Comments to the Competition Bureau of Canada Regarding Anti-competitive Conduct and Agreements Enforcement Guidelines

Contents

1. Introduction and Summary 1

2. Market Power 2

3. The Economics of Anti-competitive Conduct and Agreements 3

4. Assessing the Effects of Anti-competitive Conduct and Agreements 4

5. Remedies 4

6. Types of Anti-competitive Conduct and Agreements 6

7. The ACCA Provisions 12

Recommendations 16

Conclusion. 16

Endnotes 17

1. Introduction and Summary

On October 31st 2025, the Canadian Competition Bureau ("Bureau") issued draft Anti-competitive Conduct and Agreements Enforcement Guidelines (Draft Guidelines).[1]The Draft Guidelines follow significant changes to the Competition Act ("the Act") between 2022 and 2024. As the Draft Guidelines explain, the purpose of the Draft Guidelines is "to help firms comply with the Act."[2]

Canada faces a structural competitiveness challenge that makes the interpretation of these Draft Guidelines particularly consequential. Despite being the world's 9th largest economy, Canada ranks only 17th in productivity among OECD nations.[3]This productivity gap matters for competition policy because over enforcement can deter the very activities Canada needs to close it: building firms to efficient scale, vertical integration that drives innovation, aggressive price competition, strategic acquisitions that allow startups to reach global markets, and foreign direct investment that brings capital and expertise.

Other jurisdictions offer cautionary lessons. As the Information Technology and Innovation Foundation (ITIF) has explained, the European Union's approach to competition policy, with its focus on protecting decentralized market structures rather than consumer welfare, has contributed to a widening productivity gap with jurisdictions such as the United States in technology-driven sectors.[4]Canada should learn from that experience, not repeat it. At the same time, Canada's small domestic market and openness to trade mean that firms must achieve scale and integrate globally to compete. Competition policies that specifically penalize these strategies will leave Canadian firms trapped at subscale, unable to challenge larger rivals.

ITIF's Schumpeter Project on Competition Policy appreciates the opportunity to comment on the Draft Guidelines. ITIF commends the Bureau for its effort to put forward guidelines that further sound and administrable enforcement as well as provide businesses with greater certainty about what conduct will be found unlawful under the Act. At the same time, ITIF believes that there are several areas where the Draft Guidelines could be improved to achieve these ends.

The structure of this comment mirrors the structure of the Draft Guidelines: first, a discussion of the issues surrounding market power; second, an overview of the general ways in which business conduct can harm competition; third, an analysis of the proper standards for assessing anticompetitive effects; fourth, a brief note on the design of proper remedies; fifth, an examination of the Draft Guidelines' treatment of common business conduct including exclusive dealing, tying and bundling, refusals to deal, and self-preferencing; and sixth, the legal tests that should apply to determine violations of the various provisions of the Act. Recommendations and a brief conclusion follow.

2. Market Power

While in general the Draft Guidelines' approach to market power remains largely consistent with previous guidance, alongside traditional factors like market shares, concentration levels, and barriers to entry, the Bureau's framework for assessing market power now includes consideration of statutory factors added by recent amendments, including network effects, data advantages, and "learning by doing."[5]However, there are several areas where the Draft Guidelines could benefit from increased clarity in ways that would further sound competition law enforcement in Canada.

First, the Draft Guidelines should confirm that "the likelihood of entry and expansion" is not just relevant in an indirect case of proving market power based on market definition and market shares, but can also be used to rebut direct evidence of market power.[6]A firm that has demonstrated its ability to exercise market power in the past may not be able to do so in the future due to increased market dynamism and entry, a phenomenon that is particularly important when analyzing fast moving and dynamic high-tech markets.

Second, while it may be technically true that the Bureau "may define markets in a different way than how market participants like businesses or customers think about them," the Draft Guidelines should make clear that what constitutes a relevant market is to be fundamentally determined by demand substitution and cross-price elasticity.[7]As such, the hypothetical monopolist test should be the presumptive way by which the Bureau defines a relevant market, with a more qualitative product-focused analysis applied as a secondary measure when appropriate.

Third, rebuttable presumptions of market power based on market shares, including the 30 percent presumption in the Draft Guidelines, can help promote administrable enforcement and provide businesses with certainty.[8]However, the Draft Guidelines neglect to expressly include any corresponding safe harbor under which firms with market shares less than 30 percent are presumed not to have market power, and thus not only forgo similar benefits in terms of administrability and increased certainty, but risk chilling procompetitive behavior that poses little prospect of anticompetitive harm.

Finally, while it is true that the "effects of the anti-competitive conduct or agreement themselves can be evidence that the firm has market power," inferences should only be drawn from past similar anticompetitive conduct and not any determination that the instant behavior under consideration is anticompetitive.[9]That is, the analyses of whether a firm has market power and whether it has engaged in unlawful conduct should be treated as logically independent inquiries: That a firm has market power and may thus be able to harm competition does not create an inference that it has acted anticompetitively, and even a finding that behavior appears to be anticompetitive does not mean the firm had the power to harm consumers.

3. The Economics of Anti-competitive Conduct and Agreements

The Draft Guidelinesdescribe how various types of conduct and agreements can prevent or lessen competition by raising rivals' costs, foreclosing access to customers or inputs, facilitating coordination, or creating barriers to entry and expansion. The Bureau emphasizes that the competitive merits of business conduct is to be assessed based on its actual or likely competitive effects, and that the same conduct may be analyzed under multiple ACCA provisions. While ITIF strongly agrees with this emphasis on assessing anticompetitive effects, but takes no position here on the overlapping nature of Canada's competition provisions, there are several areas where the Draft Guidelines may lead to unnecessary confusion.

First, the Draft Guidelines' treatment of "reducing incentives to compete" mistakes a general category for a more specific one. Collusion and exclusion are the two ways that business conduct can harm competition through behavior that involves a reduction in competitors' incentives and/or abilities to compete.[10]In the case of collusion, this can take the form of the replacement of incentives to compete with incentives to cooperate in ways that harm consumers; in the case of exclusion, the incentives to compete on the merits are overcome by incentives to harm competitors in ways that also harm consumers. In short, by sufficiently explaining that incentives to compete may be reduced through either collusion or exclusion, there is no need to acknowledge any third and general "reducing incentives to compete" theory of harm.

Next, while ITIF does not object in principle to the Bureau "consider[ing] how all of the conduct or agreements we are investigating can work together to harm competition," it must make clear that these type of "broth" theories of liability are not simply a matter of condemning a firm for engaging in, for example, a pattern of exclusive dealing behavior even if the procompetitive benefits outweigh the anticompetitive harms in each instance.[11]Rather, in these cases the Bureau must show that the conduct, taken collectively, creates an overall anticompetitive effect that outweighs any procompetitive benefits in a way that demonstrates, for purposes of analyzing anticompetitive effects, that the whole is greater than the sum of its individual parts.

4. Assessing the Effects of Anti-competitive Conduct and Agreements

ITIF does not dispute the Bureau's view that when determining whether business conduct is anticompetitive, "[t]he core question is whether the conduct or agreement creates, enhances, or maintains market power."[12]The Draft Guidelines further reasonably identify that a "but-for" causation standard should apply when analyzing whether business conduct has anticompetitive effects.[13]Indeed, and consistent with the well-established literature on the primary importance of innovation to economic growth and the reality of the modern high-tech world, ITIF commends the Bureau for its statement that it will "take special care to promote dynamic competition, since it is critical to improving the lives of Canadians over time."[14]

However, while ITIF understands that the substantial lessening or prevention of competition test for abuse of dominance cases constitutes a higher threshold than the adverse effect on competition test used to adjudicate anticompetitive agreements, the Draft Guidelines should make clear that, as a general matter, abuses of dominance must not only have anticompetitive harms that disproportionately outweigh procompetitive benefits, but also must not, in some way, reflect competition on the merits. That is, unlike when evaluating the harms and benefits of an exclusive dealing agreement-which is prima facie not competition on the merits-an extra demonstration is required to show that unilateral practices do not reflect competition on the merits. This can be done, for example, by showing that unilateral conduct would exclude an equally efficient rival, make no sense but for its anticompetitive effect, involve some type of profit sacrifice, or be accompanied by a specific anticompetitive intent to exclude. Without such a requirement, the Bureau risks turning itself into a de facto regulator by taking license to weigh the harms and benefits of firms' unilateral business decisions to determine whether they are sufficiently maximizing competition and consumer welfare.

5. Remedies

ITIF broadly agrees with the Draft Guidelines that the "goal of a remedy is to fix the competition problem" and that this can be achieved through measures directed a "stopping the conduct or an agreement," "other orders to restore competition" as well as "[a]dministrative monetary penalties" that serve an additional purpose of deterring future anticompetitive behavior.[15]ITIF also concurs with the Draft Guidelines' note that "[w]here appropriate, we may seek an order stopping only part of the anti-competitive conduct or agreement," as in some cases fully stopping the anticompetitive conduct may not enhance consumer welfare and other procompetitive measures to restore competition can instead be implemented.[16]However, in general ITIF believes that the Draft Guidelines could provide more clarity about how the legal standards to obtain a given form of relief should vary with both the nature and purpose of the remedy.

For example, the Draft Guidelines misconstrue the purpose of a structural remedy as one to "restore competition" rather than to stop the behavior that was found to be anticompetitive.[17]That is, whereas relief designed to stop the anticompetitive behavior is backward looking and seeks to put the world back to the way it was had the conduct not occurred, measures that are designed to restore competition seek to create the world that would have existed had the firm instead fully engaged in competition on the merits. To that latter end, structural relief is plainly inappropriate: not only do breakups involve greater risk and uncertainty, but they also stand to unnecessarily undo valuable scale and scope efficiencies, and as such are extremely unlikely to be the least restrictive means of restoring competition.

Moreover, while the Draft Guidelines rightly note that "the choice not to do something can be anti-competitive conduct," such that to stop certain types of anticompetitive conduct like refusals to deal the Bureau may issue "[a]n order stopping such conduct may force the firm to do what they chose not to do," the Draft Guidelines should make clear that, given the greater administrative and competitive difficulties associated with this type of prescriptive relief relative to proscriptive measures, in addition to being procompetitive, there must also be a reasonable relationship between the affirmative obligation and undoing the anticompetitive conduct.[18]Moreover, in the cases where affirmative obligation remedies are utilized to restore competition, the Draft Guidelines should explain that the heightened but-for causation standard must be met so as to avoid situations where, for example, firms are required to share data in the name of restoring competition and undoing the fruits of the violation when many of the data advantages enjoyed by the firm were not the result of the anticompetitive conduct.

Finally, penalties that take the form of 3 percent of worldwide revenues create disproportionate penalties based on global scale, rather than harms to the Canadian market, in ways that may not just raise concerns about due process and extraterritorial overreach, but could even harm Canadian consumers.[19]Consider, for example, two Canadian firms committing identical conduct in the Canadian market that harms the same customers and produces the same competitive effects. Firm A generates $2 billion in annual revenue, all in Canada. Firm B is also Canadian but has expanded globally and now generates $2 billion in Canadian revenue and $100 billion worldwide. Firm A faces a maximum administrative monetary penalty of $60 million (3 percent of $2 billion in total revenue), whereas Firm B faces a maximum administrative monetary penalty of $3 billion (3 percent of $100 billion in worldwide revenue)-50 times higher despite identical Canadian market conduct, identical harm to Canadian consumers, and identical competitive effects. Such a penalty structure risks being abused to punish firms simply because they have achieved global scale and may chill procompetitive behavior by these firms that benefits Canadian consumers given the possibility of significant overdeterrence.

6. Types of Anti-competitive Conduct and Agreements

6.1 Exclusive Dealing

As the Draft Guidelines suggest, "exclusive dealing is not necessarily anti-competitive" and is in fact a common and widely used practice that often yields procompetitive benefits that outweigh any harms.[20]Indeed, exclusive dealing contracts that are either 1 year or less or foreclose a less than 30 percent share of the market are highly unlikely to result in anticompetitive harms. For this reason, the Bureau should consider whether the Draft Guidelines might be revised to provide safe harbors for businesses with greater certainty as to when exclusive dealing agreements are permissible.

Moreover, the Draft Guidelines' treatment of exclusive dealing is overbroad and includes behavior that should be evaluated under a different legal and economic framework. For example, conditional pricing practices that take the form of "volume and loyalty discounts" are more akin to price predation than exclusive dealing, as with these practices increased output does not necessarily mean that rivals' market share is restricted (in contrast with traditional exclusive dealing or discounts conditional on market share). As such, they are more likely to be procompetitive and should only be illegal if price is below cost in a way that would suggest an equally efficient rival is excluded.[21]Similarly, "making products that are not technology compatible with those from other suppliers" or related product design decisions are ubiquitous in the modern digital economy and likewise do not involve any concerted exclusionary restraint on trade.[22]In the rare cases where product design decisions are condemned, they should be determined to be a sham innovation or otherwise not make economic sense but for a demonstrable exclusionary effect.

6.2 Tying and Bundling

A similar analysis applies to the Draft Guidelines treatment of tying and bundling, another common form of procompetitive behavior. While the Draft Guidelines rightly note that "a firm with market power in the market for the tying product may be able to use a tie to exclude competitors in the market for the tied product," they should make clear that they are not sanctioning the use of a modified per se rule that presumes that tying is illegal if there is market power in the tying product market, but instead one that requires proof of anticompetitive effects in the tied (or tying) product markets.[23]And, if a prima facie case is made that an instance of tying is illegal, firms should have the ability to present procompetitive justifications that can be weighed against the anticompetitive harms given that "[t]ying can have various benefits."[24]

In addition, while the Draft Guidelines appropriately distinguish between contractual tying, technological tying, and bundling, they do not confirm that this sort of balancing framework is not, without qualification, the legal standard that should be applied to evaluate all of this behavior. For example, like other product design decisions, technological tying is by its very nature not a concerted or contractual restraint on trade and is much more akin to the widely practiced and almost always procompetitive "self-preferencing," discussed elsewhere in the Draft Guidelines, which can be anticompetitive only if the integration does not make economic sense but for the exclusion. Similarly, the predominant mechanism for exclusion in bundling, which here again is typically procompetitive, is price, not contract, which counsels that this behavior should only be deemed anticompetitive if the price is below an appropriate measure of cost in a way that would exclude an equally efficient rival.

6.3 Refusals to Supply

ITIF has significant concerns with the Draft Guidelines' treatment of refusals to supply. While ITIF does not here dispute the Bureau's position that anticompetitive refusals to deal need not always be actual rather than constructive, the two necessary conditions put forward as to when the Bureau will challenge refusals to deal must not be viewed as jointly sufficient.[25]Specifically, that a product may be competitively significant and difficult to obtain in other ways only evinces that some foreclosure may have occurred-evidence of increased power over price or consumer harm in the form of higher prices, reduced output, or diminished innovation must still be put forward to make out a prima facie case of harm before the burden of production shifts to the defendant to present procompetitive justifications for the refusal to deal.

Moreover, to ensure that competition on the merits is protected, the Draft Guidelines should also explain that refusals to deal are only anticompetitive if there is evidence of profit sacrifice-for example, a firm terminating a prior and presumably profitable course of dealing, or alternatively unreasonably discriminating with respect to the firms with whom it deals. Without such a requirement, competition law risks incentivizing the perverse results of forcing firms to engage in unprofitable behavior and turning the Bureau into a competition regulator by asking it to determine which terms of dealing maximize competition or consumer welfare.

6.4 Self-Preferencing

ITIF also has concerns with respect to the Draft Guidelines' discussion of self-preferencing, which it defines as "a firm giv[ing] favourable treatment to its own products, or other products that advantage it in some way."[26]While the Draft Guidelines are right to point out that self-preferencing is "extremely common" and "can have benefits," the conditions put forward for assessing when self-preferencing is anticompetitive are underinclusive.[27]Specifically, the Guidelines should state plainly that self-preferencing is only anticompetitive if there is both a clear demonstration of consumer harm and that the self-preferencing did not constitute competition on the merits by virtue of not making economic sense but for any exclusionary effect. This latter condition is important because without such a requirement the Bureau would here again be put in an awkward position, now of trying to micromanage a firm's product design decisions commensurate with their perceived effects on consumers-another quasi-regulatory role for which it is ill-suited.

6.5 Input Pre-Emption

The Draft Guidelines are correct that input preemption or predatory overbuying can also constitute anticompetitive conduct under certain limited circumstances where the buying did not make "commercial sense"-properly understood, situations where the firm made unprofitable purchases in a way that would have excluded an equally efficient rival by driving price below cost.[28]However, as in the case of predatory pricing, merely showing that a firm drove prices below costs and foreclosed rivals is not enough: there must also be some demonstration of harm to competition and consumers, such as by requiring proof that the firm would be in a position to recoup the costs of its behavior by increasing prices and harming consumers after driving rivals out. The Draft Guidelines, however, do not appear to view likely recoupment as a requirement of anticompetitive input pre-emption, and thus risk condemning procompetitive behavior that on balance benefits consumers with lower prices-indeed, the ability for a firm to recoup its losses does not even appear to be one of the factors as relevant for assessing claims of input pre-emption.[29]

6.6 Contracts That Reference Rivals

ITIF does not have substantial concerns with the Draft Guidelines' treatment of contracts that reference rivals, such as most favored nation clauses, which while generally procompetitive can harm consumers and competition in certain limited circumstances. The Draft Guidelines could however go further and confirm that these practices will be assessed by requiring clear evidence of consumer harm-which, as the Draft Guidelines state, typically occurs only when firms have market power and the potential foreclosure is substantial-and then allowing firms to present procompetitive justifications for their behavior in a way that allows for determination of the net competitive effect of the agreement.[30]However, here again, ITIF suggests that "reducing incentives or the ability to compete" should not be viewed as a distinct theory of harm from collusion and exclusion.[31]

6.7 Market Restriction

The Draft Guidelines are correct that market restrictions that include vertical intrabrand restraints like exclusive distributor agreements can harm competition under certain circumstances despite having a number of procompetitive justifications, including "preventing free-riding."[32]The Draft Guidelines could however go further and confirm that these practices will be assessed by requiring clear evidence of consumer harm-which, as the Guidelines state, is more likely when distributors request the market restrictions and there is other evidence of downstream collusive behavior-and then allowing firms to present procompetitive justifications in a way that allows the Bureau to determine the net competitive effect of the agreement.[33]

6.8 Price Maintenance

The Draft Guidelines are correct that price maintenance agreements reflect another vertical intrabrand restraint that can harm competition under certain circumstances despite having a number of procompetitive justifications, including by "eliminating inefficiency in competition amongst retailers."[34]The Draft Guidelines could however go further and confirm that these practices will be assessed by requiring clear evidence of consumer harm-which, as the Guidelines state, is typically collusive in nature but can also involve exclusionary effects-and then allowing firms to present procompetitive justifications for their behavior in a way that allows the Bureau to determine the net competitive effect of the agreement.[35]However, as discussed further infra, other forms of price maintenance such as a "refus[al] to supply a product to retailer because of its low pricing policy" should be evaluated using different legal standards, and in particular those applied in cases that involve a refusal to supply.[36]

6.9 Below-Cost Pricing

ITIF has significant concerns with the Draft Guidelines' treatment of below cost or predatory pricing and believes that the Draft Guidelines could go further in helping to ensure that the Act does not chill procompetitive and pro-consumer behavior. First, while ITIF agrees that prices that are above average total cost cannot constitute anticompetitive predation, the Draft Guidelines err in suggesting that prices between average avoidable cost and average total cost can be anticompetitive if it is determined that the firm acted with an anticompetitive intent.[37]This approach, which mirrors the standard in the European Union as formulated by AKZO Chemie BV v Commission and other cases, is flawed because pricing below average total cost could still be cash-flow positive for an equally efficient competitor who finds it rational to compete in the short run rather than deciding to exit the market for fear that it will not recoup its total costs in the long run, and as such can reflect competition on the merits.

Second, although the Draft Guidelines discuss recoupment, they do not clearly state that above cost pricing will only be unlawful if there is additional evidence that the firm is likely to recoup its losses, as distinct from being presumed from the mere fact of above-cost pricing. Without a recoupment requirement, the Act risks condemning pricing that benefits-not harms-consumers with lower prices and lacks any actionable anticompetitive effect.

6.10 Margin Squeezes

The Draft Guidelines are correct that while it is "rare that pricing decisions raise issues under the Act," it is possible for margin squeezes to violate the antitrust laws under very limited circumstances.[38]However, the Bureau merely states that in determining whether an anticompetitive margin squeeze exists, it "may consider if the vertically integrated firm had an obligation to supply competitors to begin with" for purposes of assessing the upstream input foreclosure, when the proper approach would be to confirm that an anticompetitive refusal to deal upstream is a necessary predicate of any margin squeeze claim.[39]The reason is simple: the standards for requiring firms to deal with their competitors should not change simply because the behavior is part of a broader margin squeeze.

Similarly, the Draft Guidelines also neglect to require a demonstration that the applicable standards for predatory pricing have been met when analyzing the downstream element of customer foreclosure associated with a margin squeeze, and instead state that "[w]hen assessing impacts on competition in the downstream market, we may consider whether the margin squeeze creates more market power than the vertically integrated firm already had through control of the input."[40]Yet, here again, the legal standards for condemning a firm for low pricing should be relaxed just because it is part of a margin squeeze-put simply, the whole is equal to the sum of its parts for assessing this type of anticompetitive behavior.

6.11 Excessive and Unfair Pricing

ITIF has expressed extensive and serious concerns with the Bureau's enforcement of excessive and unfair pricing offenses by virtue of the 2024 amendments to the Act.[41]ITIF thus greatly appreciates the Draft Guidelines' statement that "[e]xcessive and unfair pricing is likely to raise issues only if it is used to engage in other types of anti-competitive conduct or agreements," such as a refusal to deal or margin squeeze.[42]As the Draft Guidelines wisely counsel, "simply charging high prices is usually a firm exercising its existing market power" and "not usually conduct that adds to their market power," as well as can "encourage[] the competitive process" in a variety of ways.[43]

6.12 Information Sharing

The Draft Guidelines are correct that sharing competitively sensitive information can harm competition under certain circumstances depending on the type of information shared, its specificity, and time factor.[44]However, the Draft Guidelines should also expressly state that procompetitive justifications can exist even for the sharing of competitively sensitive information, which should be balanced against any harms in weighing the net effect of the information sharing-provided that the information sharing, taken with other evidence, does not rise to the level of inferring a collusive scheme that is per se unlawful.

Moreover, although the Bureau appropriately states that it "may consider any safeguards or limitations that exist for shared information," given the ubiquitous and generally procompetitive nature of information sharing, the Draft Guidelines' should go further and put forward concrete safe harbors for firms to provide them with greater certainty as to when an information sharing agreement will not be challenged. For example, Draft Guidelines could make clear that the sharing of competitively sensitive information will not raise antitrust concerns if the exchange is managed by a third party, the information is at least three months old, aggregated, and anonymized, and that least five firms contributed information, with no one firm's data constituting 25% of any statistic on a weighted basis.

Finally, while noting that "[i]nformation sharing can be unilateral or coordinated," the Draft Guidelines fail to identify the considerable differences that should apply in evaluating the legal merits of information sharing agreements and unilateral facilitating practices like signaling. Whereas the Bureau should weigh the harms and benefits associated with the former, unilateral facilitating practices should only be condemned if they do not constitute competition on the merits and are distinct from normal market behavior, such as a leader-follower dynamic-as evinced by evidence of a clear and unequivocal anticompetitive intent to collude.

6.13 Disciplinary Conduct

In contrast with the generally well established conduct categories discussed in the Draft Guidelines, ITIF is concerned that the discussion surrounding the "disciplinary conduct" offense in Section 78 is not sufficiently precise to rein in the substantial risk of false positives associated with condemning behavior like the introduction of new "fighting brands" or price reductions to compete with a competitor.[45]As such, ITIF recommends that the Bureau reframe the disciplinary conduct offense in a way similar to it's discussion of excessive and unfair pricing, whereby this behavior does not raise competitive concerns unless "it is used to engage in other types of anti-competitive conduct or agreements."[46]For example, finding that a firm engaged in the "use of fighting brands introduced selectively on a temporary basis to discipline or eliminate a competitor" should only be condemned if it constitutes an anticompetitive product design; similarly, "selling articles at a price lower than the acquisition cost for the purpose of disciplining or eliminating a competitor" would be unlawful if it constituted predatory pricing.[47]

6.14 Agreements Between Competitors

The Draft Guidelines are correct to highlight other common forms of competitor collaborations that, while most often procompetitive, may give rise to antitrust issues under certain circumstances by increasing the incentive or ability to engage in collusive or exclusionary behavior.

Commercialization and Joint Selling Agreements

ITIF agrees with the Draft Guidelines that these agreements "can have benefits" but "can also harm the competitive process," and that the harms and benefits of such an agreement should be weighed to determine the net competitive effect of the practice.[48]

Information Sharing Agreements

ITIF agrees with the Draft Guidelines' suggestion that information sharing agreements that may otherwise be unlawful can nonetheless be legal if they are an ancillary restraint that is reasonably necessary for a "procompetitive joint venture to function."[49]

Research and Development Agreements

While ITIF agrees that the harms and benefits of research and development (R&D) agreements, which are generally procompetitive, should be weighed in determining their overall net competitive effect, the Draft Guidelines do not make clear why a reasonable necessity causation test is also applied in this case to determine whether the procompetitive benefits from this arrangement are legitimate. While there are legal standards that include this sort of causation analysis, if this additional burden is to be imposed when assessing research and development agreements, the Draft Guidelines should also identify the greater difficulties associated with proving anticompetitive harm in an R&D market relative to an existing product market, and in particular that structural evidence, such as the number of firms competing in the R&D market, should not be given weight in view of both the generally uncertain nature of the relationship between market structure and innovation as well as the substantial evidence suggesting that increased concentration may actually increase innovation in many cases.

Joint Production Agreements

ITIF agrees with the Draft Guidelines that these agreements, which are generally procompetitive, "can create cost savings" but "can also harm the competitive process," and that the harms and benefits of such an agreement should be weighed to determine the net competitive effect of the practice.[50]

Joint Purchasing Agreements

ITIF agrees with the Draft Guidelines that these agreements "can be pro-competitive" but "can also harm the competitive process," and that the harms and benefits of such an agreement should be weighed to determine the net competitive effect of the practice.[51]

Non-compete Clauses

ITIF agrees with the Draft Guidelines that non-compete agreements between an employer and an employee "can serve legitimate purposes" and that the harms and benefits of such an agreement should be weighed to determine the net competitive effect of the practice.[52]By contrast, non-compete clauses that take the form of "no-poach" agreements between employers not to hire, should, as the Bureau rightly notes, be assessed "under the cartel provisions where they are a market allocation agreement."[53]

6.15 Acquisitions

ITIF agrees with the Draft Guidelines that acquisitions, which generally either pose no risk of harm to consumers or result in offsetting efficiency benefits, "can be procompetitive" but "can also harm the competitive process," and that the harms and benefits of such an acquisition should be weighed to determine its net competitive effect.[54]However, ITIF has considerable concerns with Draft Guidelines' discussion of a serial acquisition offense whereby "each single transaction may not have enough of a competitive effect to raise issues under the merger provisions" but "when examined collectively under the abuse of dominance provisions, they can raise serious competition issues."[55]

First, serial acquisition theories of harm are typically premised on the idea that a given merger is more likely to be anticompetitive if it is part of a pattern of acquisition activity, which is in turn based on the false assumption that growth by mergers is somehow more competitively suspect than growth achieved organically or through contract-mergers, contracts, and unilateral practices all encompass broad swaths of normal and generally procompetitive behavior that in some circumstances can run afoul of the antitrust laws. Similarly, with so-called "entrenchment" theories that involve a conglomerate merger which combines complementary products, transactions are usually condemned precisely because they result in efficiencies like economies of scope that allow a firm to better compete on the merits, which runs counter to the goals of sound merger policy. And, of course, unlike increased tacit collusion created through horizontal mergers, anticompetitive bundling by the merged firm can be addressed if and when it arises.

7. The ACCA Provisions

7.1 Abuse of Dominance (Sections 78 and 79)

ITIF acknowledges the three-part test set out in Section 79 that requires a demonstration of dominance, a practice of anti-competitive acts, and the effect of substantially harming competition, as a workable framework for policing anticompetitive unilateral conduct.

Dominance

ITIF nonetheless has several concerns with the Draft Guidelines' treatment of dominance. First, the Bureau must be careful not to mix the assessment of dominance and the ability for a firm to harm consumers with the further assessment of whether a firm has acted anticompetitively, which is at risk given the Draft Guidelines' statement that the Bureau will consider "market power gained through the anti-competitive conduct or agreement" in a dominance analysis.[56]Second, while ITIF does not oppose in principle the Draft Guidelines' inclusion of a rebuttable presumption of dominance for firms that have a market share of 50 percent or more, the Draft Guidelines lack any corresponding dominance safe harbor, stating only that the Bureau "may stop investigating if we find market shares are lower than these thresholds."[57]But this disparate treatment is unjustified given that the administrability benefits of a dominance safe harbor are arguably far greater than those that flow from a presumption of harm given that the number of firms that would satisfy the latter relative to the former are much fewer.

Joint Dominance

While ITIF understands that Section 79 raises the possibility that "one or more persons" may enjoy a dominant position, ITIF has concerns regarding the Draft Guidelines' statement that "an agreement between the jointly dominant firms" is "not a requirement," which creates substantial uncertainty about what practices will be condemned beyond that which is actionable as an anticompetitive agreement or unilateral abuse of dominance.[58]To ensure that conscious parallelism and other normal leader-follower behavior are not condemned, the Draft Guidelines should make clear that the Bureau will focus its enforcement in the case of joint dominance on invitations to collude and unilateral facilitating practices, like signaling, that are engaged in with a unequivocal anticompetitive intent to collude.

A Practice of Anti-competitive Acts

ITIF broadly agrees with the Bureau that, when assessing unilateral behavior, the focus must be on conduct that is "exclusionary or predatory" or otherwise "intended to have negative effect on a competitor."[59]However, the Draft Guidelines create a substantial lack of clarity with respect to their discussion of the three types of evidence they will consider-"subjective evidence of intent," "reasonably foreseeable consequences of the conduct or agreement," and "pro-competitive or efficiency enhancing justifications for the conduct or agreement"-which is at the same time both over and underinclusive.[60]

First, an assessment of reasonably foreseeable consequences goes much more directly to the question of whether conduct has the effect of harming competition, rather than the assessment of purpose. Moreover, analyzing a firm's subjective intent, while important to evaluate certain behavior-such as unilateral collusive conduct-is but one test that the Bureau may apply to analyze whether a practice constitutes competition on the merits. For example, in analyzing price-based conduct like predatory pricing, the Bureau may find it appropriate to inquire whether the conduct would exclude an equally efficient competitor; by contrast, for design decisions, the Bureau may require some showing that the design did not make economic sense but for an exclusionary effect; and, for refusals to deal, the Bureau may require some demonstration of profit sacrifice.

When assessing whether a firm has a legitimate procompetitive justification for its behavior that the Bureau "may weigh against evidence of an anti-competitive purpose," the Draft Guidelines again overlook that a weighing of anticompetitive harms and benefits is more properly situated in an analysis of anticompetitive effects, not intent-some desire to beat a rival and achieve better efficiency are common and non-exclusive rationales for business practices.[61]Furthermore, the Draft Guidelines appear to put undue restrictions on the sorts of justifications that the Bureau will consider. Most importantly, the Bureau states that it will look to see whether "there were other ways the firm could have achieved the goals of the justification that would have led to fewer competitive problems."[62]As noted supra in connection with the assessment of R&D agreements, although this sort of causation requirement is not uncommonly applied in some jurisdictions when evaluating anticompetitive agreements, it is not appropriate in the context of unilateral abuses: Competition law should not condemn unilateral behavior that is otherwise competition on the merits simply because there may have been some other tactic that could have resulted in greater competitive benefits, as doing so would create a highly unadministrable regime that risks turning the Act into a form of regulation rather than law enforcement.

Conduct That Has the Effect of Harming Competition

ITIF strongly agrees that dominant firm conduct that is prima facie exclusionary or predatory should only be condemned if there is some harmful effect on consumer welfare and competition and that this should be assessed by weighing the competitive harms against the demonstrated benefits that drive superior competitive performance. Moreover, ITIF supports the Draft Guidelines' view that there must be a substantial lessening of competition for an abuse of dominance to be found, as opposed to a mere adverse effect on competition, which helps to ensure that procompetitive behavior will not be chilled through over-enforcement in the area of unilateral conduct.

7.2 Anti-Competitive Agreements (Section 90.1)

ITIF acknowledges the test set out in Section 90.1, which requires either an agreement between competitors or an agreement that has the purpose of lessening competition, as well as substantial negative effects on competition, as a workable framework for policing anticompetitive agreements.[63]While excessive formalism in antitrust law can lead to adverse outcomes, distinguishing between horizontal and non-horizontal agreements has a well-established economic and legal basis, and the Bureau is right not to presume that non-horizontal agreements constitute a prima facie restraint of trade-although it is also true that horizontal agreements do not invariably involve some intentional restriction of competition.

ITIF also appreciates the Draft Guidelines' confirmation that "[c]onscious parallelism alone between competitors does not create an agreement that we can act on under section 90.1," which ensures that the agreement requirement will not yield an undue number of false positives.[64]The Draft Guidelines' statement that structural evidence will not be used to presume harm to competition is also prudent. [65]For example, even in an exclusive dealing case where a substantial share of the market is foreclosed, an examination of other pertinent facts, such as the duration of the agreement, may make the risk of competitive harm highly unlikely.

7.3 Refusal to Deal (Section 75)

Consistent with the discussion of this behavior supra, ITIF does not believe the standard in Section 75 for determining whether a refusal to deal is anticompetitive is inherently inconsistent with sound enforcement provided that the relevant factors analyzed by the Draft Guidelines are understood as follows: whereas factors (a) and (b) can be used to demonstrate sufficient foreclosure, (c) and (d) can evince whether profit sacrifice has occurred, and (e) requires some proof of ultimate consumer harm. However, the Draft Guidelines should also make clear that, even if these elements are shown, firms should still have the ability to present procompetitive justifications for their refusal to deal, which would ultimately be weighed against any anticompetitive effects.

7.4 Price Maintenance (Section 76)

Consistent with the discussion of this behavior supra, ITIF does not believe the standard in Section 76 for determining whether price maintenance is anticompetitive is inherently inconsistent with sound enforcement given the Draft Guidelines' statement that "[f]or section 76 to apply, the price maintenance must have the effect of harming competition" as well as an evaluation of whether "the price maintenance conduct creates, enhances, or maintains market power"-in other words, per se rules will not apply.[66]However, with respect to the relevant legal tests put forward, the Draft Guidelines fail to sufficiently distinguish between price maintenance achieved through an agreement, as in Subparagraph 76(1)(a)(i), as distinct from unilateral vertical refusals to deal in Subparagraph 76(1)(a)(ii), and the de facto group boycott behavior referenced in Subsection 76(8).

Price maintenance practices that take the form of "minimum resale prices, Manufacturers Suggested Resale Prices (MSRP), and Minimum Advertised Prices (MAP)" should only be condemned if there are anticompetitive harms that outweigh procompetitive benefits.[67]However, conduct where "a supplier refuses to supply their product to a firm because the firm's low pricing policy or otherwise discriminates against them" should only be condemned where there is a determination that the applicable standards for finding an anticompetitive refusal to deal were met, which will only occur in very rare circumstances in a purely vertical case where the refusal to deal does not involve a competitor-regardless of the intent of the supplier in refusing to deal.[68]By contrast, conduct by retailers that involves getting a supplier not to do business with another retailer, as identified in Subsection 76(8), can very much reflect anticompetitive behavior, although ITIF does not object the legal test set forth in 76(8) requiring a demonstration that this conduct have anticompetitive effects before it is proscribed.

7.5 Exclusive Dealing, Tied Selling, and Market Restriction (Section 77)

Consistent with the discussion of this behavior supra, ITIF does not believe the standards in Section 77 for determining whether exclusive dealing, tied selling and market restrictions are anticompetitive are inherently inconsistent with sound enforcement. However, the recognition that, unlike in the case of vertical interbrand restraints like tying and exclusive dealing, market restrictions do "not require a likely exclusionary effect to be shown," is problematic by virtue of encompassing vertical intrabrand restraints like exclusive distributor agreements and not just horizontal market restrictions that while anticompetitive may not rise to the level of a cartel.[69]As noted supra, vertical intrabrand restraints like exclusive distributor agreements are not by nature anticompetitive and can result in procompetitive effects that outweigh any anticompetitive harms. Indeed, the Draft Guidelines also fail to sufficiently make clear that, in the case of tying and exclusive dealing, firms should similarly be able to present procompetitive justifications for this behavior that would be balanced against any anticompetitive harms to determine the overall net effect on competition.

In addition, in several respects the Draft Guidelines' definitions of exclusive dealing, tied selling and market restrictions are overbroad in view of the legal standards that are being applied. For example, with respect to exclusive dealing, the Draft Guidelines state that the rule "also applies to situations where a supplier induces exclusivity by offering to supply on more favourable terms or conditions if a customer agrees to deal mainly or only with them," which overlooks that conduct like price-based loyalty discounts should only be condemned if they are below cost in a way that would exclude an equally efficient rival and offend competition on the merits.[70]The Draft Guidelines are similarly overbroad in applying the legal standard for tying to bundling, which should require-as the Draft Guidelines themselves suggest-some additional showing that the behavior does not reflect competition on the merits, either because it "makes no economic sense" or excludes an equally efficient rival.[71]Finally, with respect to market restrictions, the Draft Guidelines fail to sufficiently distinguish between horizontal market restrictions that are anticompetitive and do not require a demonstration of anticompetitive effects from vertical restraints that include cases where "the manufacturer of a product allows its distributors to only sell to retailers in a defined geographic area," which as noted supra should not be presumptively anticompetitive and can admit of procompetitive justifications.[72]

Recommendations

For these reasons, ITIF has concerns with the Draft Guidelines and respectfully offers the following recommendations for the Bureau to consider:

▪ Properly distinguish between different anticompetitive behaviors: In several respects the Draft Guidelines lump together under a common category (e.g., exclusive dealing) practices that are widely different from a legal and economic perspective (e.g., contractual exclusive dealing, bundling, predatory design). Revised Guidelines should distinguish between practices that vary commensurate with the predominant mechanism they use to exclude (e.g., contract, price, design), as well as whether they are concerted or purely unilateral in nature.

▪ Competition on the merits must be protected:In assessing abuses of dominance, the Draft Guidelines have an opportunity to more clearly confirm that "a practice of anti-competitive acts" only exists if business conduct departs from competition on the merits, either through excluding an equally efficient rival, profit sacrifice, not making economic sense, or a specific intent to exclude, depending on the type of abuse being that is evaluated.

▪ Procompetitive justifications should be considered when assessing effects: When determining whether an agreement has an adverse effect on competition, or whether a dominant firm's unilateral practice substantially lessens competition, Draft Guidelines should affirm that firms are able to generally provide procompetitive justifications for their behavior that would be weighed against any anticompetitive harms in determining the net competitive effect of the behavior.

Conclusion

ITIF commends the Bureau for issuing careful and comprehensive enforcement guidelines to facilitate sound competition enforcement and provide certainty for Canadian businesses in view of the recent changes to Canada's competition laws. However, while the Draft Guidelines generally and correctly focus on condemning only behavior that results in anticompetitive effects, in several specific respects they could be fine-tuned to provide for greater administrability and better limit false positives so as to ensure that innovation and competition flourish in Canada.

Endnotes