht and Ryan Decker examine whether rising markups correlate with declining business dynamism at the industry level and find no such relationship.[11] If anything, the correlation is positive: industries with the largest markup increases experienced the smallest declines in entry and other measures of dynamism. These findings align with Schumpeterian dynamics, where high returns attract entry and creative destruction, rather than signal insulation from competition. They also call into question the assumption, reflected in the draft Guidelines’ market-share thresholds, that higher margins necessarily indicate competitive harm requiring intervention.
The error-cost framework, however, admits important qualifications. Self-correction through entry may be slow or incomplete where network effects produce winner-take-all dynamics, switching costs lock in customers, regulatory barriers limit entry, or intellectual-property rights confer durable exclusivity.[12] In such settings, false negatives may persist longer than Easterbrook’s framework would otherwise predict. The appropriate response is not to abandon the framework, but to apply it with care. Where market conditions plausibly impede self-correction, the case for intervention strengthens. The burden, however, should rest with enforcement proponents to demonstrate that such conditions exist in the specific market at issue, rather than assuming their presence. The error-cost framework counsels restraint in close cases precisely because these conditions are difficult to identify ex ante and easy to assert ex post by parties seeking regulatory protection from competition.
Geoffrey Manne and Joshua Wright extended these principles to administrative decision-making in “Innovation and the Limits of Antitrust.”[13] They showed that in markets marked by rapid technological change, the cost asymmetry between error types widens further. A mistaken intervention that blocks an innovative business model or distribution method denies consumers benefits they may never recover. The loss compounds when other firms, observing the enforcement action, abandon similar innovations.[14]
The Bureau’s draft Guidelines recognise that competitive effects should drive enforcement decisions. They note that, when investigating potential issues, the Bureau “often focus[es] first on the conduct or agreement itself and how it can impact competition.”[15] Demonstrated or likely effects under the specific facts and circumstances of a given case provide the most reliable basis for sound competition analysis. The final Guidelines should articulate this principle more clearly and consistently.
The Bureau should therefore commit to prioritising enforcement where there is clear evidence of substantial competitive harm, while exercising restraint where likely effects remain ambiguous or genuinely contested. This approach would promote predictability and disciplined enforcement, reduce the risk of both Type I and Type II errors, and reflect the asymmetric costs those errors impose.
B. How Recent Amendments Affect Enforcement Error
The error-cost framework developed by Easterbrook and extended by Manne and Wright emerged in the context of U.S. antitrust enforcement, where decades of judicial development have created procedural and substantive filters that screen out weak claims before they impose significant costs on defendants.[16] Canada’s 2022–2024 amendments to the Competition Act moved in the opposite direction. Taken together, these changes increase the expected costs of false positives—mistaken or strategic enforcement against pro-competitive conduct—and therefore heighten the need for interpretive restraint in the Bureau’s Guidelines.
The amendments substantially expanded private access to the Competition Tribunal. Previously, applicants seeking leave under sections 75, 77, or 79 had to show that their entire business was “directly and substantially affected.”[17] The 2024 amendments relaxed that requirement. Applicants now need only show that part of their business is affected or that granting leave would be in the “public interest.”[18] In late 2024, the Commissioner intervened in JAMP Pharma Corporation v. Janssen Inc. to address the leave test, submitting that it should be “a fairly low bar that only screens out clearly frivolous and vexatious claims.”[19]
This expansion matters because private enforcement operates under incentives that differ from public enforcement. As William Breit and Kenneth Elzinga observed, private plaintiffs bring competition claims to advance their own interests, which may diverge from consumer welfare.[20] Terminated distributors, struggling competitors, or firms seeking to slow a rival’s expansion may recast ordinary commercial disputes as competition-law violations. Economic literature shows that private competition litigation can be used strategically in ways that hinder, rather than promote, competition.[21]
U.S. antitrust law developed the “antitrust injury” doctrine to address this risk. In Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc. (1977),[22] the U.S. Supreme Court required private plaintiffs to show injury “of the type the antitrust laws were intended to prevent”—harm flowing from reduced competition, not merely harm from a defendant’s conduct.[23] As Jonathan Jacobson and Tracy Greer later observed, this doctrine has helped preserve the pro-consumer focus of antitrust law and limit its strategic misuse.[24]
Canadian competition law contains no comparable filter. A private applicant granted leave under the amended Act need not show that its injury stems from harm to competition, rather than from vigorous competition by a more efficient rival. In this context, the Bureau’s enforcement priorities and interpretive guidance become the primary safeguard against strategic litigation. The Guidelines should acknowledge this role explicitly.
The amendments also increased maximum administrative monetary penalties to $25 million for corporations, or three times the benefit derived from the conduct, with penalties rising to $35 million or three times the benefit for repeat violations.[25] These increases raise the expected cost of every enforcement action, including mistaken ones. Economic theory on optimal penalties, developed by Mitchell Polinsky and Steven Shavell, shows that penalty magnitude must reflect both the probability and accuracy of enforcement.[26] Where enforcement is imperfect, excessively high penalties lead to over-deterrence. Firms respond by avoiding not only harmful conduct, but also efficient and pro-competitive behaviour that carries legal risk. As Herbert Hovenkamp has explained, overly expansive liability rules can condemn socially beneficial practices.[27] High penalties combined with ambiguous standards amplify this effect. When the Guidelines do not clearly identify which practices may attract liability, risk-averse firms will avoid a wider range of conduct, including efficient distribution arrangements, pro-competitive pricing strategies, and beneficial vertical integration.
Beginning in June 2025, the Competition Tribunal gained authority to order respondents to make monetary payments to private applicants and “any other person affected” by the conduct, up to the value of the benefit derived from the impugned practice.[28] This change materially alters private enforcement incentives. Where private applicants once could seek only injunctive relief, they may now pursue substantial monetary awards. The U.S. experience with treble damages illustrates the likely effects. Private antitrust litigation in the United States routinely imposes millions of dollars in legal costs, regardless of outcome.[29] These costs create pressure to settle even weak claims, as defendants may choose to pay to avoid litigation expense.
Additional uncertainty arises from extending section 90.1 to agreements among non-competitors under a “significant purpose” standard. Vertical agreements—such as distribution contracts, exclusive arrangements, and franchise terms—often affect competition as part of their ordinary commercial function. A manufacturer that grants exclusive territories does so precisely to shape how distributors compete. Whether such conduct has a “significant purpose” of preventing or lessening competition depends on interpretive choices the Guidelines must make.
Edward Iacobucci the University of Toronto has cautioned that lowering burdens of proof in competition law will predictably do more harm than good.[30] While the statutory amendments are now in force, the Bureau retains discretion in how it interprets and applies them. The Guidelines are the Bureau’s primary instrument for signalling enforcement priorities and providing compliance guidance. Given the increased costs of false positives under the amended Act—expanded private access, higher penalties, monetary awards, and broader substantive reach—the Bureau should use this discretion to establish clear limiting principles. The asymmetry between error types becomes more consequential when each mistaken enforcement action carries greater expected costs. Standards that might have been tolerable under a regime of public enforcement, modest penalties, and constrained private access become far more problematic when all three change at once.
C. Excessive Pricing and Effects-Based Competition Analysis
Competition law works best when it asks a simple question: does the challenged conduct harm consumers through higher prices, reduced output, lower quality, or diminished innovation? A consumer-welfare focus provides an objective benchmark and helps distinguish genuine competition concerns from complaints by less efficient rivals seeking regulatory protection.
Canadian competition law follows a different analytical path. In Commissioner of Competition v. Canada Pipe Co., the Federal Court of Appeal set out the governing framework for abuse of dominance under section 79 of the Competition Act.[31] The Supreme Court of Canada denied leave to appeal in 2007, leaving the Federal Court of Appeal’s decision as controlling authority. The court applied section 79(1)(c), which requires proof that the impugned conduct “has had, is having or is likely to have the effect of preventing or lessening competition substantially.”[32] It adopted a “but-for” test that compares market conditions with and without the conduct to assess whether the market would be “substantially more competitive” absent the practice.[33] This framework centres on harm to the competitive process and to competitors, rather than directly on consumer outcomes, and reflects Canada’s total-welfare orientation under section 1.1 of the Act.
Recent amendments to the Competition Act introduce concepts that lack clear economic content. section 78(1)(k) now treats the imposition of “excessive and unfair selling prices” as an anti-competitive act.[34] The Act does not otherwise define “excessive” or “unfair,” creating significant interpretive uncertainty. A price that appears high relative to cost may reflect legitimate returns on innovation, risk, or intellectual property. Conversely, a low price that benefits Canadian consumers may appear “unfair” to less efficient competitors, even when it results from superior efficiency.
The European Union’s experience with excessive-pricing enforcement under Article 102 of the Treaty on the Functioning of the European Union illustrates these difficulties.[35] The European Court of Justice established a two-part test in United Brands v. Commission: whether the price-cost margin is excessive and whether the price is unfair in itself or in comparison with competing products.[36] Despite this framework, the European Commission brings excessive-pricing cases only rarely, recognising the measurement problems and high risk of error.[37]
Comparative studies across jurisdictions reach a similar conclusion: authorities should reserve excessive-pricing intervention for exceptional cases. Surveys of EU and national case law, as well as OECD roundtables, show that such cases typically arise in regulated or quasi-regulated sectors, or where dominance flows from legal privileges or prior exclusionary conduct, rather than from pure “standalone” exploitative abuses.[38] Scholars such as David Evans and Jorge Padilla,[39] Massimo Motta and Alexandre de Streel,[40] and Lars-Hendrik Röller[41] argue that excessive-pricing claims are most defensible when tied to exclusionary conduct or durable structural barriers that competition law can identify and address. They caution that attempting to police high prices in competitive or contestable markets risks severe measurement error and effectively turns competition authorities into price regulators, a role for which they lack both the tools and institutional mandate.
The Bureau’s draft Guidelines appropriately recognise that excessive pricing will generally matter only where it links to exclusionary effects or harms the competitive process itself.[42] The final Guidelines should state this limitation clearly and treat it as a binding enforcement principle. High prices alone should not trigger liability under the Competition Act without evidence that those prices result from, or reinforce, conduct that prevents or substantially lessens competition.
D. Market Share as a Screening Tool
The draft Guidelines state that firms with market shares above 30% are “more likely to have market power” for the purposes of sections 75, 76, 77, and 90.1,[43] a reduction from the 35% threshold in prior guidance. For abuse of dominance under section 79, the Guidelines retain the 50% (single firm) and 65% (joint) dominance screens.[44]
Market-share thresholds can serve a useful screening role, but treating a 30% share as a meaningful indicator of market power lacks economic support. A market with many small firms is not necessarily more competitive than one with fewer large firms. Concentration often reflects competitive success rather than market failure.[45] As firms with lower costs and better offerings expand, competition can increase concentration as higher-cost rivals lose sales.[46] As Chad Syverson observes, “many empirical studies in varied settings have found that greater substitutability/competition—resulting from, say, reductions in trade, transport, or search costs—shifts activity away from smaller, higher-cost producers and toward larger, lower-cost producers,” increasing concentration even as margins decline.[47]
Decades of economic research undermine the view that market shares or concentration ratios reliably predict competitive harm. Harold Demsetz’s critique of the “market concentration doctrine” showed that correlations between concentration and profits often reflect efficiency rather than market power.[48] Writing in the Harvard Law Review, William Landes and Richard Posner concluded:
Since market share is only one of three factors… that determine market power, inferences of power from share alone can be misleading. In fact, if market share alone is used to infer power, the market share measure… which is determined without regard to market demand or supply elasticity… will be the wrong measure.[49]
More recently, Dennis Carlton has described market shares and concentration as “at best crude first steps” to analysis, noting that empirically there is “only a weak link between change in market share and change in competitive performance,” and that “there is no model” in which market share alone reliably predicts prices or welfare effects.[50] Static market-share data therefore cannot substitute for analysis of firm conduct and market conditions.
Modern competition analysis instead emphasises entry conditions and potential competition. The contestable markets framework developed by William Baumol, John Panzar, and Robert Willig shows that industry outcomes depend on the costs of entry and exit—particularly sunk costs—rather than the number or size of incumbent firms. Even highly concentrated markets can remain competitively disciplined where entry is feasible, because “potential competition enforces cost-minimization regardless of the nature of the industry structure.”[51] David Teece extends this insight to innovation-driven sectors, arguing that market share is a poor proxy for market power where technological change and “unseen” competition constrain incumbents.[52] OECD methodological reviews likewise find that contestability, entry barriers, and switching costs shape competitive outcomes but do not appear in standard concentration measures.[53]
Evidence from productivity[54] and innovation[55] studies reinforces this point. Measures of competitive pressure tied to entry and policy changes correlate more strongly with performance than does market structure alone. Competition authorities should therefore treat market shares and concentration as incomplete and potentially misleading first screens, to be supplemented—and where appropriate displaced—by analysis of entry, expansion, switching costs, innovation, and the durability of any competitive advantage.
Digital markets illustrate these dynamics. Firms such as Facebook, Google, and Amazon achieved high market shares by offering valuable services at low or zero prices and by improving consumer experience. Their positions remain contestable through innovation, as shown by TikTok’s rapid growth in social media and by continued competition in e-commerce from Shopify, Walmart, and specialised retailers.
The final Guidelines should make clear that market-share thresholds serve only as initial screens, not presumptions of market power. Below 30%, the Bureau should state that it will not pursue enforcement absent exceptional circumstances. Above 30%, the Guidelines should identify specific factors—such as entry conditions, rival expansion, customer switching, and innovation rates—that can rebut any inference of market power based on market-share data alone.
III. Limiting Principles for New and Expanded Theories of Liability
Recent amendments to the Competition Act expand potential liability and raise new interpretive and administrability challenges for both the Bureau and firms seeking to comply. This section addresses several of the most consequential changes, including the introduction of “excessive and unfair pricing” as an anti-competitive act, the extension of section 90.1 to vertical agreements under a “significant purpose” standard, and the treatment of self-preferencing and algorithmic pricing. In each area, the central issue is how the Bureau can provide guidance that captures genuinely exclusionary conduct without condemning common commercial practices that typically promote efficiency and consumer welfare.
These concepts depend heavily on context, measurement, and intent. Overly broad interpretations risk turning competition enforcement into de facto regulation, particularly in dynamic, innovation-driven markets where conventional screens—such as price–cost comparisons or presumed anti-competitive purpose—often mislead. This section identifies where the draft Guidelines could more clearly articulate limiting principles and evidentiary thresholds that support effective enforcement while reducing the risk of chilling beneficial pricing, contracting, and product-design decisions.
A. Interpreting ‘Excessive and Unfair Pricing’ Under Section 78
Section 78(1)(k) presents the most significant interpretive challenge in the amended Competition Act. It adds “directly or indirectly imposing excessive and unfair selling prices” to the list of anti-competitive acts. To establish abuse of dominance under section 79, the Bureau must still show that a dominant firm engaged in a practice of anti-competitive acts that has the effect of substantially preventing or lessening competition, or that the firm intended the conduct to have a predatory, exclusionary, or disciplinary effect on a competitor.[56]
1. Price levels are not abuse
Charging a “high” price may reflect existing market power, but it does not create or preserve that power. High prices serve two essential market functions. First, they allocate scarce goods to buyers who value them most. Second, they signal profit opportunities that attract new supply and entry. When an authority constrains prices below market-clearing levels, both functions break down. Artificially low prices create shortages and weaken, or eliminate, the entry incentives that would otherwise erode monopoly positions over time.
The U.S. Supreme Court recognised this logic in Verizon Communications Inc. v. Law Offices of Curtis V. Trinko, LLP.[57] Writing for the Court, Justice Antonin Scalia explained that “the mere possession of monopoly power, and the concomitant charging of monopoly prices, is not only not unlawful; it is an important element of the free-market system. The opportunity to charge monopoly prices—at least for a short period—is what attracts business acumen in the first place.”[58]
Canadian competition law takes a different approach by defining certain conduct as anti-competitive acts even when dominant firms charge competitive prices. The same economic logic nevertheless applies. If high prices alone establish abuse, competition law risks turning the Competition Tribunal into a price regulator. Tribunals lack the information, tools, and institutional mandate to set prices across diverse industries.
The Bureau should therefore focus enforcement on the conduct that enables a firm to sustain supra-competitive prices—such as exclusionary practices that raise rivals’ costs, foreclose distribution, or block entry—rather than on price levels themselves. This approach targets genuine sources of competitive harm while avoiding the institutional risks of de facto price regulation.
2. Limits of cost-based price tests
Determining whether a price is “excessive” requires a defensible competitive benchmark. The most common approach compares price to cost, often through a price–cost margin. In practice, this approach faces serious and often insurmountable obstacles.
Cost Allocation: Multi-product firms must allocate common costs—such as headquarters operations, shared research and development, and brand investment—across products. Different allocation methods produce very different margins for the same firm. A pharmaceutical company, for example, may appear to earn a 95% margin on a successful drug if analysis counts only manufacturing costs, but a much lower margin once it allocates the costs of numerous failed R&D projects that preceded that success.
Intangible Capital: Many modern firms create value through intangible assets such as software, data, brand reputation, and network effects. Accounting rules often treat these investments as current expenses rather than capital, which depresses measured costs and inflates measured margins. A software firm with a very high price–cost margin may simply reflect near-zero reproduction costs after substantial upfront investment and risk, rather than market power.
Risk Adjustment: Competitive returns must compensate investors for ex ante risk, not just observed ex post costs. Distinguishing returns that reflect market power from returns that reflect uncertainty, innovation risk, and sunk investment requires information that enforcement authorities typically do not have.
A substantial body of economic literature shows that cost-based tests for excessive or predatory pricing become unworkable in multi-sided platforms and in industries with large fixed and common costs and very low marginal costs.
For multi-sided platforms, David Evans and Richard Schmalensee’s survey of platform economics and antitrust concludes that “standard cost-based tests for detecting predatory pricing generally make no economic sense for a multi-sided business.”[59] Building on the work of Jean-Charles Rochet and Jean Tirole, as well as Mark Armstrong,[60] they show that optimal platform pricing frequently involves prices below marginal cost on one side and above cost on the other in order to balance participation across interdependent user groups. In equilibrium, both profit-maximising and welfare-maximising prices may fall below marginal cost on one side, meaning that one-sided price–cost comparisons will systematically misclassify efficient platform conduct as predatory or excessive.
Evans has made this point even more directly in submissions to the U.S. Department of Justice (DOJ) and the OECD, arguing that price-equals-cost benchmarks are not economically meaningful for two-sided platforms because prices on each side depend on demand elasticities, indirect network effects, and costs across the platform as a whole.[61] He further notes that development and manufacturing costs for core platform assets—such as video-game consoles—cannot be meaningfully allocated across user groups, rendering side-specific cost benchmarks arbitrary.[62] This literature supports the conclusion that traditional Areeda–Turner-type cost tests are not merely difficult to apply in platform markets, but conceptually mis-specified.
Similar concerns arise in high fixed-cost, low marginal-cost, and R&D-intensive industries. Herbert Hovenkamp’s reassessment of the Areeda–Turner test explains that average-variable-cost benchmarks “seriously underdeter in markets with high fixed costs.”[63] Attempts to repair this by redefining “variable” cost to include depreciation or joint costs introduce substantial measurement problems, particularly where capacity is lumpy or shared across products. When addressing digital and intellectual-property-based goods with near-zero marginal cost, such as software, music, or e-books, Hovenkamp questions how any average-variable-cost rule could operate coherently and suggests that cost-based tests may need to be abandoned altogether in such settings.[64]
Einer Elhauge reaches a parallel conclusion for industries with large common costs and low marginal costs. He argues that allocating joint costs to individual products or customers is inherently arbitrary, causing cost-based tests to misclassify efficient price discrimination and competitive above-cost price responses as predatory or excessive.[65] William Baumol and David Bradford’s classic analysis of optimal departures from marginal-cost pricing similarly shows that in decreasing-cost industries—such as utilities, telecommunications, and transport—marginal-cost pricing is infeasible and that multiple second-best pricing structures can support efficient outcomes.[66] There is therefore no single, administrable cost-based benchmark that regulators can reliably enforce.
The literature on excessive pricing in pharmaceuticals reinforces these concerns. An OECD Competition Committee roundtable notes that marginal-cost benchmarks are untenable in the presence of large sunk R&D costs, while total-cost benchmarks are also misleading because competitive firms routinely price above total cost.[67] For multiproduct originator firms, allocating joint manufacturing costs, multi-year R&D expenditures, and returns to intellectual property to a single molecule is exceptionally difficult. David Evans and Jorge Padilla describe the task of identifying a reliable cost-based excessive-price benchmark in such sectors as “a daunting, if not impossible task.”[68] Similar conclusions appear in EU law & economics scholarship, including work by Massimo Motta and Alexandre de Streel, who caution that cost-based excessive-pricing control in innovative, high fixed-cost sectors risks arbitrary allocation or de facto price regulation.[69]
Taken together, this body of work supports a clear conclusion. In multi-sided platforms, R&D-intensive industries, and other high fixed-cost, low marginal-cost settings, conventional cost-based tests for excessive or predatory pricing are not merely difficult to apply; they are fundamentally ill-suited. Reliance on such tests risks arbitrary outcomes and de facto price regulation. Where enforcement proceeds at all, it should rely on standards that focus on exclusionary conduct and competitive effects, rather than fragile and misleading price–cost benchmarks.
3. Fairness is not an enforceable standard
The Guidelines’ inclusion of “excessive and unfair pricing” as an anti-competitive act under section 78 raises fundamental concerns about the use of “fairness” as an enforceable legal standard.[70] While the Guidelines appropriately recognise that high prices alone do not amount to abuse of dominance,[71] the Bureau should interpret “unfairness” narrowly to avoid importing subjective and unworkable value judgments into competition enforcement.
The concept of “fairness” is not new to competition law. Its appeal lies in its association with equity and justice—values that are difficult to contest. The difficulty is that fairness lacks a stable and measurable meaning. As a result, it functions more as a rhetorical device than as an operational standard for identifying anti-competitive conduct. As Giuseppe Colangelo has observed, fairness is “a mere aspiration and a useful mantra for political signaling.”[72] Its ambiguity makes it attractive to policymakers because it “reliev[es] them of the burden of economic analysis” and expands the scope for discretionary intervention.[73]
Under a traditional understanding of fairness as equality before the law, there is little basis for redistributing rents away from dominant firms that achieved their position through superior business models, management, or product design. By contrast, if fairness is understood as equality of outcomes, it can be invoked to justify redistributing value across the supply chain or to less efficient competitors.[74] The statute provides no guidance on which conception should apply.
This ambiguity becomes more acute in markets with multiple stakeholder groups. What appears “fair” to one group may appear unfair to another. Terms that app developers view as fair compensation for access to a platform may not appear fair to the platform that invested substantial resources to create it, or to consumers who may face higher prices to subsidise services they do not use. The concept offers no principled way to resolve these competing claims.
An expansive fairness standard would also sit uneasily with the Competition Act’s own objectives. The Act has long focused on protecting the competitive process, not on prescribing market outcomes. As the Bureau noted in its 2022 “Future of Competition Policy in Canada” discussion paper, the Competition Act “does not proactively dictate how to conduct business, allocate resources among stakeholders, or designate participants, winners or losers in the free market.”[75]
The Guidelines should therefore state clearly that “unfair” pricing is actionable only where it forms part of recognised exclusionary conduct that harms the competitive process. The Bureau should not treat unfairness as an independent basis for intervention untethered from demonstrable competitive harm.
4. Excessive pricing requires exclusion
Section 79 requires that a practice of anti-competitive acts either substantially prevent or lessen competition or be intended to have a predatory, exclusionary, or disciplinary effect on a competitor.[76] The draft Guidelines indicate that the Bureau intends to focus excessive-pricing cases on situations where pricing facilitates or accompanies exclusionary conduct, such as margin squeezes, tied selling, or refusals to deal that entrench monopoly power by excluding rivals.[77]
This limiting principle warrants clearer and stronger expression. The final Guidelines should state unequivocally that high prices, on their own, do not constitute an anti-competitive act under section 78(1)(k). The Bureau should assess pricing only where it evidences, or contributes to, conduct that prevents rivals from offering meaningful alternatives to consumers.
The Guidelines should illustrate this approach with concrete examples. A margin squeeze may involve “excessive” upstream pricing where a vertically integrated dominant firm charges rivals high input prices while pricing the downstream product at levels that foreclose effective competition. Similarly, a tying arrangement may involve unfair pricing where a competitively priced monopoly product is bundled with a separately priced component at supra-competitive levels to extend monopoly power. In each case, the competitive concern arises from exclusionary effects, not from high prices in isolation.
The Bureau should also commit explicitly not to pursue standalone excessive-pricing cases absent clear evidence of exclusionary effects. The Guidelines should clarify that returns attributable to innovation, intellectual property, superior efficiency, or entrepreneurial risk-taking do not amount to excessive or unfair pricing. Finally, the Bureau should state that it will not rely on cost-plus methods or simple price–cost margin tests as primary evidence of excessiveness, given the well-recognised measurement and allocation problems associated with such approaches.
B. Applying Section 90.1 to Vertical Agreements
Bill C-59 expanded section 90.1 to cover agreements between non-competitors where “a significant purpose” of the agreement is to prevent or lessen competition.[78] Previously, section 90.1 applied only to agreements among competitors. This change brings vertical agreements—such as those between manufacturers and distributors, or firms operating at different levels of the supply chain—within the scope of potential civil review.
1. Pro-competitive effects of vertical restraints
A substantial body of economic research shows that vertical restraints usually enhance efficiency and consumer welfare. Francine Lafontaine of the University of Michigan, a former director of the U.S. Federal Trade Commission’s Bureau of Economics, and Margaret Slade, emeritus professor of economics at the University of British Columbia, surveyed the theoretical and empirical literature in 2008.[79] They examined exclusive territories, exclusive dealing, resale price maintenance, quantity forcing, and tying arrangements. Their core findings were as follows:
- Vertical restraints address coordination problems between manufacturers and distributors that would otherwise reduce output and harm consumers;
- Empirical evidence “generally supports the manufacturer’s claim that [vertical restraints] are used in ways that increase joint manufacturer–consumer welfare”;
- Government intervention that prohibits or mandates vertical-restraint terms often leads to higher prices or lower service quality for consumers; and
- Sound policy favours a rule-of-reason approach, supported by safe harbours based on low market shares.
These conclusions reflect a basic economic insight. Unlike horizontal agreements among competitors, which can restrict output and raise prices without offsetting benefits, vertical agreements between complementary firms align incentives to increase output. Manufacturers and distributors earn more by selling more units.
Vertical restraints commonly eliminate inefficiencies in distribution, including:
- Solving Double Marginalisation: When both manufacturers and retailers exercise market power, each may apply its own markup, resulting in prices higher than even a single monopolist would charge. Vertical integration or contractual restraints can eliminate this problem and lower prices for consumers.
- Preventing Free-Riding: Retailers may underinvest in pre-sale services, demonstrations, or inventory if discount rivals can free-ride on those efforts. Exclusive territories or minimum resale prices can address this problem and increase overall consumer value.
- Ensuring Quality: Manufacturers may use vertical restraints to protect brand reputation and maintain consistent product quality across the distribution chain.
Benjamin Klein and Kevin Murphy analyse these rationales in detail.[80] They show that vertical restraints often arise as firms seek to maximise the value of their products to final consumers. By constraining distributor behaviour, manufacturers can better align distributor incentives with consumer interests.
2. Interpreting ‘significant purpose’
The Competition Act does not define “significant purpose.” This matters because the threshold for triggering section 90.1 review will shape how often the Bureau scrutinises vertical agreements. If “significant purpose” captures any non-trivial intent to affect competition, then most exclusive dealing, territorial restrictions, and selective distribution systems would qualify. Firms adopt these arrangements precisely to shape competitive conditions—typically to make their distribution networks more effective in competing against rival networks. The relevant question is whether the arrangement, on balance, harms or benefits competition and consumers.
A narrower interpretation would better target problematic conduct. If “significant purpose” requires that preventing or lessening competition be a primary or dominant motivation, the standard would focus on arrangements that lack a legitimate business justification. This approach aligns with the U.S. rule-of-reason treatment of most vertical restraints following Leegin Creative Leather Products, Inc. v. PSKS, Inc.[81] Under the rule of reason, courts weigh “all of the circumstances” of a restraint, including its “history, nature, and effect” and “specific information about the relevant business,” to determine whether it produces “procompetitive or anticompetitive effects, depending upon the circumstances in which [it was] formed.” Higher prices alone do not end the inquiry; courts also consider whether the restraint supports interbrand competition and legitimate business objectives.
The draft Guidelines suggest that, for agreements between non-competitors, the Bureau may presume a significant anti-competitive purpose where the agreement has the effect of substantially harming competition, absent credible evidence to the contrary.[82] This approach risks reversing the appropriate analytical sequence. The Bureau should begin by assessing whether the agreement has a plausible efficiency rationale. Where it does, evidence of competitive effects should not, on its own, create a presumption of improper purpose.
3. Private enforcement raises the stakes
Section 90.1 now allows private parties to seek leave to bring applications before the Competition Tribunal.[83] This change significantly raises the stakes of ambiguous standards. A distributor terminated for poor performance can now challenge its former supplier, alleging that the termination or the underlying distribution agreement had a “significant purpose” of preventing competition. Even weak claims can impose substantial discovery costs, legal fees, and reputational harm.
Experience in the United States shows that private antitrust litigation can impose significant economic costs. U.S. courts have therefore developed doctrines to screen out meritless claims. The “antitrust injury” requirement established in Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc. requires plaintiffs to show harm to competition, not merely harm to a competitor.[84] The “indirect purchaser” rule in Illinois Brick Co. v. Illinois limits standing to direct purchasers.[85] Canadian competition law has adopted neither doctrine, increasing the risk of expansive private enforcement.
The Guidelines should therefore set out a clear analytical framework for section 90.1 cases involving vertical agreements. At a minimum, the Guidelines should state that:
- The Bureau will not challenge vertical restraints absent evidence that the parties possess substantial market power—defined more precisely than simply exceeding a 30% market share—and that the restraint forecloses a meaningful portion of the market to rivals;
- The Bureau recognises common pro-competitive justifications, such as addressing free riding, eliminating double marginalisation, and ensuring quality, and will not pursue enforcement where an arrangement has a plausible efficiency rationale absent strong evidence of net competitive harm; and
- Any presumption regarding anti-competitive purpose should be removed or confined to circumstances where no legitimate business justification exists.
C. Self-Preferencing and Competition Analysis
The draft Guidelines identify “self-preferencing” as a potential anti-competitive act.[86] The term refers to conduct by vertically integrated firms that favour their own downstream products or services over those of third parties. Common examples include Amazon promoting Amazon Basics products, Google displaying Google Shopping results prominently, and Apple pre-installing Apple applications on iPhones.
1. Self-preferencing is not presumptively harmful
The draft Guidelines adopt a suitably nuanced approach to self-preferencing. They recognise that “[c]hoices that can be seen as self-preferencing are extremely common, and they can have benefits,” and that concern arises primarily “when it is done by a firm that controls access to a market.”[87] The Guidelines identify relevant factors, including whether users multi-home, whether barriers limit effective competition, and whether rivals have feasible alternatives.[88] This framework appropriately ties enforcement to market power rather than treating self-preferencing as inherently harmful.
This approach reflects sound economic principles. Self-preferencing is a natural feature of vertical integration. Firms operating at multiple stages of production or distribution routinely coordinate those activities to maximise joint value. Economic analysis of vertical integration, dating back to Ronald Coase’s 1937 article “The Nature of the Firm,”[89] recognises that firms integrate to reduce transaction costs, internalise complementarities, and adapt more quickly to changing conditions.
In earlier submissions on Canadian competition policy, ICLE scholars emphasised that self-preferencing is not presumptively harmful and identified several pro-competitive justifications, including efficiency gains, security and privacy considerations, and maintaining incentives for platform investment and innovation.[90]
Recent empirical evidence further undermines any presumption of consumer harm. A 2025 field experiment by Chiara Farronato, Andrey Fradkin, and Alexander MacKay used a browser extension to hide Amazon’s private-label products from shoppers and measured the resulting welfare effects.[91] The study found that removing Amazon’s own brands would reduce consumer surplus by 5.4%, with roughly 10% of that loss attributable to higher prices charged by other sellers. The authors also tested whether demoting private labels in search results—a common proposed remedy for self-preferencing—would benefit consumers. It did not. As the study concludes, “demoting private labels in search results to counteract potential self-preferencing does not lead to gains in consumer surplus,” because many consumers prefer private-label products and benefit from their prominent placement.[92]
These results align with the economic logic of vertical integration. As Brian Albrecht and Geoffrey Manne explain, self-preferencing often reflects technical and efficiency considerations rather than exclusionary intent.[93] Latency falls when software is co-located with data and coordinated platform design can improve performance and reliability. Prohibiting a cloud provider from bundling its own database with compute services in the name of neutrality would likely raise prices, slow performance, and reduce market scale. The Australian Competition and Consumer Commission has reached a similar conclusion, finding that self-preferencing is “often benign.”[94]
A 2022 review by Analysis Group economists Kevin Caves, David Evans, and Richard Schmalensee reached the same general conclusion.[95] Surveying empirical studies of self-preferencing across digital platforms, the authors found mixed and context-specific effects. In some cases, self-preferencing reduced prices or improved quality; in others, competitive effects remained ambiguous. They concluded that enforcement should proceed on a case-by-case basis, grounded in specific conduct and market conditions, rather than through categorical prohibitions.
2. Multi-homing limits competitive harm
The draft Guidelines appropriately recognise that users’ ability to multi-home—use multiple platforms at the same time—shapes the competitive significance of self-preferencing.[96] Where users can easily compare offerings across platforms or switch to alternatives, self-preferencing poses little competitive risk. Most smartphone users, for example, can install competing browsers, maps, or music services alongside pre-installed applications. Online shoppers typically search across several e-commerce sites, and business sellers often list products on Amazon, eBay, Walmart Marketplace, and their own websites simultaneously. These practices limit the extent to which any single platform’s self-preferencing can foreclose competition.
The Guidelines should apply this logic consistently. Where barriers to multi-homing are low, switching costs are minimal, and effective alternatives exist, self-preferencing will rarely harm competition, regardless of a platform’s market share. Where structural features genuinely limit multi-homing or lock users into a single platform, closer scrutiny may be warranted. Even in those cases, the Bureau should identify concrete exclusionary effects rather than treating integration itself as problematic.
The final Guidelines should state clearly that self-preferencing is a form of vertical integration with generally pro-competitive effects. The Bureau should commit to challenging self-preferencing only where all three conditions are present: (1) the platform holds durable market power protected by high barriers to entry or expansion; (2) users face meaningful obstacles to multi-homing or switching; and (3) the conduct demonstrably forecloses competition in ways that harm consumers. The Guidelines should also recognise that product integration, quality assurance, and investment incentives commonly justify self-preferencing behaviour.
D. Network Effects and Competitive Analysis
The Guidelines also identify network effects as a potential barrier to entry and note that they can advantage incumbent firms.[97] Network effects are a real and well-understood economic phenomenon, but the Guidelines should not overstate their competitive significance or treat their presence as grounds for lowering intervention thresholds in digital markets.
1. Network effects are often local
Network effects are often more limited and local than is commonly assumed. Users typically value participation by their own friends, family, or colleagues, rather than the platform’s total number of users. A social network’s value to someone in a particular city, for example, depends on whether her personal contacts use the service, not on aggregate usage elsewhere. Because network effects operate at this local level, large platforms do not necessarily enjoy insurmountable advantages over smaller rivals that serve specific communities, regions, or use cases.[98]
2. Network effects do not lock in dominance
The same forces that can reinforce incumbent positions can also enable rapid displacement. Network effects may sometimes strengthen existing firms, but they can also intensify competition for the market and generate substantial welfare gains when fragmentation declines. Strong installed bases can further support disruptive innovation by giving new products immediate access to large user groups. A game such as Fortnite, for example, benefits from the installed bases of Xbox, PlayStation, iOS, and Android.[99]
Network effects also rarely confer durable or permanent dominance. TikTok’s rapid growth despite Facebook’s earlier position in social media shows that network effects do not prevent disruptive entry. Google displaced Yahoo! and AltaVista in search despite their early lead and data advantages. In messaging, iMessage, WhatsApp, Telegram, and Signal each hold strong positions in different regions and demographics, even though messaging services exhibit pronounced network effects. Similarly, vertical search platforms such as Amazon and Booking.com now capture a significant share of commercial queries that once flowed primarily to general search engines.
3. Data advantages are often overstated
The role of data in generating network effects is often overstated. Network effects and economies of scale or scope are distinct economic concepts. Economies of scale and scope reflect cost-side efficiencies, whereas network effects arise from increases in user value as participation grows. Network effects therefore depend on how users perceive value, not simply on the volume of data a firm holds.[100]
A common assumption holds that firms must accumulate large datasets and specialised expertise to compete in data-intensive markets. In practice, the advantages of data accumulation often diminish quickly. “Learning by doing” effects in data use tend to plateau, as do gains from scale and scope in data assets.[101] Claims that firms such as Amazon, Google, or Facebook succeed primarily because of superior access to data often reverse the causal relationship.[102] These firms appear to have accumulated large volumes of data because they developed successful, industry-defining business models, not because data alone conferred success. Facebook’s early growth illustrates this point: it built a highly successful social network despite having far less data than established competitors such as MySpace.[103]
4. Multi-homing limits network power
As noted above in the discussion of self-preferencing, widespread multi-homing—the ability of users to engage with multiple competing platforms at the same time—significantly constrains the market power that network effects might otherwise create. Restaurants commonly list on several food-delivery platforms, riders use both Uber and Lyft, and consumers routinely shop on Amazon and competing websites. Where multi-homing is easy and common, it weakens lock-in, lowers switching costs, and prevents network effects from translating into durable market dominance.[104]
The Bureau should therefore resist calls to apply heightened scrutiny or relaxed legal standards to digital markets based solely on theoretical concerns about network effects. Enforcement should remain grounded in demonstrated competitive effects, not conjecture about market dynamics.
E. Algorithmic Pricing and the Law of Coordination
The draft Guidelines note that information sharing among competitors can facilitate coordination and that algorithmic-pricing tools may increase the risk of coordinated outcomes.[105] In 2024, the Bureau issued a separate discussion paper on algorithmic pricing and invited public comment.[106] As explained in our submissions in Appendix B, algorithmic pricing represents an evolution of longstanding business practices—such as yield management in airlines and hotels—rather than a fundamental shift in competitive dynamics.
1. Public price monitoring is lawful
Economic theory and competition law distinguish between sharing private, competitively sensitive information and observing public market data. Firms have long monitored competitors’ public prices to inform their own pricing decisions. This practice remains lawful whether carried out by human analysts or through automated tools. As former U.S. Federal Trade Commission Chair Maureen Ohlhausen has explained, the “guy named Bob” test applies: if a pricing manager (“Bob”) can legally review competitors’ public prices and adjust his firm’s rates, an algorithm performing the same function remains lawful.[107]
The Bureau’s discussion paper raises concerns about competitors “pooling data” through algorithmic tools. That concern requires careful definition. Reputable third-party vendors design pricing systems to prevent the sharing of non-public, competitively sensitive information across firms. A hotel using revenue-management software, for example, receives market analysis based on public information, such as rates scraped from travel websites, but does not receive rivals’ proprietary booking or pricing data. This one-way use of aggregated public data differs fundamentally from the exchange of confidential strategic information that characterises cartel conduct.
2. Distinguishing agreement from parallel conduct
The Competition Act requires an “agreement” to establish liability under both section 45 (criminal cartel provisions) and section 90.1 (civil competitor agreements). The draft Guidelines correctly interpret this requirement as demanding a “meeting of the minds.”[108] That interpretation reflects established common-law principles and appropriately limits enforcement discretion.
The Guidelines also draw a clear and important distinction between agreements and parallel conduct. As the Bureau explains, “[c]onscious parallelism is a specific type of coordination. It involves each firm setting their prices with an expectation of how their competitors will respond…. Conscious parallelism occurs without an agreement between the firms. There is no ‘meeting of the minds’ between the firms.”[109] The Guidelines further confirm that “conscious parallelism alone between competitors does not create an agreement that we can act on under section 90.1.”[110]
This distinction reflects core principles of Canadian competition law. In R. v. Nova Scotia Pharmaceutical Society, the Supreme Court of Canada held that conspiracy requires proof of a “common design” involving both intention and agreement.[111] Parallel conduct—even where firms anticipate and respond to competitors’ actions—does not meet that standard. The Guidelines appropriately preserve this boundary between lawful interdependent behaviour and actionable coordination.
3. Three forms of coordination
The terminology in this area can cause confusion. Economists use the term “tacit collusion” to describe supracompetitive pricing that arises when oligopolists recognise their mutual interdependence without reaching any explicit agreement.[112] Each firm independently concludes that aggressive price competition is unprofitable because rivals will match price cuts, eliminating any gain. This economic concept, however, does not align neatly with legal standards, and it is important not to conflate distinct forms of conduct.
The Guidelines appropriately distinguish among three categories:[113]
- Explicit Agreements (Actionable): Formal contracts or informal understandings based on direct communication between competitors.
- Tacit Agreements (Actionable): Situations in which firms achieve a “meeting of the minds” through indirect communication or signalling. The Guidelines provide an example in which a firm publicly announces that it will continue a practice only if rivals do the same and will stop otherwise. If rivals respond by adopting the same conduct, this signalling may establish a tacit agreement.[114]
- Conscious Parallelism (Not Actionable): Circumstances in which firms independently set prices while anticipating how competitors will respond. As the Guidelines explain, conscious parallelism “occurs without an agreement between the firms” and involves no “meeting of the minds.”[115]
The key distinction between a tacit agreement and conscious parallelism is communication. Tacit agreements involve signalling that invites coordination and leads to a shared understanding. That communication—although indirect—creates the consensus that section 90.1 requires. Conscious parallelism, by contrast, involves no signalling or exchange of commitments. Firms act independently based on their own assessment of market conditions and rivals’ likely responses.
This distinction reflects a core principle of competition law. As the U.S. Supreme Court explained in Brooke Group, tacit coordination based solely on mutual interdependence “describes the process, not combatted by the antitrust laws, by which firms in a concentrated market might in effect share monopoly power.”[116] Treating such behaviour as unlawful would effectively require firms to price at competitive levels, a duty that neither Canadian nor U.S. competition law imposes.[117] Canadian law follows the same approach: the Competition Act targets agreements to fix prices, not independent decision-making that produces similar outcomes.
4. Algorithms do not create agreements
Algorithmic pricing does not change the statutory framework. The draft Guidelines note that algorithms may “soften[] competition among a jointly dominant group of firms,”[118] but that observation must be assessed within the legal standards described above. When competitors use the same pricing software, they form separate vertical relationships with the software vendor. To convert those vertical relationships into a horizontal conspiracy—a hub-and-spoke arrangement—the evidence must show a “rim”: an agreement among the competitors themselves to restrict competition.
The Guidelines recognise this distinction when discussing information sharing through third parties. They note that “firms can give commercially sensitive information to a third-party algorithm developer that provides pricing recommendations to industry participants,”[119] and that this “could lead to coordinated outcomes without any direct information sharing between industry participants.” Absent an agreement, however, such outcomes amount to conscious parallelism, not a violation of section 45 or an agreement actionable under section 90.1.
U.S. courts have applied this reasoning in algorithmic-pricing cases and rejected expansive hub-and-spoke theories. In Gibson v. Cendyn, plaintiffs alleged that a set of vertical software-licensing agreements, adopted independently by competing hotels at different times, produced anticompetitive effects in the aggregate—even though “neither the terms nor the operation of the licensing agreements [were] alleged to have harmed competition by affecting the competitive incentives in the relevant market, nor did the licensing agreements restrain any party’s ability to compete.”[120] The court held that conclusory allegations were insufficient even at the pleading stage; plaintiffs must still plead “a restraint of trade that causes an actual adverse effect on competition.”[121]
Similarly, in Cornish-Adebiyi v. Caesars, courts observed that firms often adopt pricing software at different times and with different customisations, which undercuts any inference of a horizontal agreement.[122] The legal question is not whether firms’ prices move in parallel, but whether the firms reached a consensus to coordinate their conduct.
5. Algorithmic pricing effects are context-dependent
The competitive effects of algorithmic pricing depend on market structure, not on the technology itself. Empirical research shows that algorithms can intensify competition in some settings and soften it in others, but the literature provides little support for concerns about autonomous coordination divorced from market conditions.
Efficiency Gains: In the airline industry, Kevin Williams finds that dynamic pricing increased output by 2.7% and total welfare by 1% by improving the match between supply and demand. Although consumer surplus fell for late-booking business travellers, overall welfare increased.[123]
Mixed Effects in Rental Housing: Sophie Calder-Wang and Gi Heung Kim study the rental-housing market and find that landlords using pricing algorithms reduced rents more quickly during demand downturns, including the Great Recession, improving occupancy rates.[124] While rents rose more quickly during expansion periods, the technology facilitated market clearing that benefited consumers in weaker markets.
German Gasoline Markets: The most prominent study is Heski Bar-Isaac Assad et al.’s analysis of German retail-gasoline stations.[125] The authors find that algorithmic pricing increased average margins by roughly 9%, with effects varying sharply by market structure. In monopoly markets, adoption had little effect. In duopoly markets where both stations adopted pricing software, margins rose by 38% relative to markets where neither adopted.[126] The authors describe this outcome as “tacit collusion,” noting that algorithms learned not to undercut because price cuts would be quickly matched.[127]
These findings reflect conscious parallelism, not tacit agreement. The German gasoline stations did not communicate, signal conditional strategies, or coordinate through software vendors. Each station independently adopted pricing software, and each algorithm independently learned that aggressive price cuts were unprofitable. The parallel outcome emerged from independent optimisation, not from any “meeting of the minds.”[128]
The policy significance of Assad et al. lies as much in what the study does not find as in what it does. The authors identify no evidence of communication among competitors, no agreement to adopt common software or pricing rules, no exchange of competitively sensitive information, and no signalling that could establish agreement. The observed margin increases resulted from independent adoption of similar tools responding to similar market conditions—the algorithmic analogue of oligopolists independently recognising their mutual interdependence. As the Guidelines correctly recognise, this form of conscious parallelism is not actionable under Canadian competition law.[129]
6. ‘Plus factors’ are still required
The Guidelines appropriately recognise that parallel conduct, even when facilitated by algorithms, requires “plus factors” to establish an agreement. As the Bureau explains, “when firms take part in parallel conduct along with practices that can facilitate coordination, this can be proof of an agreement between them.”[130] Relevant plus factors include sharing competitively sensitive information, agreeing to use common pricing parameters or data sources, or communicating in ways that establish coordination through algorithmic tools.[131]
Absent such plus factors, enforcement against algorithmic pricing would require the Bureau to condemn conscious parallelism—an approach the Guidelines expressly reject. That restraint is well-founded for several reasons.
First, the statutory text requires an “agreement.” Extending liability to tacit coordination arising from independent decision-making would require legislative change, not interpretive expansion.
Second, conscious parallelism has long existed in oligopolistic markets. Algorithms may speed price adjustments, but they do not introduce a new form of coordination. Firms observed and matched rivals’ prices long before algorithmic tools existed; software simply automates that process.
Third, enforcement against algorithmic tacit coordination would risk chilling beneficial innovation. Pricing algorithms lower barriers to entry by giving smaller firms access to tools once available only to large incumbents with dedicated revenue-management teams. Treating the use of similar software as evidence of coordination would disproportionately harm smaller and newer competitors.
Fourth, the competitive effects of algorithmic pricing remain context dependent. The same technology that Heski Bar-Isaac Assad et al. associate with higher margins in German gasoline markets also correlates with faster price declines during demand downturns in Sophie Calder-Wang and Gi Heung Kim’s rental-housing study. Categorical presumptions of harm would sacrifice efficiency gains in competitive markets to address risks that arise mainly in already-concentrated ones.
The Guidelines should therefore affirm that the legal test for agreements under section 90.1 and section 45 does not change because firms use algorithmic tools. Enforcement should focus on conduct that crosses the line from conscious parallelism to agreement, such as exchanging competitively sensitive information through shared platforms, agreeing on common pricing parameters, or communicating to establish coordinated strategies. Independent adoption of similar pricing technologies, even where it leads to parallel outcomes, remains lawful conscious parallelism beyond the reach of the Act.
F. Caution in Applying Competition Law to Labour Markets
The Guidelines recognise that firms may hold market power on the buying side of markets, including in labour markets.[132] If the Bureau intends to expand enforcement to address labour-market monopsony, it should proceed with caution. The empirical support for aggressive labour-market antitrust enforcement remains limited, and applying traditional competition frameworks to labour markets raises substantial conceptual and practical challenges.
1. Mixed evidence on labour-market power
Estimates of labour-supply elasticity—a central indicator of labour-market power—vary widely across studies. As Suresh Naidu, Eric Posner, and Glen Weyl report, residual labour-supply elasticities range from 0.1 to 4.2, implying that workers receive anywhere from 9% to 81% of their marginal product.[133] This dispersion shows that labour-market power is neither uniform nor pervasive. Low elasticity, and thus substantial monopsony power, does not characterise every labour market, and even high average estimates do not imply that power exists in all settings.
Steven Berry, Martin Gaynor, and Fiona Scott Morton caution that many empirical studies “provide interesting descriptions of concentration and wages but are not ultimately informative about whether monopsony power has grown and is depressing wages.”[134] The gap between concentration measures and competitive effects appears even wider in labour markets than in product markets.
2. Limits of concentration measures in labour markets
Labour markets have features that make concentration metrics a poor proxy for market power. Workers are highly mobile. When firms exit or reduce employment, workers often move across industries or occupations, not just to direct competitors. For many workers, the relevant market is therefore broader than a single industry or location, which makes conventional market-definition exercises unreliable.[135]
Non-competitive frictions also explain much wage variation. Search costs, geographic preferences, family obligations, and occupational licensing shape employment choices and pay, independent of employer concentration. These frictions fall outside the reach of competition law. A merger that raises concentration in a narrowly defined labour market, for example, will not affect wages if workers’ binding constraints stem from commuting distance, licensing rules, or similar non-competitive factors.[136]
3. Unresolved worker-consumer trade-offs
Traditional antitrust analysis centres on harms to consumers of final goods. Labour-market enforcement, by contrast, requires regulators to balance potential harms to workers against possible benefits to downstream consumers. Existing legal frameworks handle this trade-off poorly.[137] A merger that lowers wages may also lower consumer prices. As commentators note, “neither ignoring upstream harms nor disregarding downstream benefits provides a satisfactory solution under current doctrine.”[138]
The 2023 U.S. Merger Guidelines attempt to bypass this problem by asserting that “a merger’s harm to competition among buyers is not saved by benefits to competition among sellers.”[139] That position does not resolve the core issue. Competition law offers no principled method to compare a dollar of worker harm with a dollar of consumer benefit.
4. Labour regulation as the primary tool
Many labour-market concerns are better addressed through targeted labour regulation rather than through competition enforcement. Minimum-wage laws, collective-bargaining frameworks, limits on non-compete clauses, and reforms to occupational licensing can improve worker welfare directly. These tools allow policymakers to tailor responses to specific problems without forcing competition authorities to resolve difficult questions about market definition and competitive effects in settings where the economic relationships remain uncertain.
IV. Recommendations for Revision
This section sets out targeted revisions to make the Guidelines more predictable, administrable, and consistent with sound competition policy. It recommends an explicit effects-based enforcement focus, anchored in clear evidence of substantial competitive harm and calibrated to the asymmetric costs of enforcement errors. It also proposes practical limiting principles—such as market-power screens, safe harbours, and clarifications of novel provisions—to reduce the risk that ambiguous standards deter efficient pricing, contracting, innovation, and platform design. Together, these recommendations seek to preserve robust competition enforcement while giving businesses clearer and more reliable compliance guidance.
A. Procedural Fairness Is Essential to Effective Enforcement
As Canada implements its strengthened competition-law framework, the Bureau should commit to strong procedural protections. Experience in other jurisdictions shows that aggressive enforcement without adequate safeguards undermines both the legitimacy and effectiveness of competition policy.
Recent European cases illustrate the risk. In Qualcomm and Intel, appellate courts overturned European Commission decisions after finding procedural failures, including inadequate documentation of interviews and insufficient consideration of exculpatory evidence.[140] Those reversals imposed substantial costs: years of litigation, damage to the credibility of enforcement, and prolonged uncertainty for firms seeking to comply with the law.
The European Union’s experience with the Digital Markets Act offers a further warning. Commentators have criticised the DMA’s implementing framework for prioritising “procedural effectiveness over procedural fairness,” increasing the likelihood that courts will check enforcement through judicial review.[141] An enforcement regime that cannot withstand appellate scrutiny does not advance competition policy; it generates uncertainty and litigation costs.
Procedural safeguards protect investigated parties from administrative overreach and protect the Bureau from costly reversals. They also help ensure that enforcement resources focus on well-founded cases. Canada’s adversarial system, centred on the Competition Tribunal as an independent decision-maker, provides important structural protections that the Bureau should preserve and reinforce, not bypass in the name of efficiency.
The Guidelines should therefore include an explicit commitment to:
- providing timely access to the evidence relied on by the Bureau;
- documenting all substantive interviews and giving due weight to exculpatory evidence;
- allowing sufficient time for meaningful responses to Bureau concerns; and
- subjecting any interim measures to appropriate judicial oversight.
B. Prioritise Clear Harm, Not Close Calls
The Guidelines should state clearly, in both the introduction and the executive summary, that the Bureau will prioritise enforcement actions supported by clear evidence of substantial competitive harm. Where effects are ambiguous, disputed, or likely to be offset by efficiencies, the Bureau should exercise restraint to avoid chilling pro-competitive conduct.
This commitment does not require the Bureau to overlook conduct that meets statutory thresholds. It reflects practical enforcement realities and the need to focus limited resources on cases most likely to benefit consumers. It also aligns with an error-cost framework: because false positives impose greater and more durable costs than false negatives, the Bureau should favour non-intervention in close cases.
C. Clear Screens for Market Power
The Guidelines should identify market conditions that generally rule out a finding of market power for purposes of Competition Act enforcement:
- Market share below 30%, absent extraordinary circumstances;
- Competitive markets with several significant firms, where the largest firm’s share does not exceed the combined share of the next three competitors;
- Dynamic markets marked by rapid innovation, low entry barriers, and frequent entry or expansion; and
- Markets with easy multi-homing or switching, where customers readily use alternatives.
Where shares exceed these thresholds, the Bureau should also list factors that rebut any inference of market power based on share alone. These include ease of entry, rapid technological change, demonstrated rival expansion, strong customer bargaining power, or evidence that the firm cannot raise prices profitably without losing substantial sales.
D. Excessive Pricing Requires Exclusion
The Guidelines should state clearly that high prices, by themselves, do not constitute anti-competitive conduct under section 78(1)(k). The Bureau should assess pricing levels only as evidence of, or in connection with, conduct that prevents competitive entry or expansion. Relevant examples include:
- Margin squeezes, where a vertically integrated firm prices inputs to downstream rivals in ways that foreclose efficient competition;
- Tying arrangements, where a monopoly product is bundled with competitively supplied products at prices that extend market power; and
- Loyalty discounts or exclusive dealing, where pricing terms substantially foreclose rivals’ access to the market.
The Guidelines should also clarify that returns reflecting innovation, intellectual property, risk-taking, or superior efficiency do not amount to excessive or unfair pricing. The Bureau should commit not to rely on cost-plus regulation or simple price–cost margin tests, given the well-recognised difficulties in allocating common costs, valuing intangible capital, and adjusting for risk.
E. Safe Harbours for Vertical Agreements
For section 90.1 cases involving vertical agreements, the Guidelines should establish a presumption of legality where one or more of the following conditions apply:
- Neither party holds a market share above 30% in any relevant market;
- The restraint forecloses no more than 30% of the market to rivals;
- The restraint is limited in duration and includes reasonable termination rights;
- Rivals retain access to substantial alternative distribution channels; or
- The restraint serves a documented efficiency objective, such as preventing free riding, ensuring quality, or eliminating double marginalisation.
The Guidelines should also identify common pro-competitive rationales for vertical restraints and commit to enforcement only where evidence shows net competitive harm. They should remove any presumption that significant competitive effects alone establish an anti-competitive purpose. Instead, the Bureau should first assess whether legitimate business justifications plausibly explain the arrangement and consider purpose only where no such efficiency rationale exists.
F. Self-Preferencing Is Usually Pro-Competitive
The Guidelines should recognise that self-preferencing usually reflects vertical integration and often delivers pro-competitive benefits. The Bureau should challenge self-preferencing only where all three of the following conditions are met:
- The platform holds durable market power protected by substantial barriers to entry or expansion;
- Users face meaningful obstacles to multi-homing or switching, allowing the platform to foreclose competition profitably; and
- The conduct causes demonstrable consumer harm—such as higher prices, reduced quality, or diminished innovation—after accounting for any efficiency gains.
The Guidelines should also state clearly that product integration, quality control, security, and investment incentives commonly justify self-preferencing. Difficulty faced by rivals on a particular platform does not, on its own, establish competitive harm where those rivals can reach customers through viable alternative channels.
G. Algorithms Don’t Replace Agreement
The Guidelines should affirm that section 90.1 requires an agreement—a meeting of the minds—regardless of whether pricing decisions are made by humans or algorithms. Parallel pricing, even when produced by similar algorithms, does not establish an agreement without evidence of coordination.
The Bureau should therefore focus on cases with clear indicia of coordination, such as the exchange of competitively sensitive information, agreements to use common pricing parameters or data sources, or communications that establish coordinated strategies through algorithmic tools. Independent adoption of similar pricing strategies remains lawful conscious parallelism.
H. Commitment to Review and Adapt
The Guidelines should state that the Bureau will monitor how business practices respond to the ACCA Guidelines and assess whether enforcement actions achieve their intended outcomes. The Bureau should commit to updating its guidance as needed, drawing on decisional experience, economic research, and stakeholder feedback.
This commitment reflects appropriate regulatory humility. Competition policy operates in dynamic markets where business models, technologies, and competitive conditions evolve quickly. A willingness to adjust positions as evidence develops will strengthen both the credibility and the effectiveness of the Guidelines.
V. Conclusion
The Competition Bureau faces a demanding task in implementing the amended Competition Act. Parliament has expanded the Bureau’s authority—most notably through the new excessive-pricing provision and the extension of section 90.1 to vertical agreements—while leaving key concepts open-textured and highly context dependent. Without clear limiting principles, these provisions risk deterring conduct that often benefits consumers, including efficient pricing, vertical coordination, product integration, and innovation.
Our comments urge the Bureau to anchor the Guidelines in an explicit error-cost framework. False positives—mistaken condemnation of pro-competitive conduct—impose durable and often irreversible costs by chilling investment, innovation, and rivalry. By contrast, false negatives are more likely to self-correct through entry, expansion, and technological change. In close or ambiguous cases, restraint is therefore not leniency but sound enforcement policy.
Several areas warrant particular care. Excessive pricing under section 78(1)(k) should apply only where pricing forms part of exclusionary conduct that blocks entry or expansion, not where high prices reflect innovation, intellectual property, risk-taking, or superior efficiency. Vertical agreements reviewed under section 90.1 should benefit from safe harbours and a presumption of legality, consistent with extensive evidence—surveyed by Francine Lafontaine and Margaret Slade—that such arrangements usually enhance consumer welfare. Self-preferencing should be analysed as a form of vertical integration, with enforcement limited to cases involving durable market power, meaningful limits on multi-homing, and demonstrable consumer harm. Algorithmic pricing should not dilute the Act’s core requirement that the Bureau prove an agreement—a meeting of the minds—rather than lawful conscious parallelism, regardless of whether pricing decisions are made by humans or software.
More broadly, the Guidelines should reinforce procedural fairness, clear screens for market power, and an effects-based enforcement focus grounded in evidence of substantial competitive harm. They should also reflect regulatory humility by committing to review and adaptation as markets, technologies, and business practices evolve.
Taken together, the revisions we propose would make the Guidelines more predictable, administrable, and economically grounded. They would help focus Bureau resources on conduct most likely to harm competition and consumers, while reducing the risk that uncertain standards chill efficient pricing, contracting, innovation, and platform design.
We appreciate the Bureau’s openness to public input through this consultation. ICLE remains available to provide further analysis or clarification on any of the issues raised in these comments.
[1] The International Center for Law & Economics (ICLE) has received financial support from numerous companies, foundations, and individuals, including firms with interests both supportive of and in opposition to the ideas expressed in this and other ICLE-supported works. Unless otherwise noted, all ICLE support is in the form of unrestricted, general support. The ideas expressed here are the authors’ own and do not necessarily reflect the views of ICLE’s advisors, affiliates, or supporters.
[2] Competition Bureau, Anti-Competitive Conduct and Agreements Enforcement Guidelines (Oct. 31, 2025), https://competition-bureau.canada.ca/en/how-we-foster-competition/consultations/anti-competitive-conduct-and-agreements [hereinafter “Draft ACCA Guidelines”].
[3] Geoffrey Manne, Dirk Auer, Daniel Gilman, & Lazar Radic, Comments from the International Center of Law and Economics on The Future of Competition Policy in Canada (Mar. 31, 2023), https://laweconcenter.org/wp-content/uploads/2023/03/ICLE-Canada-Comments.pdf [hereinafter “Appendix A”].
[4] Geoffrey Manne, Dirk Auer, Brian Albrecht, Daniel Gilman & Lazar Radic, Comments of the International Center for Law & Economics: Canadian Competition Bureau Discussion Paper on Algorithmic Pricing and Competition (Aug. 4, 2025), https://laweconcenter.org/wp-content/uploads/2025/08/Canada-Algorithmic-pricing-and-competition-1.pdf [hereinafter “Appendix B”].
[5] Competition Act, R.S.C. 1985, c. C-34, s. 78(1)(k). See also An Act to Amend the Excise Tax Act and the Competition Act, S.C. 2023, c. 31, pt. 2 [hereinafter “Bill C-56”] and Fall Economic Statement Implementation Act, 2023, S.C. 2024, c. 15, pt. 5, div. 6 [hereinafter “Bill C-59”].
[6] Competition Act, s. 90.1(1.01).
[7] Competition Act, s. 79(3.1).
[8] See Eric Fruits, Do Biased Stats Provide Bogus Economics? A Primer on Publication Bias and Power, Truth on the Mkt. (Mar. 29, 2018), https://truthonthemarket.com/2018/03/29/do-biased-stats-provide-bogus-economics-a-primer-on-publication-bias-and-power.
[9] Frank H. Easterbrook, The Limits of Antitrust, 63 Tex. L. Rev. 1, 2 (1984) (“If the court errs by condemning a beneficial practice, the benefits may be lost for good. Any other firm that uses the condemned practice faces sanctions in the name of stare decisis, no matter the benefits. If the court errs by permitting a deleterious practice, though, the welfare loss decreases over time. Monopoly is self-destructive. Monopoly prices eventually attract entry.”)
[10] Id. 15.
[11] Brian C. Albrecht & Ryan A. Decker, Markups and Business Dynamism across Industries, Int’l J. Indus. Org. (forthcoming 2026), available at https://www.briancalbrecht.com/Albrecht_Decker_Markups.pdf.
[12] See Michael L. Katz & Carl Shapiro, Network Externalities, Competition, and Compatibility, 75 Am. Econ. Rev. 424 (1985) (analyzing how network effects can create barriers to entry); Joseph Farrell & Paul Klemperer, Coordination and Lock-In: Competition with Switching Costs and Network Effects, in 3 Handbook of Industrial Organization 1967 (Mark Armstrong & Robert Porter eds., 2007) (surveying conditions under which switching costs and network effects impede competitive displacement of incumbents). Easterbrook himself acknowledged that “the ability of the market to correct monopoly depends on the existence of conditions that allow entry.” Easterbrook, supra note 9, at 15.
[13] Geoffrey A. Manne & Joshua D. Wright, Innovation and the Limits of Antitrust, 6 J. Competition L. & Econ. 153, 167 (2010) (“In the innovation context, successfully challenging business or product innovations is likely to dampen innovation across the economy, whereas Type 2 errors are at least mitigated in part by entry and other competition. When viewed through the error-cost lens, the combination of the anti-market bias in favor of monopoly explanations for innovative conduct that courts and economists do not understand, and the increased stakes of antitrust intervention against innovative business practices, is problematic from a consumer-welfare perspective.”).
[14] Id. 168 (“the consequences of Type 1 error are magnified by the threat of erroneous intervention further deterring subsequent, similar innovation or applications of innovations in novel settings”).
[15] Draft ACCA Guidelines Preface vii.
[16] See Joshua D. Wright & Murat C. Mungan, The Easterbrook Theorem: An Application to Digital Markets, Yale L. J. Forum 622, 626–30 (Jan. 18, 2021) (discussing the development of U.S. case law based on Easterbrook’s framework); see also id. at 625 n. 12 & 626 n. 13 (listing cases).
[17] Competition Act, s. 103.1(7); see also Competition Bureau, Information Bulletin on Private Access to the Competition Tribunal (Jun. 2025), https://competition-bureau.canada.ca/en/information-bulletin-private-access-competition-tribunal.
[18] Competition Act, s. 103.1(7); see also Norton Rose Fulbright, Competition Act Amendments Expand Opportunities for Private Litigation (Jun. 2024), https://www.nortonrosefulbright.com/en/knowledge/publications/68b7a977/competition-act-amendments-expand-opportunities-for-private-litigation.
[19] Norton Rose Fulbright, Competition Act: Expanded Private Enforcement Rights Now in Force (Jun. 20, 2025), https://www.nortonrosefulbright.com/en-ca/knowledge/publications/c5f0e8cd/competition-act-expanded-private-enforcement-rights-now-in-force (describing the commissioner’s intervention in JAMP Pharma Corp. v. Janssen Inc., CT-2024-005).
[20] William Breit & Kenneth G. Elzinga, Private Antitrust Enforcement: The New Learning, 28 J.L. & Econ. 405 (1985).
[21] R. Preston McAfee, Hugo M. Mialon, & Sue H. Mialon, Private Antitrust Litigation: Procompetitive or Anticompetitive?, in The Political Economy of Antitrust 453, 453 (Vivek Ghosal & Johan Stennek eds., 2007), https://www.sciencedirect.com/science/article/abs/pii/S0573855506820171.
[22] Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 489 (1977)); see also, Cargill, Inc. v. Monfort of Colorado, Inc., 479 U.S. 104, 111–13 (1986); Califano v. Yamasaki, 442 U.S. 682, 702 (1979).
[23] Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 489 (1977).
[24] Jonathan M. Jacobson & Tracy Greer, Twenty-One Years of Antitrust Injury: Down the Alley with Brunswick v. Pueblo Bowl-O-Mat, 66 Antitrust L.J. 274 (1998).
[25] Competition Act, s. 79(3.1)-(3.2).
[26] A. Mitchell Polinsky & Steven Shavell, The Optimal Tradeoff Between the Probability and Magnitude of Fines, 69 Am. Econ. Rev. 880 (1979); A. Mitchell Polinsky & Steven Shavell, The Theory of Public Enforcement of Law, in 1 Handbook of Law and Economics 403 (A. Mitchell Polinsky & Steven Shavell eds., 2007).
[27] Herbert Hovenkamp, A Primer on Antitrust Damages 8 (Univ. of Iowa Legal Studies Research Paper No. 10-45, 2011), at 6, https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1685919.
[28] Competition Act, s. 75(1.2), 76(10.1), 77(3.1), 79(3.3), and 90.1(1.2); see Osler, Hoskin & Harcourt LLP, The Dramatic Expansion of Private Enforcement of Canada’s Competition Laws (2024), https://www.osler.com/en/insights/reports/the-dramatic-expansion-of-private-enforcement-of-canadas-competition-laws.
[29] See Global Competition Review, United States: Private Antitrust Litigation (2022), https://globalcompetitionreview.com/review/the-antitrust-review-of-the-americas/2022/article/united-states-private-antitrust-litigation (documenting combined plaintiff-defendant expenditures exceeding $13 million in single-plaintiff cases litigated through trial).
[30] Edward Iacobucci, Uncertainty and the Burden of Proof in Canadian Competition Law, C.D. Howe Inst. Commentary No. 659, at 4, 5 (May 2024), https://cdhowe.org/publication/uncertainty-and-burden-proof-canadian-competition-law.
[31] Comm’r of Competition v. Can. Pipe Co., (2006) F.C.A. 233.
[32] Id.
[33] Id.
[34] Competition Act, s. 78(1)(k).
[35] Treaty on the Functioning of the European Union art. 102, May 9, 2008, 2008 O.J. (C 115) 47.
[36] Case 27/76, United Brands Co. v. Comm’n, ECLI:EU:C:1978:22, ¶ 252 (Feb. 14, 1978) (“The questions therefore to be determined are whether the difference between the costs actually incurred and the price actually charged is excessive, and, it the answer to this question is in the affirmative, whether a price has been imposed which is either unfair in itself or when compared to competing products.”)
[37] See, e.g., Aline Blankertz, Todd Davies, Justine Haekens, & Nicholas Shaxson, The European Commission Can and Must Act on Excessive Pricing, ProMarket (Oct. 8, 2025), https://www.promarket.org/2025/10/08/the-european-commission-can-and-must-act-on-excessive-pricing.
[38] See Org. for Econ. Coop. & Dev., Excessive Prices, OECD Doc. DAF/COMP(2011)18 (Feb. 7, 2012), https://www.oecd.org/content/dam/oecd/en/publications/reports/2012/02/excessive-prices_372e092c/8e1fd82e-en.pdf; Raphaël De Coninck, Excessive Prices: An Overview of EU and National Case Law, E-Competitions Bull. (Jun. 2018), https://ecp.crai.com/wp-content/uploads/2018/06/Excessive-Pricing-R.-de-Coninck-e-Competitions-2018.pdf.
[39] David S. Evans & A. Jorge Padilla, Excessive Prices: Using Economics to Define Administrable Legal Rules, 1 J. Competition L. & Econ. 97 (2005).
[40] Massimo Motta & Alexandre de Streel, Excessive Pricing and Price Squeeze under EU Law, in The Pros and Cons of High Prices 14 (Konkurrensverket ed., 2007), https://www.konkurrensverket.se/globalassets/dokument/informationsmaterial/rapporter-och-broschyrer/pros-and-cons/rapport_pros-and-cons_2007_high_prices.pdf.
[41] Lars-Hendrik Röller, Exploitative Abuses, (ESMT Business Brief No. BB-107-002, 2007), https://d-nb.info/1012903311/34.
[42] Draft ACCA Guidelines, ¶¶ 263-267 (“Excessive and unfair pricing is likely to raise issues only if it is used to engage in other types of anti-competitive conduct or agreements… In such cases, we would generally investigate the pricing as part of the other form of conduct or agreement (e.g., refusal to deal, bundling, margin squeezing).”)
[43] Draft ACCA Guidelines, ¶ 81.
[44] Draft ACCA Guidelines, ¶ 336.
[45] Brian Albrecht, Competition Increases Concentration, Truth on the Mkt. (Aug. 16, 2023), https://truthonthemarket.com/2023/08/16/competition-increases-concentration.
[46] See Brian Albrecht, What Competition Scholars Should Know About the 2025 Economics Nobel, Truth on the Mkt. (Oct. 14, 2025), https://truthonthemarket.com/2025/10/14/what-competition-scholars-should-know-about-the-2025-economics-nobel/ (“Market structure is an outcome of this competitive process, not just a cause of competitive behavior.”).
[47] Chad Syverson, Macroeconomics and Market Power: Context, Implications, and Open Questions, 33 J. Econ. Persps. 23, 34 (2019).
[48] Harold Demsetz, Industry Structure, Market Rivalry, and Public Policy, 16 J.L. & Econ. 1 (1973); see also Harold Demsetz, The Market Concentration Doctrine: An Examination of Evidence and a Discussion of Policy (AEI–Hoover Policy Studies No. 7, 1973), https://masonlec.org/site/rte_uploads/files/GAI/Readings/Economics%20Institute/Demsetz_Market%20Concentration%20Doctrine.pdf.
[49] William M. Landes & Richard A. Posner, Market Power in Antitrust Cases, 94 Harv. L. Rev. 937, 947 (1981); see also Mario Zúñiga, ‘Market Power in Antitrust Cases,’ by William M. Landes and Richard A. Posner, Truth on the Mkt. (Jan. 16, 2026), More recently, Dennis Carlton has described market shares and concentration as “at best crude first steps,” noting that empirically there is “only a weak link between change in market share and change in competitive performance,” and that “there is no model” in which market share alone reliably predicts prices or welfare effects. Static market-share data therefore cannot substitute for analysis of firm conduct and market conditions. https://truthonthemarket.com/2026/01/16/market-power-in-antitrust-cases-by-william-m-landes-and-richard-a-posner.
[50] Dennis W. Carlton, Market Definition: Use and Abuse, 3 Competition Pol’y Int’l 3 (2007), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=987061.
[51] William J. Baumol, John C. Panzar & Robert D. Willig, Contestable Markets and the Theory of Industry Structure 222 (rev. ed. 1988), available at https://archive.org/details/contestablemarke0000baum_e1g8 (“[F]reedom of entry, indeed the mere threat of incursions by entrants into the market, may effectively discipline the monopolist, even if entry is never successful. It can force the monopolist to curb his avarice and forgo profits he might otherwise have enjoyed. Indeed, in the absence of entry barriers, in perfectly contestable markets, it can force him to accept earnings no higher than those available under perfect com¬ petition. Potential competition can also force the monopolist to produce with maximal efficiency, and to hunt down and utilize fully every opportunity for innovation.”); see also David J. Teece, Understanding Dynamic Competition: New Perspectives on Potential Competition, “Monopoly,” and Market Power (May 22, 2025), available at https://ssrn.com/abstract=5356023.
[52] Id.
[53] Org. for Econ. Coop. & Dev., Competition in the Pharmaceutical Industry, OECD Doc. DAFFE/CLP(2000)29 (Feb. 6, 2001) (“For many products which face few rivals in their therapeutic class, the primary competitive threat is the threat that rival firms will develop substitutes.”)
[54] Stephen J. Nickell, Competition and Corporate Performance, 104 J. Pol. Econ. 724 (1996).
[55] Richard Blundell, Rachel Griffith, & John Van Reenen, Market Share, Market Value and Innovation in a Panel of British Manufacturing Firms, 66 Rev. Econ. Stud. 529 (1999).
[56] Competition Act, s. 79(1).
[57] Verizon Commc’ns Inc. v. Law Offices of Curtis V. Trinko, LLP, 540 U.S. 398 (2004).
[58] Id. 407.
[59] David S. Evans & Richard Schmalensee, The Antitrust Analysis of Multi-Sided Platform Businesses (Nat’l Bureau of Econ. Rsch. Working Paper No. 18783, 2013), http://www.nber.org/papers/w18783.
[60] Jean‑Charles Rochet & Jean Tirole, Platform Competition in Two‑Sided Markets, 1 J. Eur. Econ. Ass’n 990 (2003); Mark Armstrong, Competition in Two‑Sided Markets, 37 RAND J. Econ. 668 (2006).
[61] David S. Evans, Two-Sided Platforms and Analysis of Single-Firm Conduct, Comment to the U.S. Dep’t of Justice Single-Firm Conduct Hearings (Sep. 2006), https://www.justice.gov/archives/atr/single-firm-conduct-hearings-comment-david-evans-two-sided-platforms-and-analysis-single-firm.
[62] David S. Evans, The Antitrust Economics of Multi‑Sided Platform Markets, 20 Yale J. Reg. 325, 334–35 (2003).
[63] Herbert Hovenkamp, The Areeda–Turner Test for Exclusionary Pricing: A Critical Journal, 46 Rev. Indus. Org. 209, 213-215 (2015).
[64] Id. 216.
[65] Einer Elhauge, Why Above‑Cost Price Cuts to Drive Out Entrants Are Not Predatory, 112 Yale L.J. 681, 727–31 (2003).
[66] William J. Baumol & David F. Bradford, Optimal Departures from Marginal Cost Pricing, 60 Am. Econ. Rev. 265 (1970); see also Gus Hurwitz, ‘Optimal Departures from Marginal Cost Pricing’ by William J. Baumol & David F. Bradford, Truth On The Mkt. (Sep. 30, 2025), https://truthonthemarket.com/2025/09/30/optimal-departures-from-marginal-cost-pricing-by-william-j-baumol-david-f-bradford.
[67] OECD, Excessive Prices in Pharmaceutical Markets ¶¶ 22, 104–22, DAF/COMP(2018)12 (2018), https://one.oecd.org/document/DAF/COMP(2018)12/en/pdf.
[68] Id. 21-22, citing David S. Evans & A. Jorge Padilla, Excessive Prices: Using Economics to Define Administrable Legal Rules, 1 J. Competition L. & Econ. 97, 118 (2005).
[69] Motta & de Streel, supra note 41.
[70] Draft ACCA Guidelines, § 6.11.
[71] Id. at ¶ 264.
[72] Giuseppe Colangelo, In Fairness We (Should Not) Trust: The Duplicity of the EU Competition Policy Mantra in Digital Markets, 68 Antitrust Bull. 618, 635 (2023).
[73] Id. at 621.
[74] See Eliana Garcés and Giuseppe Colangelo, Markets, Competition, and Fairness in Research Handbook on Competition and Corporate Law 83, 96–102 (Thepot & Tzanaki, eds. 2025) (discussing the implications of fairness in the process of the allocation of rights and privileges versus fairness of outcome).
[75] Innovation, Sci. & Econ. Dev. Can., The Future of Competition Policy in Canada (2022), https://ised-isde.canada.ca/site/strategic-policy-sector/sites/default/files/attachments/2022/The-Future-of-Competition-Policy-eng.pdf.
[76] Competition Act, s. 79(1).
[77] Draft ACCA Guidelines, ¶ 266.
[78] Competition Act, s. 90.1(1.01).
[79] Francine Lafontaine & Margaret E. Slade, Exclusive Contracts and Vertical Restraints: Empirical Evidence and Public Policy, in Handbook of Antitrust Economics 391 (Paolo Buccirossi ed., 2008); see also Francine Lafontaine & Margaret E. Slade, Transaction Cost Economics and Vertical Market Restrictions—Evidence, 55 Antitrust Bull. 587 (2010); Geoffrey Manne, Kristian Stout, & Eric Fruits, The Fatal Economic Flaws of the Contemporary Campaign against Vertical Integration, 68 U. Kan. L. Rev. 923 (Jun. 2020).
[80] Benjamin Klein & Kevin M. Murphy, Vertical Restraints as Contract Enforcement Mechanisms, 31 J.L. & Econ. 265 (1988).
[81] Leegin Creative Leather Prods., Inc. v. PSKS, Inc., 551 U.S. 877 (2007).
[82] Draft ACCA Guidelines, ¶ 430.
[83] Competition Act, s. 90.1(1).
[84] Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477 (1977) (The injury must be “of the type the antitrust laws were intended to prevent,” meaning the harm must result from a reduction in competition, not just a loss of profits. Also, the injury must “flow from that which makes the defendant’s acts unlawful.”).
[85] Ill. Brick Co. v. Illinois, 431 U.S. 720 (1977).
[86] Draft ACCA Guidelines, s. 6.4, see ¶ 197 (“In most cases we assess self-preferencing as exclusionary conduct. We may investigate self-preferencing under the abuse of dominance provisions.”).
[87] Draft ACCA Guidelines, ¶¶ 196, 198.
[88] Draft ACCA Guidelines, ¶ 199.
[89] Ronald H. Coase, The Nature of the Firm, 4 Economica 386 (1937).
[90] Appendix A s. II.
[91] Chiara Farronato, Andrey Fradkin, & Alexander MacKay, Vertical Integration and Consumer Choice: Evidence from a Field Experiment (Nat’l Bureau of Econ. Rsch. Working Paper No. 34135, 2025), https://www.nber.org/papers/w34135.
[92] Id. at 3.
[93] Brian Albrecht & Geoffrey A. Manne, Self-Preferencing Isn’t a Sin. It’s Often the Way Competition Works., Truth on the Mkt. (Aug. 20, 2025), https://truthonthemarket.com/2025/08/20/self-preferencing-isnt-a-sin-its-often-the-way-competition-works.
[94] Australian Competition and Consumer Commission, Digital Platform Services Inquiry—Interim Report No. 5: Regulatory Reform 94 (Sep. 2022), https://www.accc.gov.au/system/files/Digital%20platform%20services%20inquiry.pdf (“Although self-preferencing conduct is often benign, self-preferencing conduct that leverages market power over a key online service into a related service, which is not justified by a procompetitive rationale, can distort competition and decrease consumer welfare.”).
[95] Juliette Caminade, Juan Carvajal, & Christopher R. Knittel, An Economic Analysis of the Self-Preferencing Debate, 32 Competition 30 (2022), https://www.analysisgroup.com/globalassets/insights/publishing/2022-caminade-et-al-an-economic-analysis-of-the-self-preferencing-debate.pdf.
[96] Draft ACCA Guidelines, ¶ 72.
[97] Draft ACCA Guidelines, ¶ 90(g).
[98] See generally Stan J. Liebowitz & Stephen E. Margolis, Winners, Losers & Microsoft: Competition and Antitrust in High Technology (1999).
[99] Matthew T. Clements & Hiroshi Ohashi, Indirect Network Effects and the Product Cycle: Video Games in the U.S., 1994–2002, 53 J. Indus. Econ. 515 (2005).
[100] Catherine Tucker, Digital Data, Platforms and the Usual [Antitrust] Suspects: Network Effects, Switching Costs, Essential Facility, 54 Rev. Indus. Org. 683 (2019).
[101] See, e.g., Geoffrey Manne & Dirk Auer, Antitrust Dystopia and Antitrust Nostalgia: Alarmist Theories of Harm in Digital Markets and Their Origins, 28 Geo. Mason L. Rev. 1279 (Summer 2021).
[102] See Geoffrey A. Manne & Ben Sperry, Debunking the Myth of a Data Barrier to Entry for Online Services, Truth on the Mkt. (March 26, 2015), https://truthonthemarket.com/2015/03/26/debunking-the-myth-of-a-data-barrier-to-entry-for-online-services.
[103] See Harrison Jacobs, Former MySpace CEO Explains Why Facebook Was Able to Dominate Social Media Despite Coming Second, Business Insider (May 9, 2015), https://www.businessinsider.com/former-myspace-ceo-explains-why-facebook-was-able-to-dominate-social-media-despite-coming-second-2015-5.
[104] See, e.g., Bruno Jullien, Alessandro Pavan, & Marc Rysman, Two-Sided Markets, Pricing, and Network Effects, in 4 Handbook of Industrial Organization 485 (Kate Ho, Ali Hortaçsu & Alessandro Lizzeri eds., 2021).
[105] Draft ACCA Guidelines s. 6.12.
[106] Competition Bureau Canada, Algorithmic Pricing and Competition: Discussion Paper (Jun. 10, 2025).
[107] Maureen K. Ohlhausen, Acting Chairman, Should We Fear the Things That Go Beep in the Night? Some Initial Thoughts on the Intersection of Antitrust Laws and Algorithmic Pricing, Fed. Trade Comm’n (May 23, 2017), at 10, https://www.ftc.gov/system/files/documents/public_statements/1220893/ohlhausen_-_concurrences_5-23-17.pdf.
[108] Draft ACCA Guidelines, ¶ 417 (“An agreement involves a ‘meeting of the minds’ between two or more parties. That is, those parties have reached a consensus regarding their intentions.”).
[109] Draft ACCA Guidelines, ¶ 121.
[110] Draft ACCA Guidelines, ¶ 418.
[111] R. v. Nova Scotia Pharmaceutical Society, [1992] 2 S.C.R. 606, at 651-52 (holding that the conspiracy offense requires proof of “intention” and “agreement” constituting a “common design”).
[112] See Louis Kaplow, Competition Policy and Price Fixing 1-4 (2013) (explaining that “tacit collusion” describes oligopolistic interdependence rather than illegal agreement); William E. Kovacic et al., Plus Factors and Agreement in Antitrust Law, 110 Mich. L. Rev. 393, 400-01 (2011) (distinguishing “express” and “tacit” collusion and noting that only the former violates antitrust law).
[113] See Draft ACCA Guidelines, ¶¶ 121, 417-18 (distinguishing explicit agreements, tacit agreements, and conscious parallelism).
[114] Draft ACCA Guidelines, ¶ 417.
[115] Draft ACCA Guidelines, ¶ 121.
[116] Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209, 227 (1993).
[117] See Donald F. Turner, The Definition of Agreement Under the Sherman Act: Conscious Parallelism and Refusals to Deal, 75 Harv. L. Rev. 655, 663 (1962) (“The rational oligopolist is doing nothing more than what he would do in any event…. To say that this conduct should be unlawful is to say that the firms are under a duty to compete, which is hardly a duty imposed by the present antitrust laws.”).
[118] Draft ACCA Guidelines, fn. 106.
[119] Draft ACCA Guidelines, ¶ 272 (“firms can give commercially sensitive information to a third-party algorithm developer that provides pricing recommendations to industry participants”).
[120] Gibson v. Cendyn Grp., LLC, 148 F.4th 1069 (9th Cir. 2025).
[121] Id.
[122] Cornish-Adebiyi v. Caesars Entm’t (D.N.J. 2024).
[123] Kevin R. Williams, The Welfare Effects of Dynamic Pricing: Evidence from Airline Markets, 90 Econometrica 831 (2022).
[124] Sophie Calder-Wang & Gi Heung Kim, Algorithmic Pricing in Multifamily Rentals: Efficiency Gains or Price Coordination? (Aug. 16, 2024), https://ssrn.com/abstract=4403058.
[125] Stephanie Assad, Robert Clark, Daniel Ershov, & Lei Xu, Algorithmic Pricing and Competition: Empirical Evidence from the German Retail Gasoline Market, 132 J. Pol. Econ. 723 (2024).
[126] Id. at 726 (finding that algorithmic adoption increases margins by roughly 9% overall, with no meaningful effect in monopoly markets but a 38% increase in duopoly markets where both stations adopt).
[127] Id. at 727 (“Margins start to increase only about a year after market-wide adoption, suggesting that algorithms in this market learn tacitly collusive strategies…. suggests that algorithms learn that undercutting will not be profitable, since lower prices will be followed.”).
[128] Id. at 760 (“To the best of our knowledge, this occurs without explicit communication between competitors, making it legal according to current competition laws in many countries.”).
[129] Draft ACCA Guidelines, ¶ 121.
[130] Draft ACCA Guidelines, ¶ 418 (“when firms take part in parallel conduct along with practices that can facilitate coordination, this can be proof of an agreement between them”).
[131] See Draft ACCA Guidelines, ¶¶ 271-75 (discussing information sharing through third parties, including algorithm developers, as a potential facilitating practice).
[132] See id. at ¶¶ 31, 33, 51.
[133] Geoffrey A. Manne, Brian C. Albrecht, & Dirk Auer, Labor Monopsony and Antitrust Enforcement: A Distorting Mirror, 74 DePaul L. Rev. 1119 (2025); Suresh Naidu, Eric A. Posner, & Glen Weyl, Antitrust Remedies for Labor Market Power, 132 Harv. L. Rev. 536 (2018).
[134] Steven Berry, Martin Gaynor, & Fiona Scott Morton, Do Increasing Markups Matter? Lessons from Empirical Industrial Organization, 33 J. Econ. Persp. 44, 57 (2019).
[135] Manne et al., supra note 134 at 1153 (“It is virtually impossible to know what outside options to include in the relevant market, and it may not always be possible to identify even where such potential employers are located. . . . These market-definition issues are far more acute in monopsony cases than in traditional monopoly cases, both because the intrinsic question of substitutes is more complicated and because there is far less precedent to guide parties and enforcers.”).
[136] Id. at 1138.
[137] Id. at Part III.
[138] Manne et al., supra note 134 at 1173.
[139] U.S. Dep’t of Justice & Fed. Trade Comm’n, Merger Guidelines § 4.3 (2023).
[140] Case T-235/18, Qualcomm Inc. v. European Commission, ECLI:EU:T:2022:358 (Jun. 15, 2022) (in which the General Court of the European Union annulled the European Commission’s €997 million fine against Qualcomm. The court’s decision was largely based on “procedural irregularities” that affected Qualcomm’s rights of defense); Case T-286/09 RENV, Intel Corp. v. European Commission, ECLI:EU:T:2022:19 (Jan. 26, 2022); Case C-240/22 P, European Commission v. Intel Corp., ECLI:EU:C:2024:911 (Oct. 24, 2024) (Court of Justice of the European Union set aside an initial ruling in 2017, the General Court in 2022 annulled the Commission’s €1.06 billion fine).
[141] See, e.g., Alfonso Lamadrid, The DMA: Procedural Afterthoughts, Chillin’ Competition (Sep. 5, 2022), https://chillingcompetition.com/2022/09/05/the-dma-proceduralafterthoughts.
