Chapter 11: Big Data and Competition / Anti-Trust Law
Big Data Law in Canada
Chapter 11:
Big Data and Competition / Anti-Trust Law
Chetan Phull · December 12, 2019
Chapter 11 is provided below. See also our service offering related to this chapter:
“Competition / Anti-Trust Issues”.
Special thanks to Idan Levy for for his valuable legal research and editorial work in the preparation of this book.
Tip: Read this book in Fullscreen mode.
Full Book Online
Big Data Law in Canada may be read online in full.
Chapter-by-Chapter
Big Data Law in Canada may be read chapter-by-chapter.
HTML VERSION
Chapter 11:
Big Data and Competition / Anti-Trust Law
Big Data and Competition / Anti-Trust Law
I. Overview of Legal Risks in Big Data Markets
Data is a core economic asset. It can be leveraged to gain crucial business insights, thereby creating a competitive market advantage. If your business has a vested interest in the Canadian data marketplace, competition-related risks are important to understand.
[See OECD, Supporting Investment in Knowledge Capital, Growth and Innovation (Paris: OECD Publishing, 2013) at 319; Government of Canada, “Competition Bureau statement regarding its investigation into alleged anti-competitive conduct by Google” re Google Adwords API restrictions (Apr 19, 2016): “The collection, analysis and use of data is increasingly becoming an important source of competitive advantage in the digital economy, driving innovation and product improvement”; OPC, “Privacy Law Reform - A Pathway to Respecting Rights and Restoring Trust in Government and the Digital Economy” (Dec 10, 2019) at 21: “the stockpiling of personal information is increasingly seen as a competitive advantage.” See also the discussion of training data in the context of AI in subsection I of Chapter 10.]
The business of big data is “cut-throat,” partly because data ownership is not always clear. For example, consider datasets that could indirectly identify a person. For these datasets, a company’s right to store, analyze, process, sell, and license out such data is tenuous. The data source is usually free to stop contributing data and transition to a competitor’s platform. Moreover, as discussed in Chapter 7, the data source may also be entitled to a right of erasure, with respect to any data that identifies them personally.
It is consequently tempting for a big data business to grow or maintain market share through anti-competitive acts. Regulators, meanwhile, are increasingly wary of anti-competitive agreements and abuse of market dominance. As discussed below, these and other competition law infringements can create exposure in the hundreds of millions of dollars. For smaller players, this could mean getting shut down just as the business begins to skyrocket.
[See Government of Canada, “Competition Bureau sues to shut down business directory scam” (Jul 28, 2011), re several companies getting shut down within their second year of operation.]
Even lawful and common business conduct is subject to the competition regulator’s review, and can spark competition-related exposure. At the very least, a big data company may be required to attend a lengthy deposition, simply because six persons complained about competition infractions. Indeed, in the big data context, the Competition Tribunal has previously stated that evidence in the form of witness statements and testimonies may be necessary because of a “greater need for … qualitative evidence in innovation cases.” An attempt to obstruct such deposition could, in a worst case, result in a fine and/or imprisonment for up to 10 years.
[See Nikiforos Iatrou, ed., Litigating Competition Law in Canada (Toronto: LexisNexis, 2018) at §6.5; The Commissioner of Competition v. Direct Energy Marketing Limited, 2015 CACT 2 at para. 27 (CanLII); Competition Act, ss.9-11, 64(1)-(2); The Commissioner of Competition v. The Toronto Real Estate Board, 2016 Comp. Trib. 7 (“TREB case (upheld tribunal decision)”) at para. 471, aff’d 2017 FCA 236 at para. 3, leave to appeal to the SCC denied (Aug 23, 2018).]
Moreover, the anti-competitive acts of directors and officers may trigger personal liability, or at least give rise to negative press that could damage professional reputations. Such liability and/or bad press can arise from various sources, including a whistleblower or a dissenting insider.
[See Competition Act, ss.49(1), 52.1(8), 53(5), 65(4), 66.1-66.2.]
Foreign parties can also be subject to Canadian competition law. In fact, there is a longstanding statutory mechanism for international competition law enforcement. More recently, an international competition law framework has also been created. International enforcement efforts are likely to increase when a foreign party is charged with a criminal offence for ongoing behaviour, for example, with respect to “hardcore cartels” or misleading public representations.
[See Competition Act, Part III; Government of Canada, “Competition Bureau statement regarding its investigation into alleged anti-competitive conduct by Google” re Bureau’s investigation through the FTC (U.S.) and European Commission; Government of Canada, “Competition Bureau sues to shut down business directory scam” (Jul 28, 2011) re example of international cooperation; Nikiforos Iatrou, ed., Litigating Competition Law in Canada (Toronto: LexisNexis, 2018) at §§4.47-4.55, 6.3-6.4; Chetan Phull, “Cross-Border Crypto Related Litigation” (Smartblock Law, Apr 26, 2018) re conflict of laws overview; Bureau, “International community creates new framework for fair and effective competition law enforcement” (May 1, 2019); Competition Act, ss.46(1), 52(2.1).]
On the basis of the foregoing, it is no wonder that the Competition Bureau of Canada has warned: “[i]n the context of big data, market players should remain vigilant to ensure that the use of new technology, including algorithms, does not result in anti-competitive conduct.”
[See Bureau, “Big data and Innovation: Implications for competition policy in Canada” (Sep 18, 2017) at 34.]
To fully appreciate and heed this warning, an understanding of anti-competitive conduct in the context of big data is required. Such understanding begins with the fundamentals of Canadian competition law and enforcement.
II. Competition Law Fundamentals: Administration and Enforcement
Competition law is focused on eliminating activities that reduce competition in the marketplace. The American counterpart is known as “antitrust” or “anti-trust” law.
[See Competition Act, s.1.1; G.M. v. City National Leasing, [1989] 1 S.C.R. 641 at 676, 1989 CanLII 133 (SCC); U.S. Federal Trade Commission, “The Antitrust Laws” (as of Dec 3, 2019).]
This area of law in Canada is administered and enforced by the Commissioner of Competition (the “Commissioner”). The Commissioner leads the Competition Bureau of Canada (the “Bureau”) for this purpose. Further to his or her enforcement powers, the Commissioner will pursue litigation through the Competition Tribunal (“Tribunal”) or the courts, if negotiated settlement is not possible.
[See: Competition Act, s.7(1); Government of Canada, What is the Competition Bureau (Sep 25, 2019); Government of Canada, Competition and Compliance Framework (Nov 10, 2015), s.3.1.2.]
The competition law infringements discussed herein may be heard by the courts, or the Tribunal, depending on the severity of the infringement. For example, criminal proceedings will be heard by a court. Less severe infringements or reference proceedings with respect to abuse of a dominant market position, anti-competitive agreements, deceptive marketing practices and other matters, may be heard by the Tribunal or a court. The Tribunal is also an option for private parties disputing a competition law matter.
[See: Competition Act, references to “court” in Part VI, Part VII and Part VIII; inclusion of “Tribunal” within the definition of “court” in Part VII.1 and Part IX, per ss.74.09, 123.1(4); reference proceeding before the Tribunal, as commenced by agreement or unilaterally by the Commissioner, per s.124.2. See also the discussion of statutory versus inherent jurisdiction in Antonio Di Domenico, Competition Enforcement and Litigation in Canada (Toronto: Emond, 2019) at 7; Competition Act, s.124.2(3) re reference proceeding for private actions.]
Competition law works to ensure that small- and medium-sized businesses, both domestic and foreign, have an equitable opportunity to participate in their chosen market. This means that enforcement will be pursued when market advantage is gained through means other than investment, ingenuity, and competitive performance.
[See Competition Act, s.1.1; Bureau, “Big data and Innovation: Implications for competition policy in Canada” (Sep 18, 2017) at 3.]
With regard to data-driven markets, the Bureau does not regulate prices, profits, market shares, or the amount of data collected. The Bureau’s work focuses on “market forces (as opposed to market outcomes), thereby ensuring that companies compete on the merits, customers are able to make well-informed decisions, and regulations are minimally intrusive.”
[See Bureau, “Big data and Innovation: Implications for competition policy in Canada” (Sep 18, 2017) at 2, 9.]
There are presently 25 distinct criminal offences under the Competition Act, arising from “cartel agreements, arrangements and conspiracies, false and misleading representations and deceptive marketing, multi-level marketing and pyramid selling, obstruction, failure to notify the Commission of a notifiable transaction, and the contravention of orders.” Criminal liability can also arise from non-active participation in these acts.
[See Antonio Di Domenico, Competition Enforcement and Litigation in Canada (Toronto: Emond, 2019) at 103-104.]
It should be noted that, even if the criminal law standard of “proof beyond a reasonable doubt” is not met, related private claims may still give rise to civil damages “on the balance of probabilities”.
[See Competition Act, ss.36(1), 62.]
In such cases, a torrent of class actions should be expected within two years of the criminal proceeding’s end. The private claims would closely resemble many of the criminal claims, and if a criminal finding was previously made, civil damages may automatically follow.
[See Competition Act, s.36(2), (4).]
We pause here to note that, “potential anti-competitive conduct in any global industry seems inexorably to trigger an avalanche of competition class actions in multiple jurisdictions around the world….” On this basis, exposure in such cases “can quickly run into the tens or hundreds of millions.” It is therefore unsurprising that such class actions are “flourishing in Canada.”
[See Nikiforos Iatrou, ed., Litigating Competition Law in Canada (Toronto: LexisNexis, 2018) at §§1.54-1.56, 6.2-6.5. See also EC, “Antitrust: Commission fines Google €1.49 billion for abusive practices in online advertising” (press release, Mar 19, 2019): “The new EU Antitrust Damages Directive makes it easier for victims of anti-competitive practices to obtain damages.”]
There are also numerous civil regulatory offences under the Competition Act. They include entering into an agreement with a competitor, where such agreement is likely to suppress competition substantially. Note that a legitimate business purpose will be overshadowed by the anti-competitive effect of such agreement.
[See Competition Act, s.90.1.]
Other civil regulatory offences arise from mergers, agreements between financial institutions, vertical agreements between suppliers and customers, bid-rigging, and abuse of a dominant market position.
[See Competition Act, ss.92, 49(1), 76-77, 47, 78-79.]
Most of these offences are among the 24 “reviewable practices” provided in the Competition Act. Additional reviewable practices include:
- refusal to supply a product to a particular purchaser; and
- a manufacturer’s control over the sale price of a product in the hands of other parties in the chain of commerce.
[See Competition Act, ss.75-77; Antonio Di Domenico, Competition Enforcement and Litigation in Canada (Toronto: Emond, 2019) at 351.]
III. Competition and Big Data
(a) Focusing on Recent Big Data Competition Cases
This chapter does not cover the entire framework of competition law in Canada. Nor does it exhaustively survey common competition law issues between jurisdictions. The focus here is on the major competition issues arising from big data practices outside of the M&A context.
To clarify, the history of general data cases in competition law is actually quite rich. Although most of these cases have not dealt with data as stored or transferred according to modern standards, they are still instructive for modern day analysis. Their detailed consideration, however, would needlessly expand our review beyond the key concepts most often arising in digital market competition law.
[See Bureau, “Big data and Innovation: Implications for competition policy in Canada” (Sep 18, 2017) at 6; Bureau, “Big data and innovation: key themes for competition policy in Canada” (Feb 19, 2018) at 4-5.]
With this understanding, we begin with the following starting point: the Bureau understands that modern, large datasets can pertain to individuals, internal performance of an organization, competitor particulars and strategy, and the market environment. The Bureau has also recognized that data may be either a product for direct sale, or an input into a product or service for sale.
[See Bureau, “Big data and Innovation: Implications for competition policy in Canada” (Sep 18, 2017) at 5-6; Bureau, “Big data and innovation: key themes for competition policy in Canada” (Feb 19, 2018) at 6-7.]
For example, Facebook values data as a product for direct sale to third-party advertisers. Google values browsing history data for the same purpose. Meanwhile, Uber values data as an input into its service, for predictive analysis of demand for rideshare services at any given time and place. Google values web search data the same way, for predictive analysis of demand for search terms.
The recent cases in big data have repeatedly involved the following specific issues, which now will be discussed in turn:
- abuse of a dominant market position;
- agreements that suppress competition; and
- deceptive marketing practices.
(b) Abuse of a Dominant Market Position
i. Overview
This violation “is of particular importance for the protection of consumers, new entrants and, in particular, the small business community.”
[See Canada v. D & B Companies of Canada Ltd., [1995] C.C.T.D. No. 20 (“Neilen case”) at 5.]
The test for abusing market dominance involves an assessment of market dominance, an identifiable practice of anti-competitive acts, and an anti-competitive effect of that practice in a market. These factors cannot be conflated.
[See Competition Act, s.79(1); Canada (Commissioner of Competition) v. Canada Pipe Co., 2006 FCA 233 (“Canada Pipe case”) at paras. 27-28, 92, leave to appeal to SCC refused, 31637 (Jun 1, 2017); TREB case (upheld tribunal decision) at paras. 41-45.]
Various sub-issues must be considered, and those sub-issues may further break down into numerous sub-sub-issues, some of which will now be discussed.
ii. Liability Issues and Cases
An assessment of market dominance requires analysis of the particular “market” at issue, and market power. The former may be further considered with respect to geographical and definitional breadth of the relevant market, and the latter may be further considered with respect to product substitutability.
[See Neilen case; Canada v. Tele-Direct, [1997] C.C.T.D. No 8 (“Tele-Direct case”); Canada Pipe case;TREB case (upheld tribunal decision) at paras. 117-132.]
In the big data context, the “market” will require certain special consideration. For example, the Bureau has recognized that collection of “free” data, which is subsequently used “as an input into other services such as the sale of advertising,” involves competition “on dimensions other than price, such as quality.” Moreover, “quality” in the big data context may require consideration of non-quantifiable factors, like privacy. Note that privacy as an enhancer of quality will arguably require a demonstrable track record with respect to sound privacy practices.
[See Bureau, “Big data and Innovation: Implications for competition policy in Canada” (Sep 18, 2017) at 12-14, 23; Bureau, “Big data and innovation: key themes for competition policy in Canada” (Feb 19, 2018) at 8-9 and fn 11 referencing the DuckDuckGo search engine; TREB case (upheld tribunal decision) at paras. 25, 406.]
With respect to big data market power, the Bureau has acknowledged that novel issues arise from online platforms. Namely, that a high price on one side of a platform may “subsidize investments to attract users and their data to the platform….”
[See Bureau, “Big data and Innovation: Implications for competition policy in Canada” (Sep 18, 2017) at 14.]
Moreover, “market power” requires the existence of barriers to enter the market for competitors. The test also requires the power to engage in exclusionary behaviour, such as preventing rivals from introducing products to the market, with a corresponding incentive to operate such market power.
[See TREB case (upheld tribunal decision) at paras. 165, 173-176, 181, 187-90, 194-95.]
Furthermore, anti-competitive acts may involve consideration of anti-competitive “squeezing” (e.g. forcing retailers to deal exclusively, or forcing manufacturers into long-term contracts), selling under acquisition cost, tied selling, agency through consultants, etc.
[See s.78(1); Tele-Direct case; Canada Pipe case at paras. 62-92.]
The practice of anti-competitive acts must be directed to one or more specific competitors (as opposed to competition in general), without regard to impact. The focus here is on intention, inclusive of whether the acts were motivated by legitimate business reasons.
[See TREB case (upheld tribunal decision) at paras. 270, 275-284, 294, 316-318.]
Substantial suppression of competition is a question of degree and duration. The standard for causation will differ from case to case.
[See TREB case (upheld tribunal decision) at paras. 456-459, 483.]
Most of the above factors arose in the Google Adwords case, before numerous competition authorities. In 2012-13, the U.S. Federal Trade Commission (“FTC”) found that Google’s contractual conditions “made it more difficult for an advertiser to simultaneously manage a campaign on [the Google AdWords platform] and on competing ad platforms….” To satisfy the FTC’s concerns, Google agreed to implement an “opt out” option and change its Adwords API contractual conditions in the U.S.
[See FTC, “Google Agrees to Change Its Business Practices to Resolve FTC Competition Concerns In the Markets for Devices Like Smart Phones, Games and Tablets, and in Online Search” (Jan 3, 2013).]
However, the contractual conditions were not changed in Canada. As expected, three years later, the Bureau in Canada also found that Google inhibited the ability of customers to advertise on competing search engines. It was found that Google’s intent was clearly to exclude rival search engines from the market. In response to the Bureau’s concerns, Google committed to excluding such clauses from its API contracts—this time in Canada—for a period of five years.
[See Government of Canada, “Competition Bureau statement regarding its investigation into alleged anti-competitive conduct by Google” re Google Adwords API restrictions (Apr 19, 2016).]
After another three years, the same competition law conclusions were reached by the European Commission, which found that: “Google's rivals were not able to compete on the merits, either because there was an outright prohibition for them to appear on publisher websites or because Google reserved for itself by far the most valuable commercial space on those websites, while at the same time controlling how rival search adverts could appear.”
[See EC, “Antitrust: Commission fines Google €1.49 billion for abusive practices in online advertising” (press release, Mar 20, 2019).]
To combat data-opolistic practices as exemplified above, a Canadian Parliamentary committee recommended, toward the end of 2018, that PIPEDA be amended to add “principles of data portability and system interoperability.”
[See ETHI Committee Report, “Democracy Under Threat: Risks and Solutions in the Era of Disinformation and Data Monopoly” (Dec 2018) at 58, Recommendations 11.]
For future cases of data market dominance, counsel should consider the Bureau’s more recent enforcement guidelines regarding abuse of market dominance, in addition to past jurisprudence.
[See Bureau, “Abuse of Dominance Enforcement Guidelines” (Mar 7, 2019), particularly at 43-45 re “exclusive dealing” and panopticon data aggregators, and at 51-53 re data aggregation and “disciplinary conduct”.]
iii. Penalties / Remedies
In Canada, a finding of abused market dominance can result in an order:
- prohibiting the anti-competitive practice;
- divesting assets or shares as necessary to overcome the anti-competitive market effects; and
- issuing a fine of up to $10 million ($15 million for subsequent orders).
[See Competition Act, ss.79(2), (3.1).]
When the abuse is not outright and manifests as a more tacit form of market restriction, an order may issue to remove the restriction, and restore or stimulate competition in the relevant product. A consent agreement (i.e. a voluntary agreement to comply moving forward) is also possible.
[See Competition Act, ss.77(3), 105.]
Fines for abusing market dominance can also arise in foreign jurisdictions. For example, in the last couple of years, the European Commission has fined Google:
- €1.49 billion in the Adsense matter discussed above;
- €2.42 billion for similar transgressions in the Google Shopping case; and
- €4.34 billion for tied-selling and illegal profit-sharing agreements involving Android mobile devices.
[See EC, Google Search (Adsense), press release re Case AT.40411 (March 20, 2019); EC, Google Search (Shopping), press release re Case AT.39740 (June 27, 2017); EC, Google Android, press release re Case AT.40099 (July 18, 2018).]
An order may also issue for “the compulsory licensing of intellectual property” in a given dataset. The purpose of such an order would be to provide uniform access to a given dataset among competitors. However, depending on the nature of the dataset, the consent of data subjects may be required, as briefly discussed in subsection I of Chapter 10.
[See Bureau, “Big data and Innovation: Implications for competition policy in Canada” (Sep 18, 2017) at 28-29.]
Moreover, licensing terms may require copyright to be established, which in turn will require consideration of various factors, including the dataset’s originality. Originality in the data is unlikely to be established simply because the data was mined using artificial intelligence in some novel or creative way.
[See Toronto Real Estate Board v. Commissioner of Competition, 2017 FCA 236 at para. 3 and paras. 192-195, clarifying 2016 Comp. Trib. 7 at paras. 26, 718-720, 731-735, 738-742, 753-754, leave to appeal to the SCC denied (Aug 23, 2018).]
An order to license out patents may also be made. For example, in 2012, Google acquired Motorolla and its extensive portfolio patents, inclusive of “industry standards used to provide wireless connectivity and for internet-related technologies.” To address the FTC’s concerns, Google agreed to license its standard-essential patents, on fair and reasonable terms, to enable competition in the market for wireless devices.
[See Bureau, “Big data and Innovation: Implications for competition policy in Canada” (Sep 18, 2017) at 27 re IP licensing as a quasi-structural remedy; FTC, “Google Agrees to Change Its Business Practices to Resolve FTC Competition Concerns In the Markets for Devices Like Smart Phones, Games and Tablets, and in Online Search” (Jan 3, 2013).]
Under Canadian competition law, the purpose of such orders is to force compliance, not to impose penal consequences. Various aggravating consequences are prescribed for the purpose of assessing a fine.
[See Competition Act, ss.79(3), (3.2)-(3.3); Bureau, “Big data and innovation: key themes for competition policy in Canada” (Feb 19, 2018) at 5.)
Additional remedies are available when the market abuse results from compliance with the laws of a foreign nation, or with the involvement of foreign suppliers.
[See Competition Act, ss.82(c)-(d), 83(1)(c)-(d), 84.]
(c) Cartels and Agreements that Suppress Competition
i. Overview
Hardcore cartel agreements, which expressly aim to restrict or lessen competition, are the most egregious form of anti-competitive conduct.
[See Bureau, “Big data and Innovation: Implications for competition policy in Canada” (Sep 18, 2017) at 29.]
Collusive agreements with data are not new. In the late 1980s, Air Canada and Canadian Pacific Airlines dominated the Canadian domestic airline market. Their respective computer reservation systems merged to create Gemini, which had an 80% share of the flight-booking market.
As noted by one of the intervening parties, the merger would make it possible for Air Canada and Canadian Pacific “to observe each other’s discount seat inventories, thereby reducing competition….” As a result, a consent order issued for Gemini to share “look and book” functionality with all customer reservation systems operating in Canada.
[See Director of Investigation & Research v. Air Canada et al., Consent Order (Jul. 7, 1989) at 4-5, 11, 13, 19-20, 43.]
More recently, the Bureau has noted, “conspirators may agree to use the same algorithm to maintain prices for a large array of products.” The Bureau has also noted that, “a cartel could share large data sets of inventory information to facilitate an agreement to restrict output….” These observations were followed by a discussion of recent cartel examples through the Amazon Marketplace and Eturas.
[See Bureau, “Big data and Innovation: Implications for competition policy in Canada” (Sep 18, 2017) at 29-30.]
ii. Liability and Penalties
A company’s involvement in a “hardcore cartel” can give rise to criminal charges. The offence applies with respect to agreements to control production or supply or supply prices, and agreements to allocate markets related to production or supply. Although the criminal offence must be proven beyond a reasonable doubt, this standard may be met with circumstantial evidence. The penalty can involve imprisonment for up to 14 years, a fine up to $25 million, and an interim injunction. Moreover, the reach of the hardcore cartel offence and penalty provisions expressly extends beyond the borders of Canada.
[See Competition Act, ss.33(1), 45(1)-(3), 46(1).]
A lesser violation applies in respect of other unfair influence over the price of a product, further up or down the chain of commerce. One of the available remedies for this offence is an order requiring the violator(s) to do business on usual trade terms. Additional remedies are available when the cartel-like effects result from compliance with the laws of a foreign nation, or with the involvement of foreign suppliers.
[See Competition Act, ss.75(1), 76(1)-(2), (8), 82(c)-(d), 83(1)(c)-(d), 84.]
A prosecutor may also pursue a “facilitating practices” claim—either as a criminal or civil offence. The criminal version of this claim may involve unilateral acts that facilitate, or circumstantially indicate, the presence of an anti-competitive agreement.
[See Competition Act, s.45(3).]
The civil version of a “facilitating practice” falls within a general catch-all prohibition against agreements that “lessen competition substantially in a market.” It involves an open set of analytical factors, including foreign competitors and barriers to market entry. One of the available remedies is an order that the offending party not perform under the anti-competitive agreement.
[See Competition Act, s.90.1(1)-(2).]
The Bureau has noted the following cautionary examples of big data facilitating practices:
- the unilateral disclosure of a pricing algorithm to competitors;
- an airline’s fare-dissemination service that signals upcoming price increases for competitors; and
- a price reporting service among petrol retailers.
[See Bureau, “Big data and Innovation: Implications for competition policy in Canada” (Sep 18, 2017) at 34.]
iii. The Defence of Conscious Parallelism
One defence against the presence of a hardcore cartel is “conscious parallelism”. This is a defence against the presence of an agreement to suppress competition. Conscious parallelism involves unilateral price discovery and market actions. Competitors thereby adopt the same business or pricing practices based on their own market observations.
[Bureau, “Big data and Innovation: Implications for competition policy in Canada” (Sep 18, 2017) at 30.]
Automated pricing algorithms, potentially executed through bots, are poised to change the traditional market dynamic of conscious parallelism. Bots can now react instantaneously to competitor price changes, which may result in further third-party price changes through third-party bots. The result could be market-wide price stability at the lowest justifiable price, or alternatively, continual upwards price pressure for a prolonged period.
In most cases, algorithm-driven conscious parallelism does not trigger a criminal offence. However, it may still give rise to civil penalties if it amounts to “parallelism by tacit agreement.” This, in turn, may reverse the onus of proof and require a defending party to disprove collusion.
[See Bureau, “Big data and Innovation: Implications for competition policy in Canada” (Sep 18, 2017) at 32; Bureau, “Big data and innovation: key themes for competition policy in Canada” (Feb 19, 2018) at 9; Atlantic Sugar Refineries Co. Ltd. et al. v. Attorney General of Canada, [1980] 2 SCR 644 at paras. 656-57.]
Having to disprove collusion in these circumstances would be complex and expensive for a defendant. The legal issue is whether a tacit agreement probably existed. The defendant’s job would be to prove that the probability of a subjective “meeting of the minds”, to the nebulous standard of a “tacit” agreement, is less than 51%.
Tacit agreements were considered by the European Court of Justice in the Eturas case. In that case, a Lithuanian travel booker sent an e-mail to various travel agents, proposing a uniform 3% limit on discounts. The Court held that the e-mail established a tacit agreement if the receiving parties were aware of the content of the message. The receiving parties were then tasked with rebutting the tacit agreement, for example with evidence that they:
- publicly distanced themselves from the proposed anti-competitive act; or
- reported the proposed act to the administrative authorities.
[See “Eturas” UAB and Others v Lietuvos Respublikos konkurencijos taryba (Jan 21, 2016), Case C-74/14 at paras. 15, 28, 39-41, 45-46, 50.]
(d) Deceptive Marketing Practices
i. Overview
The Bureau has stated: “The collection and use of data that go beyond what consumers would reasonably expect increases the likelihood of deception.”
[See Bureau, “Big data and Innovation: Implications for competition policy in Canada” (Sep 18, 2017) at 39.]
Insufficient communication and consent regarding the collection and use of data may therefore amount to a deceptive marketing practice. Such deception has implications in privacy law, consumer protection law, and competition law.
[See subsection I of Chapter 2, Chapter 6 and Chapter 7; provisions in provincial consumer protection legislation re deceptive representations, e.g. Ontario’s Consumer Protection Act, 2002, SO 2002, c 30, Sch A, s.14.]
The most conceivable scenario, within the big data context, is for deceptive marketing to trigger shared jurisdiction between the Bureau and federal Privacy Commissioner.
[See PIPEDA, Sch.1, Principle 4.3.5 re consent obtained through deception; Competition Act, s.74.01(1)(a) re misleading representations; Bureau, “Big data and innovation: key themes for competition policy in Canada” (Feb 19, 2018) at 13 and fn 20.]
In the absence of such shared jurisdiction, the Bureau’s investigation into deceptive marketing may still give rise to an international, multi-body investigation. Recall the international competitional law framework, discussed in subsection I of this chapter. Moreover, recall that PIPEDA amendments are expected to establish a framework for collaboration between the Bureau and federal Privacy Commissioner, and between the federal Privacy Commissioner and any other regulator in the world. In fact, inter-jurisdictional cooperation among privacy regulators is already a norm.
[Bureau, “International community creates new framework for fair and effective competition law enforcement” (May 1, 2019); ETHI Committee Report, “Democracy Under Threat: Risks and Solutions in the Era of Disinformation and Data Monopoly” (Dec 2018) at 59 and 73, Recommendations 13 and 25; OPC, “Privacy Law Reform - A Pathway to Respecting Rights and Restoring Trust in Government and the Digital Economy” (Dec 10, 2019) at 59-60: “Our Office regularly collaborates with privacy regulators from other countries.”]
ii. Liability and Penalties
A public misrepresentation, whether in electronic form or otherwise, may amount to a criminal offence and grounds for the Bureau to commence a review proceeding.
[See Competition Act, ss.52(1), 52.01(1)-(3), 74.01(1)-(3), 74.011(1)-(3).]
The criminal offence involves an intentional or reckless representation to the public, that is false or misleading, concerning the supply or use of a product. Such conduct can result in imprisonment for up to 14 years, a fine in the court’s discretion, and an injunction. Moreover, disobeying certain Tribunal orders can trigger criminal consequences, resulting in up to 5 years’ imprisonment and/or an additional fine, along with an injunction.
[See Competition Act, ss.33(1)-(1.2), 52(1), 52(5)-(6), 66.]
The lesser reviewable penalty may include a fine of $1 million for an individual, and $15 million for a corporation. The terms for such an order will be aimed at enforcing compliance, rather than punishment.
[See Competition Act, ss.74.1(1), 74.1(4), 74.111(1).]
Note that both the criminal and civil offences also apply to representations made from outside Canada.
[See Competition Act, ss.52(2.1), 74.03(2).]
iii. Example Cases and Remedies
Over the last 20 years, the “online lead generation” business model has too often involved data aggregation and brokerage without adequate consent being obtained. In this context, data is collected surreptitiously by a public-facing entity, which sells that data to a third-party entity, which pools and processes the data for sale to other entities. This business model was squarely at issue, albeit with a privacy emphasis, in the FTC’s recent landmark settlement with Facebook for $500 billion.
[See DOJ, “Facebook Agrees to Pay $5 Billion and Implement Robust New Protections of User Information in Settlement of Data-Privacy Claims” (Jul 24, 2019), Press Release No. 19-801.]
Even when collected data is not sold to third parties, the surreptitious collection itself will trigger deceptive marketing issues. For example, the FTC pursued BLU for illegally exercising control over its devices in the manner of “firmware over-the-air” updates. The updates involved regular collection of location data and text messages, without adequate protection, user knowledge, or user consent.
[See FTC v. BLU Products, Inc. et al. (Sep 8, 2018), FTC File No. 1723025 (complaint, order and consent agreement).]
Another deceptive marketing example involves online marketplaces. The Bureau has criticized Amazon.ca with respect to its display of savings on various items. In particular, it criticized the practice of inaccurately displaying comparisons between regular prices and sale prices. The Bureau confirmed that savings can only be displayed outright if the supplier sells a substantial volume at the list price, or offers the product for a substantial period of time at the list price. Amazon was forced to agree to a consent order, involving a $1 million fine and $100,000 payable to the Bureau in legal costs.
[See Bureau, “Competition Bureau statement regarding its inquiry into Amazon’s price advertising in Canada” (Jan 11, 2017); Government of Canada, “Ordinary Price Claims” in the “Enforcement guidelines” (Oct 16, 2009), s.4.2.]
More recently, the Bureau levied an even greater fine—for essentially the same conduct—against the Hudson’s Bay Company (“HBC”). In that case, HBC likewise marketed Sleep Sets at substantially reduced prices displayed alongside misleading reference prices. In a consent agreement, HBC agreed to pay a $4 million fine and $500,000 to the Bureau in legal costs.
[See Commissioner of Competition v. HBC (May 8, 2019), CT-2017-008 (consent agreement).]
The online payment-processing sector has also been scrutinized for deceptive marketing. Venmo, an affiliate of PayPal, operated a peer-to-peer payments service involving customer deception. Upon receiving payment, users were unable to transfer funds to their bank accounts as promised. Upon complaint, Venmo subjected users to unfair document demands, and unfair freezing of funds during the ticket review process. Venmo also failed to timely verify or approve consumer transactions, without providing an explanation or resolution for the problem. To make matters worse, Venmo also employed misleading privacy practices regarding transaction sharing, which sparked unauthorized account access and fraudulent transactions.
[See FTC v. Paypal (May 24, 2018), FTC File No. 1623102 (complaint, order and consent agreement).]
Deceptive marketing practices are obviously not limited to the above example cases. Additional examples include, but are not limited to:
- privacy apps that are not private as advertised;
- consumer-targeting by unvetted entities;
- fake online reviews to promote brand awareness and quality; and
- click-farms working to generate falsely positive market reception (e.g. through “likes” on Facebook).
IV. Conclusion: The Future of Big Data Competition
Competition and anti-trust laws aim to promote ingenuity in the marketplace by discouraging market unfairness. In the context of big data, the Bureau has stated that competition policy will remain the same, notwithstanding the potential need for new methods of analysis.
[See Bureau, “Big data and innovation: key themes for competition policy in Canada” (Feb 19, 2018) at 4-5.]
Meanwhile, there is a clear trend—following a growing need—that competition rules and enforcement relating to big data are expanding beyond national borders.
[See “Ottawa Declaration, May 28, 2019”, appearing as “Appendix A” in ETHI Committee Report, “International Grand Committee on Big Data, Privacy and Democracy” (Jun 2019) re “market power arising from the concentration of data in large digital platforms”; G7 Competition Authorities, “Common Understanding of G7 Competition Authorities on ‘Competition and the Digital Economy’” (Jun 5, 2019); “Chair’s Summary: G7 Finance Ministers and Central Bank Governors’ Meeting” (Jul 18, 2019) re “competition rules and … enforcement can and should adapt as appropriate to new challenges, including those raised by the digital transformation or new economic understanding.”]
Regulatory scrutiny over big data is at an all-time high, and there are no signs of abatement given that data is now essential to modern life. Business risk is accordingly high for ventures—of all sizes—hoping to capitalize on anti-competitive strategies that previously went unnoticed.
Deals and business practices that “walk the line” of competition law boundaries should be carefully considered. The best approach is to retain counsel who understands big data markets, domestic competition law, international trends in competition law within the big data context, and the related privacy and consumer protection issues.
The copyright and disclaimer, as contained in the publication page of Big Data Law in Canada, applies to the content of this webpage.