On July 29, 2020, the Competition Bureau published a draft revised version of its Competitor Collaboration Guidelines, inviting interested parties to provide their comments no later than September 28, 2020. The revised draft represents the first update to the Competitor Collaboration Guidelines – one of only a few foundational guidance documents in the Canadian competition law framework – since their initial publication in 2009.

The most noteworthy thing about the revised draft may be the modest nature of the changes; clearly, the Bureau remains confident that its guidance in this area is working well. However, there are a few additions and deletions that are noteworthy as potential thematic indications of the Bureau’s evolving approach to enforcement in this area, including with respect to mergers incorporating anticompetitive competitor agreements, hub-and-spoke cartels, AI-assisted parallelism, remedial preferences in respect of civil agreements and the scope of potential competition.

New Guidelines Signal No Major Change in Approach

The Competitor Collaboration Guidelines were initially published in 2009 as part of a major overhaul to the Competition Act that, among other things, created a new two-track approach to competitor agreements that provided two enforcement options: a “full rule of reason” civil framework that could lead to remedial orders where the requisite anti-competitive effect is proven, and a new per se criminal provision applying to “hard core” cartels, potentially punishable by imprisonment and significant financial liability.

At the time of the amendments, there was widespread concern from the competition bar and the business community that the new per se criminal framework could, without being limited to conduct that prevented or lessened competition “unduly” (which was the old test), be unfairly harsh, penalizing normal business practices and chilling pro-competitive collaboration, while others expressed the view that the boundaries of the civil track were too vague, or that the two-track system could lead to confusion, unfairness and delay.

Eleven years on, however, such criticism has largely quieted, with most agreeing that the system works fairly well, or at least has not resulted in a material number of legitimate competitor collaborations with pro-competitive rationales being pursued on a criminal basis. On the contrary, some have argued that the civil provisions may be subject to a degree of under-enforcement, as there have been only a very few cases brought by the Bureau since the 2009 amendments. It had been anticipated that the new, flexible civil provisions would serve a useful role in testing and adjudicating collaborations having a mix of pro-competitive and anti-competitive features; this has not happened.

Overall, while the Bureau has been willing to call for legislative change with respect to other areas of the Act, they appear largely to be satisfied with the competitor collaboration provisions. Consistent with the smooth functioning of the Act in this area, the new draft guidelines do not contemplate a major change in enforcement approach. Instead, the vast majority of the proposed changes are purely cosmetic in nature, such as necessary updates to references and statutory language or minor modifications to clarify wording. This suggests that the Bureau believes that its guidance struck the right balance in the first place and is unlikely to make dramatic changes for the foreseeable future.

Some Areas Updated to Reflect Recent Bureau Experiences

Nevertheless, the revised guidelines do reflect some interesting changes, including updates that almost certainly reflect the Bureau’s approach in certain specific cases, which will be immediately recognizable to those in the bar.

Cartel enforcement in merger cases

For example, the guidelines now make clear that although mergers will generally be assessed under section 92 of the Act rather than the civil or criminal competitor collaboration provisions, “Where parties enter into any agreement(s) that goes beyond the acquisition, amalgamation or combination agreement, whether within or outside said agreement, the Bureau will consider under which provision(s) of the Act any investigation or inquiry should be pursued.” This appears to be directed at the Bureau’s ongoing investigation into a 2017 transaction between Torstar and Postmedia involving the reciprocal transfer of 41 local newspapers, 36 of which were then closed, which was commenced under the merger provisions of the Act and continues as a criminal conspiracy investigation.

Scope for hub-and-spoke conspiracies

Similarly, in the discussion of the criminal conspiracy provision, the general statement that “Where an agreement involves competing and non-competing parties, the fact that some parties are not competitors does not insulate the competing parties from prosecution under section 45” is now accompanied by the more specific statement that “a wholesaler who facilitates a price-fixing conspiracy among its retail clients may be a party to the conspiracy even if it does not compete in the retail market.” This new language seems to clearly contemplate the Bureau’s ongoing investigation into potential price-fixing in respect of bread products, in which suppliers are alleged to have facilitated a “hub-and-spoke” agreement between competing retailers.

Other new language explaining that “Parallel conduct coupled with facilitating practices, such as sharing competitively sensitive information or activities that assist competitors in monitoring one another's prices, may be sufficient to prove that an agreement was concluded between the parties”, may also be motivated in part by the bread case. A new hypothetical example included in the revised guidelines also appears to be directed at this type of agreement.

Non-compete clauses and customer cartels

It is also noteworthy that the draft guidelines include a new warning in the discussion of non-compete clauses to the effect that, “in rare instances”, such agreements may constitute market allocation agreements that could be subject to scrutiny under the criminal provisions. As a general matter, the Bureau has taken the view that “customer cartels” (i.e., agreements to coordinate in the purchase or procurement of goods and services) fall outside of the criminal provision, which appears sound based on the actual statutory text of the criminal provision, and the draft guidelines do not clearly signal a change to this approach. Given the potential significance of this issue, however, it would be useful for the Bureau to provide additional guidance as to when a non-compete clause (or other customer cartel) might constitute a market allocation agreement.

Looking Ahead to Algorithmic Collusion

Some of the proposed modifications to the guidelines also reflect the Bureau’s enforcement priorities on a forward-looking basis. In particular, a new footnote states that “Canada’s cartel provisions are per se and prohibit agreements between competitors and potential competitors to, among other things, fix or control prices – the use of algorithms (Artificial Intelligence or “AI”) could form the basis of a cartel offence, however, the existence of an agreement—actual or tacit—to fix or control prices is necessary, and conduct that amounts to conscious parallelism is not sufficient.”

However, as another new passage clarifies, “Parallel conduct coupled with facilitating practices, such as sharing competitively sensitive information or activities that assist competitors in monitoring one another's prices, may be sufficient to prove that an agreement was concluded between the parties.” While the Bureau has not yet publicly announced a case involving an algorithmic theory of price-fixing, its disclosure and public comment clearly signals an interest in this area.

Potential Change for Potential Competitors

Finally, the new guidelines appear to envision a possibly broader approach to the identification of potential competitors, and thus a wider scope for the identification of potentially improper agreements. In particular, a new section explains how the Bureau will look to “business and strategic plans prepared in the ordinary course of business, marketing and communications with potential customers, and evidence of actual competition for similar customers in neighbouring regions or in respect of similar products”, including “records prepared in the ordinary course of business that consider who the parties perceive to be their competitors and the extent to which the parties monitor and/or respond to the competitive behaviour of each other.”

Another modification replaces a statement that the civil agreements provision could apply to parties “who are potential competitors with respect to the products that are the subject of the agreement” with the broader reference to potential competitors “with respect to a product(s)”. This broader view is consistent with recent practice and serves as a reminder as to the potential scope of the competitor collaboration provisions.


As noted at the outset, the most remarkable thing about the revised guidelines is how few changes the Bureau has proposed. However, as such guidance documents provide valuable insights into potential shifts in the Bureau’s approach, they are often worth reading closely. We will be watching closely to see whether any further changes are made in response to comments received by the Bureau before the new guidelines are finalized.