INTRODUCTION
In the ever-evolving realm of electronic trade and digital marketplaces, the intersection of sophisticated technologies and commercial tactics has spurred a phenomenon that questions the fundamental nature of equitable rivalry: algorithmic pricing collusion. This intricate concern orbits around the intricate interaction of mechanized pricing algorithms, market fluctuations, and unintentional consequences detrimental to healthy competition. At its heart, algorithmic pricing collusion denotes the circumstance wherein rival enterprises, propelled by their mechanized pricing algorithms, either inadvertently or purposely synchronize their pricing tactics in manners that subvert competition and potentially impair the well-being of consumers. The ascent of algorithmic pricing is steered by its capacity to amplify efficiency, adaptability, and financial gains, equipping enterprises to successfully navigate the intricacies of the digital marketplace.
UNDERSTANDING ALGORITHMIC COLLUSION
Price-setting algorithms facilitate the immediate adjustment of prices based on a consumer’s past choices and actions. These self-enhancing algorithms for pricing enable users to amass and utilise various demand-related information, considering multiple factors, to establish the optimal price for a particular product for that specific customer at that given moment.[1] Additionally, these algorithms enable quicker responses to shifts in demand, supply, and other variables than what can be achieved manually. The prevalent incorporation of algorithms in e-commerce has not only rendered dynamic pricing feasible but also commercially viable, as companies can gather and analyze abundant personal and behavioural data from online users with minimal transaction expenses. Similarly, algorithms empower companies to modify prices with minimal time and exertion.
It is widely acknowledged that contemporary pricing algorithms possess the potential to not just simplify collusion but also to introduce novel modes of coordination that were previously unobserved or unattainable. This occurrence is termed “algorithmic collusion,” and it has been elaborated upon by Ezrachi and Stucke, who outlined four conceivable methods through which algorithms might be employed to facilitate or result in collusion.[2]
1] Messenger Algorithms
In the context of messaging platforms, market participants utilize computers or a single algorithm to engage in collusion. An illustration of this can be seen in the case of United States v. David Topkins [3], where the individuals involved chose to employ a unified pricing algorithm for selling their posters. The intention behind this was to ensure uniform pricing, but this arrangement was deemed unlawful. A similar situation emerged in 1994 when six airlines shared a computerised online booking system, which facilitated collusive pricing and was consequently deemed to violate anti-competitive regulations.[4] However, it’s important to note that the software itself merely carries out the directives of humans who are pursuing collusion.[5] The fundamental collusive agreement continues to be a product of human decision-making.
2] Hub and Spoke Conspiracy
Numerous companies opt to delegate the development of dynamic pricing algorithms to a shared IT provider, or they incorporate an identical dataset into their respective algorithms. Consequently, this gives rise to a network of comparable vertical agreements involving a multitude of industry rivals.[6] In this scenario, the algorithm’s developer acts as the central ‘hub’, while the competitors represent the ‘spokes’. The developer furnishes these competitors with nearly identical iterations of the algorithm. The employment of similar algorithms and identical datasets streamlines the synchronization of pricing choices, ultimately resulting in what can be described as a hub-and-spoke conspiracy.
3] Predictable Agent
In this particular situation, each company independently develops its own machine to produce consistent results and respond in specific manners to shifting market dynamics. There is no collaborative agreement among competitors; instead, the motivation for creating and utilizing the algorithm stems from each company’s individual economic interests. While these firms might possess insights into the probable functionalities of algorithms employed by their rivals, n formal arrangement or cooperation is established.
Nevertheless, the utilisation of algorithms in such a manner has the potential to generate anticompetitive consequences through interconnected operations of these algorithms.
4] Autonomous Machines
Autonomous machines represent the pinnacle of algorithmic implementation in price determination. In this context, businesses independently develop and employ algorithms to attain specific objectives, such as profit maximisation or sales enhancement.[7]
LEGAL PERSPECTIVE
Algorithmic collusion is becoming a growing global concern, particularly within the realms of online marketplaces and digital advertising. This phenomenon holds the potential to significantly impact competition and the well-being of consumers, as it can result in diminished competitive forces, escalated prices, and restricted consumer options. In the United States, entities such as the Federal Trade Commission (FTC) and the Department of Justice are tasked with addressing this issue.[8] Likewise, in Europe, the European Commission (EC) is responsible for upholding competition law and countering activities that undermine fair competition. In India, the Competition Commission of India (CCI) is vested with the authority to enforce legislation aimed at fostering competition and safeguarding consumers against practices that hinder fair competition. [9]
India has encountered several cases related to algorithmic collusion, each shedding light on various aspects of this issue.
- Matrimony.com v. Google LLC[10]: In this case, it was brought to attention that Google had control over its search algorithms, allowing it to manipulate the relevance of search suggestions. This was alleged to be misleading to consumers. The Competition Commission of India (CCI) found Google’s practices to be discriminatory, unfair, and manipulative toward users.
- Allegations against Ola and Uber[11]: Another case revolved around hub-and-spoke collusion accusations involving ride-sharing apps Ola and Uber. It was alleged that these apps were using Pricing Algorithms to coordinate and fix prices among their drivers, effectively creating a cartel in violation of Section 3 of the Act. CCI laid down the criteria for establishing hub-and-spoke collusion, emphasising the need for sensitive information exchange and a conspiracy or agreement to fix prices.
Until the new amendment act section 3 of the act did not necessitate the presence of an explicit agreement. Collusion or pricing cartels could only be established if a party could demonstrate the existence of an agreement, whether implicit or explicit. Such an agreement could also be inferred from circumstantial evidence.[12] This made Section 3 broad enough to encompass intentional collusion through pricing algorithms, even without an overt agreement.
Building upon the established concepts of traditional cartels, both horizontal and vertical, the Competition Commission of India (CCI) has introduced the notion of a “Hub and Spoke Cartel” within the context of Section 3 of the Act’s latest amendment.[13]This recent amendment explicitly recognises and addresses the concept of ‘hub and spoke’ cartels. The primary objective of this provision is to encompass collusive and anti-competitive agreements between parties that do not operate within the same or similar sectors. Specifically, it covers agreements between a facilitator, platform, intermediary, or agent (referred to as the ‘hub’) on one side, and one or more competing entities (termed the ‘spokes’) on the other side.[14]
The incorporation of hub-and-spoke cartels is poised to disrupt existing cartels and foster a culture of competition within the country’s market. Its implications carry substantial weight for businesses, consumers, and various stakeholders.
CONCLUSION
To summarise, the emergence of algorithmic pricing collusion represents a significant convergence of technology and the intricate landscape of competition and consumer protection. As we navigate the digital era, the interplay between autonomous machines, data analytics, and market dynamics brings about unprecedented opportunities and unforeseen challenges. The discussed cases and considerations highlight the intricate nature of addressing algorithmic collusion within the regulatory framework. As legal provisions and competition authorities adapt to encompass these novel scenarios, maintaining a balance between promoting innovation and ensuring fair competition becomes crucial. A collaborative approach involving policymakers, industry stakeholders, and technology experts becomes indispensable in formulating strategies that discourage anti-competitive behaviours, promote transparency, and ensure that algorithmic potential aligns harmoniously with competition and consumer welfare principles. This collaborative effort can harness the transformative influence of algorithms to reshape industries while upholding competitive and equitable market ideals for the benefit of all.
Author(s) Name: Bhoomi Jain (ILS Law College, Pune)
References:
[1]‘Dynamic pricing: In-depth Guide to Improved Margins’ (2018) <https://blog.appliedai.com/dynamic-pricing/>
accessed 22/08/2023c
[2] Ariel Ezrachi and Maurice E Stucke, ‘Artificial Intelligence & Collusion: When Computers Inhibit Competition’ [2015]
[3] No. CR 15-00201 WHO (2015, US NDC)
[4] Martin Tolchin, Six Airlines Settle Suit by Government on Fares, The New York Times, 18-3-1994, available at <https://www.nytimes.com/1994/03/18/business/six-airlines-settle-suit-by-government-on-fares.html> (last visited on 22-8-2023)
[5] Ezrachi and Stucke, ‘Artificial Intelligence & Collusion: When Computers Inhibit Competition’ (n 31)
[6] Ibid
[7] Ezrachi and Stucke, ‘Artificial Intelligence & Collusion: When Computers Inhibit Competition’ (n 31) 9.
[8] Indulia B and Bhardwaj P, ‘Changing Dynamics of Algorithmic Collusion: An Analytical Study’ (SCC Blog, 19 May 2023) <https://www.scconline.com/blog/post/2023/05/18/changing-dynamics-of-algorithmic-collusion-an-analytical-study/> accessed 27 August 2023
[9] Ibid
[10] Matrimony.com v. Google LLC [2012] CCI
[11] Samir Agarwal v. ANI Technologies Pvt. Ltd. [2018] CCI
[12] Director General (Supplies & Disposals) v. Puja Enterprises [2012] CCI
[13] ‘Analyzing the Competition Amendment Bill Vis-a-Vis Regulation of Digital Market’ (CBCL, 14 September 2022) <https://cbcl.nliu.ac.in/competition-law/analyzing-the-competition-amendment-bill-vis-a-vis-regulation-of-digital-market/> accessed 27 August 2023
[14] Kakkar A ed., ‘2023 Amendments to Indian Competition Law: Bringing Down the Hammer on Anti-Competitive Conduct (Part 2)’ (Kluwer Competition Law Blog, 4 May 2023) <https://competitionlawblog.kluwercompetitionlaw.com/2023/05/04/2023-amendments-to-indian-competition-law-bringing-down-the-hammer-on-anti-competitive-conduct-part-2/> accessed 27 August 2023
