Examining Refusal to Deal in Indian Digital Ecosystems

[Mustafa is a student at Rajiv Gandhi National University of Law.]

The dominance of big tech companies can be broadly justified as an outcome of strategizing and reaping the advantage of a package of third-party services in order to deliver the most attractive products to consumers. Where dominant digital platforms (gatekeepers) act as facilitators for a variety of services, access to gatekeepers may involve interaction with multiple players across different markets depending on their interoperability, resulting in the formation of a digital ecosystem. The value of such platforms increases with complementary contributions from third-party providers. For instance, akin to the facts of the European Union’s (EU) Case C-22/23 (Android Auto), automobile infotainment services may be facilitated by Google’s Android Operating System, while other amenities such as music may be provided by companies like Spotify. 

The rise of digital ecosystems entails an interesting phenomenon of requiring antitrust enforcers to strike a balance between the dominant entity’s incentive to invest, innovate and its right to profit from its business, and third parties’ ability to access the dominant player’s infrastructure and meet consumer demand and interest. 

In this article, the author shall examine the concept of refusal to deal as a form of abuse of dominance in the Indian digital antitrust regime.

Legal Position in India

Section 3(4) of the Competition Act 2002 (Act) prohibits vertical supply or distribution agreements that refuse to deal with a person or classes of persons, if such agreements cause or are likely to cause an appreciable adverse effect on competition (AAEC) in India. 

Although refusal to deal as a form of abuse of dominant position has not been explicitly incorporated within the ambit of Section 4 of the Act, available literature suggests the possibility of such refusal to constitute abuse (see here and here). In Competition Commission of India v. Schott Glass India Private Limited (Schott Glass), it was argued by the respondents that “absolute refusal to supply means abuse under Section 4(2)(c) of the Act.” The Supreme Court of India (SC), without specifically deliberating upon this, mandated an effects-based analysis to establish abuse of dominance under Section 4. While this ratio is much welcome as it aligns India towards a more economic approach, there is no clarity on how this will play out for enforcement within digital ecosystems. 

In the context of the EU, Geradin and Huijts write about the dangers of the effects-based approach rendering competition-enforcement proceedings extremely lengthy, and a resultant burdensome intervention process by authorities in case of an excessively rigid implementation of the approach. They note that only the AAEC analysis in the European Commission's Intel decision of 2009 spans over 150 pages, and the case took almost 9 years from the date of filing to the adoption of the final decision. Similarly, in Schott Glass, the SC passed a judgement more than a decade after the Competition Commission of India (CCI) had first passed a contravention order against Schott Glass India Private Limited. 

The economies of scale and data-driven network effects of dominant players in digital ecosystems, as discussed extensively in Report of the Committee on Digital Competition Law (Report) are too complex and elaborate. In the absence of interim measures, lengthy enforcement will simply lead to delayed justice, and the harm caused to competition during that period would be irreparable. In lieu of this, implementing an unbridled effects-based analysis for cases of refusal to deal as abuse of dominance in the digital sphere warrants reconsideration.

Analysis of the Law in European Union and Other Jurisdictions

In the landmark EU cases C-241/91 (Magill) and C-7/97 (Bronner), it was held that refusal constitutes an abuse “where the resource is indispensable to the exercise of an activity on a downstream market, the refusal excludes any competition on that market, and the defendant fails to provide an objective justification.” However, recent judgements pronounced by the EU related to big tech companies have restricted the application of the criteria developed by Magill and Bronner. For example, in the Google Shopping case of 2024,  the EU distinguished the facts from Bronner to state that access to resources had been provided, albeit on discriminatory conditions, which could not be dealt under the latter’s criteria. 

Similarly, in the Android Auto judgement, the EU noted that the circumstances for the applicability of the Bronner criteria differed significantly from the original case. Instead, as analyzed by Philipp Hornung, the exceptions to the duty to not refuse to deal were derived from the interoperability provisions of the Digital Markets Act (DMA) in Articles 6(4), 6(7), and 7 – highlighting that not only is the DMA influenced by Article 102 of the Treaty on the Functioning of the European Union that deals with abuse of dominance, the DMA also influences the latter. Notably, both Android Auto and Google Shopping expressed concerns regarding specific functioning and characteristics of digital markets, before overruling the applicability of Bronner or Magill. 

From a reading of the aforementioned provisions of the DMA, the following exceptions can be developed, provided they are duly justified by the dominant entity: 

1. Ensuring interoperability is mandatory, but the gatekeeper can be allowed to take necessary and proportionate measures to ensure that third-party software applications do not endanger the integrity of the its product;

2. The gatekeeper can be allowed to implement necessary and proportionate measures/settings beyond the default ones to effectively protect security in relation to the third-party’s product.

The requirement of compulsory interoperability is reflected in antitrust legislations of other jurisdictions as well. Section 20(3) of the United Kingdom’s Digital Markets, Competition and Consumers Act 2024 provides for the imposition of conduct-specific requirements on designated undertakings to prevent them from restricting interoperability. In the same vein, the 10^th amendment of the German Competition Act outlines ex-ante obligations of “undertakings of paramount significance for competition across markets,” and prohibits refusal of interoperability under Section 19a(2)(5), provided that such conduct is “objectively justified.” In essence, these provisions encapsulate the crux of refusal to deal in digital ecosystems. 

Gaps in India’s Digital Ecosystems and Antitrust Regime

India’s Digital Competition Bill (DCB) contained within the Report includes Section 13, whereby largescale tech companies designated as ‘Systematically Significant Digital Enterprises’ (SSDE) have the obligation to not impede access to their platforms for consumers and third-party providers.  Furthermore, the DCB adopts a ‘broad principles’ model (see paragraph 3.37 of the Report) and allows the CCI to spell specific regulations and separate conduct requirements for each entity based on the nature of the market, number of users and other factors. 

Even if the DCB operates as an ex ante framework, the SC in Schott Glass (See Paragraph 59) stipulated that Section 4 of the Act contemplates two separate findings: “that the impugned practice falls within one of the descriptive clauses (a)-(e) of sub-Section (2), and that it results in, or is likely to result in, an appreciable adverse effect on competition” (emphasis supplied). There is an apparent lack of clarity; the CCI is allowed take preventive measures under both the Act and the DCB against an SSDE (see paragraph 3.50 of the Report), and impose penalties under both legislations. This can lead to disproportionate penalties, extensively affecting an SSDE’s economic interests. 

Furthermore, Section 19(3) of the Act lists down various factors that can be considered by the CCI to determine if AAEC has been caused by a dominant firm. The same factors can also be rebutted under Section 19(3)(d), (e) and (f), by proving for instance, that the conduct promoted technical development. However, Section 19(3) unequivocally states that the factors are applicable in the context of determining AAEC caused by agreements under Section 3. Thus, without the mandate of  Section 19(3), conducting an effects-based analysis or determining AAEC for refusal to deal under Section 4 of the Act is difficult, if not impossible. Even Schott Glass did not explicate how an effects-based analysis under Section 4 can be carried out. Additionally, the lack of exceptions for SSDEs, similar to the ones available to gatekeepers under the DMA provisions, obfuscates the possible defenses that can be invoked by them. 

Concluding Remarks

The dynamism of digital ecosystems is transcending boundaries. To protect consumer  and third-party interests against a refusal to deal abuse, dual penalty system under the Indian antitrust regime exists. However, to safeguard the legitimate business interests of dominant entities, the author advocates for the incorporation of defenses based on principles of necessity and proportionality similar to the DMA, within the ambit of Section 13 of the DCB. In a nutshell, dominant entities can refuse to deal based on objective justification, if such a measure is necessary, and only that which is necessary can be implemented. Dependent on CCI’s enforcement practice, the influence of Section 4 of the Act on the DCB and vice-versa is not lucid – only time will tell.