Canada: Comment: Ohio v. American Express - "Multi-Sided Platforms"

'Multi-sided platforms' and 'Big Data' are given significant attention by the Competition Bureau's guidelines and discussion papers. Both regulators and practitioners are debating the appropriate lens through which 'Multi-Sided Platforms' should be viewed, whether through the traditional approaches of market definition and the Hypothetical Monopolist Test (the SSNIP test) or through novel, specialized tools and methods. On June 25, 2018, the US Supreme Court ruled in a 5/4 decision that American Express' 'anti-steering provisions' with merchants did not violate U.S. antitrust laws. The Majority opinion defined the relevant market as the credit card network—a transaction platform constituting both sides of a two-sided platform, facilitating a single simultaneous transaction between merchants and cardholders. The Majority rejected the plaintiff's argument that increasing merchant fees was proof of anticompetitive acts, because it did not demonstrate anticompetitive effects on both sides of the credit card market. The Dissent, however, dismissed such a non-traditional market definition as un-precedented in antitrust law. This article analyzes how the US Supreme Court's decision may affect the way Canada's Competition Bureau and Competition Tribunal perceive 'Multi-Sided Platforms'. Will the traditional or novel approach to competition law prevail?

«  Plateformes multifaces  » et «  mégadonnées  » sont des sujets récurrents dans les lignes directrices et les documents de travail du Bureau de la concurrence. Les autorités de réglementation et les praticiens débattent encore de la question de savoir sous quel angle il convient d'aborder ces plateformes  : par les méthodes traditionnelles de définition des marchés et le critère du monopoleur hypothétique (de la SSNIP), ou bien par des méthodes et outils spécialisés d'un genre nouveau? Le 25  juin  2018, la Cour suprême des États-Unis statuait, dans une décision partagée à cinq voix contre quatre, que les clauses «  anti-incitatives  » imposées aux commerçants par American Express ne contrevenaient pas aux lois antitrust américaines. Dans l'opinion majoritaire, on définissait le marché en cause comme étant le réseau de cartes de crédit—une plateforme transactionnelle qui constitue les deux versants d'une plateforme biface, facilitant une transaction unique instantanée entre le commerçant et le titulaire de la carte. La majorité a rejeté l'argument du demandeur selon lequel la majoration des frais payés par les commerçants constituait une preuve d'agissements anticoncurrentiels, estimant que l'existence d'effets anticoncurrentiels de part et d'autre du marché n'avait pas été démontrée. Les juges dissidents, pour leur part, ont rejeté cette définition non traditionnelle du marché, invoquant l'absence de précédent en droit antitrust. Le présent article analyse en quoi la décision de la Cour suprême des États-Unis pourrait changer le regard que portent le Bureau de la concurrence et le Tribunal de la concurrence du Canada sur les «  plateformes multifaces  ». Tradition contre modernité  : quelle approche triomphera?

Introduction

'Big Data' is a term that describes collection and commercial use of large quantities of information by technology companies such as Alphabet (Google), Facebook and Amazon.²

'Multi-Sided Platforms'/'Two-Sided Platforms'' (collectively MSPs) are often mentioned in the context of 'Big Data'. MSPs sell distinct products or services to two or more distinct but interdependent customer groups and connects them. The coordination facilitated by the platform creates value for all participants which could not be offered through traditional means of market interaction.³ To name a few examples of MSPs: the credit card network that connects cardholders and merchants, a ride sharing platform such as Uber that connects drivers and passengers, and Airbnb that connects owners and renters.

'Big Data' and innovation's pervasiveness in the Canadian and global economy affects the day-to-day life of Canadians, and therefore MSPs are given significant attention by the Competition Bureau (the "Bureau") in guidelines and discussion papers.⁴ Both regulators and practitioners are debating the appropriate lens through which MSPs should be viewed, whether through the traditional approaches of market definition and the Hypothetical Monopolist Test (the SSNIP Test)⁵ or through novel specialized tools and methods such as a single market definition encompassing two sides of a MSP.

The debate has been furthered in the recent U.S. Supreme Court decision, Ohio v American Express ("Amex"). A divided Court ruled in a 5/4 decision that American Express' 'anti-steering provisions' with merchants did not violate U.S. federal antitrust laws.⁶ The Majority opinion defined the relevant market as the credit card network—a transaction platform constituting both sides of a MSP, facilitating a single simultaneous transaction between merchants and cardholders.⁷

The Amex Majority rejected the plaintiff's argument that increasing merchant fees was proof of anticompetitive acts because it did not demonstrate anticompetitive effects on both sides of the credit card market.⁸ Amex resonated locally and globally and sparked economic and political debate between supporters of the traditional views and approaches to antitrust and those advocating for more specialized methods and approaches.⁹

How might Amex affect the way Canada's Bureau and the Competition Tribunal (the "Tribunal") perceive MSPs? Will the traditional or alternative approach to competition law prevail? This article will address these questions. First, an explanation of the nature of MSPs and the competition issues they raise will be presented. Second, the Bureau's discussion papers and guidelines, as well as Tribunal decisions dealing with MSPs will be examined. Third, a more detailed analysis of Amex will be offered, followed by; Fourth, an analysis of Amex's potential implications on Canadian competition law.

1. 'Multi-Sided Platforms'—What They Are, And What Competition Issues Do They Raise

MSPs (such as Airbnb, American Express and Uber) sell distinct products or services to two or more distinct but interdependent customer groups and connect them. The coordination facilitated by the platform creates value for all participant groups, which could not be offered through traditional means of market interaction.¹⁰ To illustrate further, another example of a MSP is the credit card network—a 'transaction platform' run by credit card companies that brings card holders and merchants together. When a cardholder uses a credit card to purchase a product or a service from a merchant, the credit card network facilitates the transaction by providing separate but inter-related services to cardholders and merchants. For cardholders, the credit card network extends them credit which allows them to defer payments and earn rewards based on the amount they spend. To merchants, the credit card network processes the transaction, guarantees the payment and increases the number and value of sales.¹¹

Competition law scholars found 'Feedback Effects' between MSP users: steps taken by the platform may affect one user group, which necessarily affects the second user group and in turn affects the first user group again.¹² For example, price increases on one side of the platform risk losing participants on the other, which in turn, would decrease the value for the first side and create a feedback loop of declining demand. In the context of the credit card network, increasing costs for cardholders would mean fewer purchases, thereby decreasing the platform's value for merchants.¹³

MSPs are usually discussed in the context of merger review and monopolistic practices. These platforms as well as their users, like traditional businesses, may conspire to fix prices, acquire market power through mergers and attempt to obtain monopoly power through unilateral practices.¹⁴ Nevertheless, the traditional tools of analysis may need to be modified in handling MSPs.¹⁵ For instance, prices below variable cost on one side of a MSP cannot be evidence of predatory pricing, because it is a constant characteristic of many such platforms seeking to attract users to the platform, unrelated to competitive conditions.¹⁶

2. The Current Views of The Competition Bureau And Tribunal On MSPs

MSPs have been addressed by the Tribunal and Bureau through cases, discussion papers, and guidelines. The 2013 Visa-Mastercard¹⁷ and 2016 TREB¹⁸ Tribunal decisions are covered below, followed by a review of the Bureau's Abuse of Dominance Enforcement Guidelines (The "Dominance Guidelines")¹⁹, and the Big Data and Innovation Discussion Paper (the "Big Data Paper")²⁰.

2.1 Visa-Mastercard

The Bureau alleged that Visa and MasterCard's 'merchant rules' (no discrimination, honour all cards and no surcharge rules) discouraged the reduction of 'merchant discount fees' (interchange, acquirer network and acquirer service fees) and breached the civil prohibition against price maintenance in s. 76 of the Competition Act, RSC 1985, c C-34 (the "Competition Act").²¹ The Tribunal considered one side of the MSP as the relevant product market and applied the SSNIP test to the price charged to the merchant. ²² The Tribunal mentioned, however, that when a hypothetical monopolist may profit from a price increase, it may be necessary to account for cross platform demand interdependence and feedback effects and changes in profit on both the customer and acquirer sides of the platform.²³

2.2 The Toronto Real Estate Board Tribunal Decision  (the "TREB Tribunal")²⁴

The Bureau alleged that certain information sharing practices of TREB prevented competition substantially in the supply of residential real estate brokerage services in the Greater Toronto Area, while disadvantaging innovative brokers who operate virtual offices breaching s. 79(1) of the Competition Act.²⁵ The Tribunal expressed its opinion that it will often be neither possible nor necessary to define the product market in s. 79 cases.²⁶ The TREB Tribunal decided that the supply of real estate brokerage services to both home sellers and home buyers constitutes a single market. In other words, it included both sides of the platform in a single market definition, somewhat similarly to the Amex Majority's market definition²⁷

2.3 The Dominance Guidelines:

In 2019, the Bureau published its Dominance Guidelines, which discuss the Bureau's view of the appropriate way to analyze a MSP when assessing dominance under S. 79 of the Competition Act. Specifically, these guidelines propose different strategies in how to approach market definition, which is an analytical process used to assess whether a participant has dominance within it. The Dominance Guidelines offer the following strategies:

1. Not define the market at all: The Bureau recognizes that the market is at times impossible to define, nor is its definition necessary in every case. As an example, when services are free ('Zero Monetary Price' —e.g. free use of search engines by a MSP user), then prices are irrelevant and thus the SSNIP test would be unusable;²⁸
2. Define the market as one side of a MSP,²⁹ which effectively defines the market the traditional way.
3. When a hypothetical monopolist would profit from a price increase, the Bureau may define the market as one side of a MSP while accounting "for the interdependence of demand, feedback effects and changes in profit on all sides of the platform".³⁰
4. Define the market to include multiple sides of a MSP.³¹

In the Dominance Guidelines, the Bureau left all options on the table and is open to novel approaches to market definition, such as defining the market as multiple sides of a MSP.

2.4 The Big Data Paper

In 2018, the Bureau released its Big Data Paper, which states that the traditional and fundamental analytical frameworks of Canadian competition law, including market definition, market power and competitive effects, continue to apply to 'Big Data' and MSPs.³² However, it also accepts that examining MSPs may require specialized tools and methods if the nature of the transaction or price differs from non-platforms.³³

To illustrate this point, the Bureau provides a scenario in the paper in which a "high" price on one side of the platform might not be indicative of market power or anti-competitive effects when resulting from a "low" price on the other side.³⁴ As an example, a ride sharing platform bringing together drivers and riders might charge the passenger a higher rate during high demand times—the difference would be paid to the driver, while the benefit to the platform remains unchanged.³⁵ In this example, there is no apparent competitive harm—the change in rate is a means to regulate the platform by lowering demand and increasing supply to better allocate the scarce supply of drivers.³⁶

The Big Data Paper also discusses 'Network Effects', which exist where the value of the MSP to a group of participants depends on how many members of the group participate.³⁷ As an example, a search engine user benefits from use of other users which improves matching search results to searchers' queries, based on the search result's popularity. Thus, there is benefit to the customer when the consumer base increases.³⁸ Importantly, 'Network Effects' can be both an efficiency and a barrier to entry. Just as competition law enforcement does not challenge use of economies of scale to develop an innovative product while raising barriers to entry if anti-competitive acts are not involved, it would not challenge a firm using 'Network Effects' in a similar fashion, absent an anti-competitive act.³⁹

It is apparent from the Visa-MasterCard and TREB Tribunal decisions, as well as the Dominance Guidelines and Big Data Paper, that the Bureau and Tribunal are open to novel and specialized market definitions such as a market definition that includes two sides of a MSP, similar to the opinion of the Amex Majority, as will be elaborated below.

Footnotes

1 Elad Travis is an associate at Bennett Jones LLP in Toronto. Prior to joining Bennett Jones and the Ontario bar, Elad worked as a competition lawyer in Tel Aviv for nearly seven years. The author would like to thank Mark Katz and Shira Sasson for their insightful comments on earlier drafts of this article.

2 US Department of Justice, "'Big Data' and Competition for the Market: Deputy Assistant Attorney General Antitrust Division U.S. Department of Justice", remarks as prepared for delivery at the Capitol Forum and CQ: Fourth Annual Tech, Media & Telecom Competition Conference (New York: Department of Justice, 2017) at 3.

3 David S. Evans and Michael Noel, "Defining Antitrust Markets When Firms Operate Two-Sided Platforms" (2005) Colum Bus L Rev 102 [Evans & Noel].

4 Competition Bureau, The Abuse of Dominance Enforcement Guidelines Mar 7, 2019 (Ottawa: Industry Canada, 2019) at paras 1.3, 1.16–1.17 [Dominance Guidelines]. Competition Bureau, The Big Data and Innovation: Key Themes for Competition Policy in Canada (Ottawa: Industry Canada, 2018) at 4–6 [Big Data Paper].

5 'SSNIP': the hypothetical monopolist test—imposing and sustaining a small but significant and non-transitory increase in price. See: Competition Bureau, Merger Enforcement Guidelines (Ottawa: Industry Canada, 2011) at 11, para 4.3.

6 Ohio v American Express Co., 585 US at 20 (2018) Thomas J [Amex].

7 Ibid at 11–12, Thomas J.

8 Ibid at 15, Thomas J.

9 As examples of such polarised debate, see: Noah Feldman, "Conservative Justices Don't Much Care For Antitrust Law " Bloomberg Opinion (25 June 2018), online: https://www.bloomberg.com/opinion/articles/2018-06-25/ supreme-court-curbs-antitrust-law-in-american-express-case. The Editorial Board, "Another Antitrust Bust" WSJ—Opinions (25 June 2018), online: https://www.wsj.com/articles/another-antitrust-bust-1529968151.

10 Evans & Noel, supra note 3 at 102.

11 Amex, supra note 6 at 1–2, Thomas J.

12 Evans & Noel, supra note 3 at 103.

13 Amex, supra note 6 at 12, Thomas J.

14 Evans & Noel, supra note 3 at 104.

15 Ibid.

16 Ibid.

17 The Commissioner of Competition v. Visa Canada Corporation and MasterCard International Incorporated (23 July 2013), CT-2010-010, online: Competition Tribunal https://www.ct-tc.gc.ca/CasesAffaires/CasesDetails-eng. asp?CaseID=333 [Visa-MasterCard].

18 The Commissioner of Competition v The Toronto Real Estate Board (27 April 2016), CT-2011-003 at para 471, online: Competition Tribunal https:// www.ct-tc.gc.ca/CMFiles/CT-2011-003_Reasons%20for%20Order%20and%20 Order_385_66_4-27-2016_7296.pdf [TREB Tribunal].

19 Dominance Guidelines, supra note 4 at paras 1.3, 1.16–1.17.

20 Big Data Paper, supra note 4 at 4–6.

21 Visa-MasterCard, supra note 17 at paras 29, 46. For further analysis of the Visa-MasterCard decision in the context of S. 76, see: Mark Katz & Erika Douglas, "Resale Price Maintenance in Canada: Where Do We Stand After the Visa/Mastercard Case?" (2013) 1 CPI Antitrust Chronicle, online: http://www.albertasenator.ca/flashblocks/data/BT%20Money%20Matters/ VisaMastercardCase13November2013.pdf.

22 Visa-MasterCard, supra note 17 at para 205.

23 Ibid at para 189.

24 TREB Tribunal, supra note 18.

25 Ibid at para 2.

26 Ibid at para 132.

27 Ibid at para 138.

28 Dominance Guidelines, supra note 4 at paras 1.3, 1.17. Citing TREB Tribunal, supra note 18 at para 132.

29 Dominance Guidelines, supra note 4 at paras 1.3, 1.16.

30 Ibid at para 1.15, n 16 cites Visa-MasterCard, supra note 17 at para 189.

31 Dominance Guidelines, supra note 4 at para 1.16.

32 Big Data Paper, supra note 4 at 4–7.

33 Ibid at 7.

34 Ibid.

35 Ibid.

36 Ibid.

37 Ibid.

38 Ibid.

39 Ibid.

To view the full article please click here

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