A Cross-Comparative Analysis of the Google’s Self Preferencing and Android Cases in India,EU and the US

Abstract

This paper explores and compares two significant antitrust battles faced by Google across the EU, India, and the United States: the Google Search Bias case and the Google Android case. It delves into the allegations of “self-preferencing,” where Google is accused of steering search results to favour its own services, and examines how courts in different jurisdictions have approached this issue, balancing the protection of competition against potential benefits for consumers. The paper also examines the complexities of the Google Android case, investigating claims that Google used its dominant Android operating system to stifle competition and lock users into its ecosystem. The analysis reveals notable differences: while the EU focuses on preventing the foreclosure of competitors, often without requiring evidence of actual harm to consumers, the United States places consumer welfare and pro-competitive benefits at the centre of its scrutiny. India’s stance aligns with the EU’s in the Android case but remains inconsistent regarding self-preferencing. Through this comparative study, the paper highlights the challenge of maintaining a delicate balance between preserving competition, fostering innovation, and protecting consumer interests in the fast-evolving digital landscape. It also touches on the growing relevance of data privacy in antitrust discussions, highlighting the broader implications for regulating Big Tech.

Introduction

Google has faced charges of abuse of dominant position for several issues in many jurisdictions across the world. The two primary cases are the (Google Search Bias case) and the (Google-Android case). The Google Search Bias case ¹ sheds light on the prevalence of ‘self-preferencing’ in antitrust laws. Google is accused of engaging in ‘self-preferencing’, despite the absence of an explicit reference to the term ‘self-preferencing’ in the court’s decision. Self-preferencing refers to actions taken by a company to favour its own products or services over those of its competitors (Hunt et al., 2023). In the context of the Google Search Bias case, it involves a dominant horizontal search provider leveraging its market power to give preference to its own vertical search services, often at the expense of competing vertical search providers (Persch, 2021). This preference can manifest through adjustments to the indexing of horizontal search results, or in extreme cases, by excluding vertical search competitors from the horizontal search index (Persch, 2021). Self-preferencing can result in two primary harms: foreclosure, making it more challenging for competitors to compete fairly, and harm to consumers, reducing choice, quality, and innovation while increasing prices (Bougette et al., 2022). This article focuses on the issue of self-preferncing as examined in the Google Search Bias case, which has garnered attention of competition regulators worldwide and the approaches taken have varied from one jurisdiction to another. ²

The second infringement decision against Google discussed in this paper is the Google Android case. Courts and competition authorities have found that Google engaged in anti-competitive behavior and abused its dominant position through its android operating system (Android OS) and other online products which is provided free of cost to the users in various jurisdictions. ³ These products include Android OS, google chrome, google search, play store and other such applications, popularly known as the Google Mobile Services (GMS). The company, Google, licenses these products to Smartphone manufacturers/Original Equipment Manufacturers (OEMs). In 2018, the European Commission (EC), whose decision was later affirmed by the General Court (EU Decision), ⁴ imposed a fine on Google and held it liable for abusing its dominance and foreclosing competition by entering into several anti-competitive agreements with the OEMs. A similar conclusion was arrived at by the Competition Commission of India (CCI) in 2022 which was affirmed by the National Company Law Appellate Tribunal (NCLAT) in an appeal. ⁵ Further, a similar case, which is yet to be decided has been bought by the United States (US) government and 37 states alleging that Google has entered into restrictive contracts to unlawfully maintain a monopoly for its app store on Android phone. ⁶

The paper aims to provide a comparative analysis of the above two Google cases in the European Union (EU), India and the US to showcase how different approaches have been adopted by the courts on similar facts in determining abuse of dominance. First, the paper analyses the Google Search Bias cases, limited to the issue of self-preferencing and argues for a detailed consideration of pro-competitive benefits and consumer welfare. Second, the paper discusses the Google-Android case and argues that the limited analysis of the pro-competitive benefits and consumer welfare by courts in the EU and Indian cases may cause harm to the consumers and hamper innovation. Lastly, consumer harm in terms of data privacy in digital markets is discussed.

Google Search Bias Case

The pertinent facts presented to the EC, the CCI, and the U.S. Federal Trade Commission (FTC) were quite similar. These facts revolved around the allegation that Google was exploiting its dominant position in general search services to distort competition within specialized search verticals. ⁷ This manipulation involved promoting Google’s own specialized verticals, such as Google Flights, Google Comparison Shopping, Google Maps, more prominently.

The assessment for identifying potential abuse of a dominant position or monopolization in these various jurisdictions follows a three-step process: (i) defining the relevant market, (ii) establishing the firm’s dominant position, and (iii) evaluating whether the conduct qualifies as an abuse of that dominant position. ⁸ Despite the similarities in the factual scenarios and the overarching legal framework applied across all three jurisdictions, the divergent outcomes can be attributed to the influence of distinct legal and economic paradigms in interpreting the third leg of this evaluation. This section presents a cross-jurisdictional analysis of the Google Search Bias case, examining the contrasting positions taken by regulators in India, the US, and the EU. The analysis delves into whether self-preferencing should be inherently considered anticompetitive or be subjected to scrutiny under the rule of reason standard.

The EU Search Bias Decision

Within the EU’s legal framework, greater emphasis has been placed on the foreclosure theory of harm. The foundation for prohibiting self-preferencing was established through the EC’s decision in the Google Shopping case. ⁹ The General Court subsequently upheld this decision by employing specific criteria to assess the potential for discriminatory behaviour by a search engine. ¹⁰ This examination revealed that Google, through the implementation of its ‘Panda’ algorithm, reduced the visibility of competing comparison-shopping services in search results. ¹¹ Consequently, Google’s own offering became more prominent to consumers, who typically focus on the top search results. Google was found to have “exploited” consumers’ reliance on default search results (represented by these top results), effectively trapping users within Google Shopping instead of encouraging them to explore alternative comparison-shopping websites. ¹² This created a self-reinforcing cycle, enabling Google to display increasingly relevant results to users, thus attracting more traffic, resulting in competitors exit from the market.

The EC determined that it was not necessary to prove the existence of actual anti-competitive effects. ¹³ Instead, they ruled out the potential anti-competitive benefits. It was considered sufficient to demonstrate that the conduct had the potential or likelihood to result in foreclosure effects. ¹⁴ The General Court emphasized that the list of abusive practices outlined in Article 102 of the Treaty on the Functioning of the European Union (TFEU) is non-exhaustive. ¹⁵ Therefore, even if self-preferencing was not explicitly mentioned within this provision, it was still encompassed by the broad scope of the non-exhaustive list and considered a distinct form of abuse (Deutscher, 2021).

The judgment clarified that overall consumer welfare was not the sole criterion for evaluating the conduct. This became evident when the EC addressed Google’s argument that the conduct aimed to enhance product design and, consequently, consumer welfare as follows:

“There is no indication in the case law that alleged improvements in product designs should be assessed under a different legal standard to that developed to assess the use of a dominant position on one market to extend that dominant position to one or more adjacent markets.” ¹⁶

In essence, EC’s reasoning can be interpretated as even if the conduct was geared toward enhancing products and, by extension, consumer welfare, it would still be subjected to the test of “exclusion of competitor”. This stance appeared consistent with the principles of the Harvard school. Under the Harvard School approach, “the courts and agencies presumed the illegality of any mergers, joint ventures, or agreements that allowed firms to obtain, enhance, or exercise market power, regardless of whether the conduct had the potential to benefit consumers by lowering prices or increasing output” (Piraino, 2007).

The Indian Scenario

In India, in Matrimony. com Ltd v. Google LLC case, ¹⁷ Google has faced allegations of bias in search result rankings. The CCI undertook an examination of Google’ dominant position, with “search bias” forming one of the issues of the complaint. “The CCI delineated two distinct relevant markets: “online general web search services” and “online search advertising services” in India. In this context, CCI’s decision predominantly favoured Google. However, it did identify an instance where Google had abused its dominant position. This abuse pertained to Google’s prominent display of its Flights Unit, which exclusively showcased results from Google Flights, its in-house vertical search engine. This is very similar to the Comparison Shopping ¹⁸ case before the EC as explained above.

The issue of self-preferencing also came up in the case of Digital News Publishers Association v. Alphabet Inc.,19 where the informants, engaged in the business of news media, alleged that Google engaged in self-preferencing by giving preference to its own news verticals over competitors in the search engine results page. Google used its position as a dominant search engine to segue into online news market by bringing in Google News, Google news showcase etc. CCI formed a prima facie opinion that such conduct leads to abuse of dominance and directed an investigation. CCI opined that Google’s dominance in the relevant market for online search services, and zero-click searches, among others, allows it to ensure that the consumer stays on Google and eliminates the need to visit another player or use their services. ²⁰ The reasoning of CCI in this case mirrors with the EU case.

Comparison With the EU Case

The rationale of the Court in Matrimony. com Ltd v. Google LLC case, ²¹ for the conclusion in the flight vertical segment was twofold. First, the CCI noted that Google’s actions, by unfairly diverting traffic, hindered third-party travel verticals from gaining a substantial volume of business. The consequence of Google’s search design was either pushing these verticals down in search rankings or, in some cases, out of the market entirely. This approach aligns with the EC’s perspective, which prioritizes the protection of competitors. Second, the CCI observed that Google’s search design, which favoured Google Flights regardless of relevance, contradicted Google’s public claims regarding ranking results based on relevance. This distinction between Google’s claim and its actual practice was a key differentiator between the CCI’s decision and the EC’s approach, where the latter’s decision primarily emphasized competitor protection.

Further, the CCI did not conduct an empirical study to support its conclusion that third-party flight verticals and individual airline websites predominantly or significantly relied on traffic from Google's general search. This situation is in contrast with EC case, where substantial data was employed to demonstrate the percentage of views attributed to Google search for various search verticals. ²² Therefore, the basis of CCI’s conclusion remains unclear.

Inconsistency in CCI Judgment

The CCI did not accept the argument that Google could not crawl and index flights from third-party verticals within its Flights Unit in contrast to the situation in Universal Results. ²³ However, the CCI did not attempt to distinguish between these varying conclusions nor independently counter Google’s arguments on confusion and impossibility. The CCI, in fact, acknowledged that many of the allegations against Google, including those related to OneBoxes and Universal Results, lacked merit, primarily because they contributed to innovation and enhanced user experiences, even if this came at the cost of competitors receiving lower rankings.

Consequently, the CCI’s approach reflects elements of both the Harvard and Chicago schools of thought (reflected in the US judgment which will be explained below). Pursuant to the Harvard School’s influence it condemned Google Flights for diverting traffic from competitors, leading to market foreclosure irrespective of consumer benefits. Whereas on the other hand, it dismissed allegations against features like OneBoxes and Universal Results because it enhanced user experience and innovation. This is in alignment with the Chicago School of thought which focuses on consumer benefit even if it results in foreclosure of competition. Hence, this results in an inconsistency in the CCI’s judgment.

The above inconsistency could be due to several underlying factors. Firstly, unlike Europe, no empirical analysis was done by CCI to find out the dependence of third party flight services on Google’s search engine or how the product was an improvement despite exclusion of competitors. Such empirical analysis was necessary for reaching a definitive conclusion. Secondly, economic considerations may have also led to this stance. If CCI would have ruled completely against Google then it could have discouraged foreign investment in the technological sector which is rapidly developing. Hence, it may have opted to acknowledge the benefits of Universal Results and OneBoxes. Lastly, the question of whether self-preferencing is inherently anti-competitive remains unsettled worldwide (Caminade et al., 2022). Hence, the divergent approaches with respect to self-preferencing worldwide made it difficult for CCI to settle on a definitive conclusion. Overall, this suggests that, the inconsistency was not only due to the interpretation of the law but also due to broader, institution and policy related challenges.

The US Position

The United States v. Google case (Federal Trade Commission, 2013) is a testament to the application of the Chicago School’s principles in antitrust enforcement by the FTC. This case reaffirms the traditional Chicago School doctrines that “shuns over deterrence, trusts unregulated market forces and attempts to maximize consumer efficiency” (Pyatt, 2023).

The FTC investigated allegations relating to Google unfairly preferencing its own content on the Google search results page and selectively demoting its competitor’s content from those results. FTC held that “design changes to the algorithm improved the quality of search results and the negative impact on actual or potential competitors was incidental to that purpose” (Federal Trade Commission, 2013). Thus, FTC was of the opinion that it would not be wise to second guess a firm’s product design decision where there could be pro-competitive effects.

The ‘consumer welfare’ standard was adopted in this case and the overall strategy of Google was scrutinized to ultimately conclude that “Google’s primary goal in introducing this content was to quickly answer, and better satisfy, its users’ search queries” (Federal Trade Commission, 2013). Therefore, the regulator focused on whether Google’s actions ultimately benefited consumers by providing them with better search results and by putting consumer welfare at the forefront of its investigation.

Comparison Between Jurisdictions: The Debate of Pro-competitive Justifications and Theory of Harm

The debate over pro-competitive or anti-competitive justifications and the theory of harm in the context of self-preferencing reveals a fundamental difference between the effects-based approach embraced by the US in line with the Chicago school and the per se approach adopted by the EU. While India has taken an inconsistent approach. ²⁴ This dichotomy has the potential to overshadow the pro-competitive benefits and hinder innovation and competition. To effectively assess self-preferencing claims, it is essential to distinguish between pro-competitive and anti-competitive biases. Google contends that its conduct may harm competitors but not competition or consumers. ²⁵ The crucial missing component is a clear harm theory, as this conduct can enhance both consumer welfare and competitive conditions. The core of the debate lies in whether Google’s conduct promotes monopolization through unlawful means or constitutes meritorious rivalry based on superior product quality. Impact on consumer welfare and competitive structure are used as the two parameters for assessing the pro-competitive or anti-competitive justifications for Google’s conduct (Singh, 2018).

Objective Justifications and Pro-competitive Benefits

In EU law, there is an ongoing debate regarding whether justifications for self-preferencing should be considered by the Commission before deeming it an abuse, rather than treating them as a defence (Akman, 2009). Presently, the burden of proving an objective justification, supported by all necessary evidence, rests with the dominant undertaking, which in this case is Google. ²⁶

Conversely, the CCI faced the initial task of exploring the potential inclusion of pro-competitive justifications under Section 4 of the Act. Section 4(2)(a) explicitly states that imposing discriminatory conditions “to meet competition” does not amount to an abuse of a dominant position. This provision allows for pro-competitive justifications. However, it is unclear that whether this encompasses the concept of “consumer welfare” or remains limited to an analysis of competitive structure merited examination by the CCI.

On the other hand, Sections 4(2)(b) and 4 (2)(c) of the Competition Act, 2002, the provisions under which charges were brought against Google’s search practices, do not explicitly provide for pro-competitive effects as valid justifications for the actions. Therefore, the CCI had to determine the relevance of pro-competitive justifications in this specific case and delve into the substance of such justifications. There appears to be a lack of comprehensive discussion in the CCI order, ²⁷ regarding the impact of Google’s conduct on consumer welfare and competitive structure. Although explicit discussions on these aspects are absent, it could be inferred that these two concepts were indirectly addressed through proxy measures (Singh, 2018). It seems that the “relevance of results” or “misleading consumers” were employed as parameters to assess consumer welfare, while bias was used as a method to evaluate the impact on competitive structure. However, reliance on these proxies for discussing the objectives of competition falls short of providing a thorough exploration of these critical aspects.

Consumer Welfare and Innovation

In the case of Google Search, given the inherent nature of the search engine business model — listing and ranking results according to a specific algorithm, it can be argued that if certain websites are not displayed or ranked in specific ways due to Google’s algorithm, this can be justified. This justification lies in a search engine’s business model, as not every site can be displayed equally in the results due to technical and practical constraints (Akman, 2017). Consequently, Google’s conduct can satisfy the requirements of “being objectively necessary and also generating efficiencies due to the nature of the business model of a search engine”. In fact, FTC found that Google’s “prominent display of its own vertical search results aimed at quickly satisfying users’ search queries by providing directly relevant information” (Federal Trade Commission, 2013), as stated above. This position aligns with the view that enhancing the quality of goods is an example of increasing efficiency that can satisfy the objective justification defence under Article 102 of the TFEU.

This perspective is supported by evidence from the FTC, indicating that Google consistently tested and monitored the impact of “introducing specialist (vertical) content on the quality of its general search results and adjusted its display when necessary to improve the user experience” (Federal Trade Commission, 2013). The German Monopolies Commission (2015) shares a similar view. For them, the inclusion of specialized services in horizontal search results constitutes a refinement of the search platform, leading to product innovation. This not only makes the platform more appealing to users but also provides added value by displaying relevant content directly, eliminating the need for inconvenient additional searches on other websites (German Monopolies Commission, 2015).

An empirical research on Google’s entry into the photography apps market on its own android platform suggests that this entry caused a “9.6% increase in the likelihood of major updates for apps affected by Google’s entry, compared to similar but not affected apps” (Foerderer et al., 2018). Furthermore, the research also suggests that the increased consumer attention for photography apps triggered complementary innovation instead of anti-competitive effects. An analogy can also be drawn with the retail market, where retailers sell their own private label brand along with other brands at their store and tend to keep their private label more visible (Long, 2022). As per a study, the prices of these private label product tends to be approximately 21.6% lower than other brands which are sold by the retailer (Call, 1967). This provided cost effective solution to the customers and also fostered overall economic welfare (Long, 2022). In response to the competition other private label products and brands then try to improve their product for an increased profit and sales, leading to innovation (Long, 2022).

The EC’s approach to the Google Search case can be critiqued for potentially protecting competitors at the expense of innovation and competition. It is important to note that there is almost near consensus now that the objective of competition law is not to protect competitors but to protect competition and consumer interests (Akman, 2017). ²⁸ Conversely, the CCI embodies a principle of non-interventionism. The following excerpts from the judgement indicate this:

“In view of the above, the Commission is cognizant of the fact that any intervention in technology markets has to be carefully crafted lest it stifles innovation and denies consumers the benefits that such innovation can offer. This can have a detrimental effect on economic welfare and economic growth, particularly in countries relying on high growth such as India.” ²⁹

This principle resonates throughout the decision and ultimately results in the CCI deciding to not intervene in Universal Results and only finding abuse on one count. However, such non-interventionist approach is not based on an actual finding of the conduct amounting to product improvement or consumer welfare but is based on the potential of restricting innovation.

Moreover, Akman (2017) ³⁰ argues that the Google’s conduct does not eliminate effective competition. Merchant platform sites like Amazon, retailer sites, and sites of retailers offering products in a given search are present in Google’s organic results, along with relevant sponsored links from sites that have purchased advertisements. All these sites are accessible by visiting their websites without conducting a search on Google, which further supports the conclusion that effective competition is maintained. In UK court proceedings, Streetmap, a provider of an online mapping service who complained about Google giving unfair advantage to Google Maps in its search results, lost the case also because it was not able to convince the court that it would have been commercially viable if Google had not advantaged its own mapping service. ³¹

The EC decision as upheld by the General Court and CCI decision recognize that “the Commission does not need to prove that the conduct has the actual effect of causing certain competing comparison shopping services to cease operating; it is sufficient to demonstrate that the conduct is capable of having such an effect”, ³² thus ignoring the pro-competitive benefits to consumers and seeks to protect competitors. The main focus of the EC was that there was unequal treatment of competitors, while not focusing on actual consumer harm.

Google Android Case

The following section first analyses the EU Android decision and the recent similar decision passed by the CCI. It then highlights how these courts dismissed the objective justifications given by Google, the defendant, due to the high burden of proof placed upon it. Furthermore, a comparison is drawn between analysis of consumer harm that is done by the US courts and the restricted approach adopted by the EU and Indian authorities. In conclusion, it is argued that the courts’ relatively narrow examination of the pro-competitive advantages and consumer well-being in the Google cases of the EU and India might potentially detrimentally impact consumers and hamper innovation.

The EU Google Android Case

In 2018, the EC fined Google 4.343 billion euros for violating Article 102 of the TFEU and Article 54 of the Agreement on the European Economic Area (EEA), which prohibit the abuse of a dominant market position. ³³ This case reaffirmed by the General Court ³⁴ sets a significant precedent for other countries to follow (Höppner & Westerhoff, 2022).

Anti-Competitive Agreements

The commission based on its investigation concluded that, Google imposed three types of restrictions on OEMs ³⁵ –

(1) Google compelled OEMs to pre-install Google Chrome and search apps in exchange for a license to use its ‘Play Store' through Mobile Application Distribution Agreements (MADAs). This created a “status quo” bias in favour of Google, as users rarely changed their default apps.

(2) Google’s “anti-fragmentation agreements” (AFAs) required OEMs to use Google Play Store and search apps only on devices running the original Android OS and not on devices which were run on Android Fork. This aimed to reduce the risk of Android platform fragmentation as it was open source, in turn making it the most closed open-source platform.

(3) Revenue Sharing Agreements (RSAs) involved sharing advertising revenue with OEMs, contingent on not pre-installing competing search apps. The RSAs, along with MADAs, had significant anti-competitive consequences. The object of the RSAs was to prevent other competitors from entering the market.

Hence, through these tying practices Google ensured that there was a preinstallation of the search engine and browser on practically all Google Android devices, thereby foreclosing competition. ³⁶ Further, the exclusivity payments strongly reduced the incentive to pre-install competing search engines. ³⁷

Abuse of Dominance

Google was held to be dominant in the relevant markets of app stores, market for licensable OS and the market for general search services with a share of 90%, 95% and 90% respectively in each of these markets (European Commission, 2018). Google’s argument that it faced competitive constraints from Apple’s IOS was rejected on the fact that the Apple IOS market was different due to its vertical integration. Third-party mobile manufacturers cannot obtain license for the same. ³⁸ Also, the switching costs for the customers are high.

Hence, as per the court, this was not a case of pure leveraging, here the company was trying to create a protective moat around the already existing dominant position in the general search application market which generated the maximum advertising revenue for google (Persch, 2022). Hence, a fine of 4.125 billion was upheld by the General Court and Google was required to “stop and to not re-engage in any of the three types of practices” or do anything else that might have the same effect. ³⁹

Consideration of Objective Justifications by the EC

Once the conditions of abuse according to Article 102 of TFEU are met, the burden of proof is on the dominant entity to demonstrate the existence of any objective justification. ⁴⁰ Google contended that the AFA agreements, which allowed GMS to be installed only on Android OS devices and not on any other devices was to ensure that there is quality and uniformity in the consumer experience and correct functioning of its proprietary apps. The EC acknowledged that the said justification was fair and reasonable, but this reason did not justify Google’s interference with the freedom of OEMs to sell devices based on Android forks that do not pre-install Google proprietary apps. ⁴¹

Secondly, Google argued that tying of google chrome and google search with Play Store is justified because it allows Google to monetise its investments in Android and its non-revenue-generating apps. This justification was dismissed by the EC because Google had enough revenues from Play Store are enough to monetize its investments. Also, it would still be able to benefit from the user data without such tying. ⁴²

Other justification provided by Google was that this arrangement also allows the company to license Play Store for free to OEMs as the amount of the license fee could be directly co-related to the amount that Google will have to pay for asking the OEMs to promote its products. Otherwise, the company will have to charge an upfront uniform license fee which would be disproportionate for the lower-end devices. The EC dismissed this argument on the ground that, it is a general practice for app stores to provide free services to OEMs and generate revenue through app downloads by customers. Further it also observed that Google could set lower license fees for lower-end devices, instead of setting the same fixed fee for all devices. ⁴³

In addition to this the company also claimed that such tying facilitates better consumer experience and allows it to compete with Apple. The commission dismissed this claim on the view that the consumer would benefit from the installation of any general search engine, it does not necessarily have to be Google. On the contrary, users would benefit if the device manufacturers had the flexibility to pre-install competitive products for all their devices. This would allow them to have different choices of devices. ⁴⁴

Hence, from the above reasoning of the EU Court it can be observed that the commission focused on the effect that Google’s conduct would have on the competition, competitors, and consumer choice rather than the possible effect that it could have on the increase in device costs due to the imposition of a license fee.

The Indian Scenario

Comparing With the EU Judgement

In 2022, a similar judgement was passed by the CCI in India. ⁴⁵ However, the scope of the judgement is much wider than the European Counterpart as the latter was concerned was only with google search application. On the other hand, in India, Google’s conduct in the online video hosting platform (OVHP) market through YouTube and other applications in the GMS bundle were challenged. Google was held liable for abuse of its dominant position pursuant to Section 4(2)(a)(i), Section 4(2)(c) to Section 4(2)(e) and Section 4(2)(b)(i) of the Competition Act, 2002 (the Act). A fine of Rs. 1337 Crore was imposed. It was noted that Google controlled 98% of the digital smartphone market in the country. The commission reached the same conclusion as the EU Court as stated above with respect to the anti-competitive agreements.

Consequently, along with the penalty a cease-and-desist order was passed against Google prohibiting it from entering the above anti-competitive agreements. The company had to guarantee that users could freely uninstall pre-installed apps without any restrictions. Additionally, users should have the option to customize their default settings on their devices. Furthermore, Google was obligated to enable developers to offer their app stores through the Play Store and not limit developers’ ability to distribute their apps through sideloading.

The remedies in this case went beyond the EC decision in 2018. Google expressed apprehension regarding India’s Android decision because the directives appeared to have broader implications compared to the ones imposed in the EC’s significant 2018 ruling against the operating system (Chaturvedi, 2023). Google has emphasized that no other jurisdiction has previously requested such extensive alterations, and it has consistently asserted that the expansion of its Android ecosystem in India will be hindered by this decision.

The NCLAT Decision

The decision was appealed to the NCLAT, which upheld the CCI’s ruling, although it overturned 4 out of the 10 directives issued by the CCI. One of the overturned directives pertained to the requirement of ensuring users could uninstall applications. ⁴⁶ The NCLAT contended that this requirement was unnecessary because users had the option to disable the applications. However, this contradicts the CCI’s reasoning where it had determined that the “mandatory pre-installation of the entire GMS suite under MADA (with no option to uninstall the same) and their prominent placement amounts to imposition of an unfair condition on the device manufacturers, thereby contravening the provisions of Section 4(2)(a)(i) of the Act.” ⁴⁷ In this context, the inability of users to uninstall the applications also perpetuates a status quo bias, a primary reason for considering MADAs anti-competitive. The NCLAT’s assertion that users could disable these apps challenges the argument of a status quo bias.

Further, it is important to note that NCLAT concluded that here, though the challenge was under Section 4 of the Act, the appreciable adverse effect on the competition was to be established as the abuse was exclusionary in nature. It noted that, “Commission is obliged to carry out effect analysis to the extent as to whether the abuse of dominant position is anti-competitive or harms the competitor. As noted above, learned ASG has submitted that the Commission has carried out effect analysis and found that the contract of the Appellant is anti-competitive and harmful to the competitors and such requirement is also complete.” ⁴⁸

Thus, it can be asserted that the tribunal upheld the CCI’s findings that Google’s behaviour imposed unfair conditions on the OEMs, leading to competition foreclosure. CCI dismissed Google’s argument that the OEMs entered into these agreements voluntarily, emphasizing that the OEMs had limited bargaining power. With no viable alternative to accepting the terms and conditions of the MADA, as omitting Google apps would reduce marketability, the CCI concluded that Google not only restricted the market for competitors but also curtailed consumer choice. Therefore, the focus of the findings remained on the impact of Google’s conduct on competing app store providers, search engines, and Android forks, i.e., the competitors. While consumer choice was considered, the analysis did not delve deeply into the potential effects on prices or innovation.

Comparing with the US

A preliminary examination of Article 102 of the TFEU, which resembles Section 4 of the Act, does not necessitate the proof of any adverse impact on competition when a dominant entity engages in an abusive tying agreement. If a dominant entity is found to be involved in any such abusive conduct, it can be inferred that it results in the stifling of competition. However, there has been a shift in the EU as well as the Indian jurisprudence where the commission is required to prove anti-competitive effect in case of an exclusionary abuse, but the same is not true for an exploitative abuse (Clark, 2017). The courts do consider the objective justifications given by the defendants, however generally the analysis done by the commission in this regard is accepted. Further as observed in the EU Google Android case, the burden of proof is on the defendant to prove any objective justification (Clark, 2017).

However, in U.S., there is a requirement of proving direct harm to the consumers either due to increased prices or reduced outputs in the relevant markets. The U.S. Supreme court has on several instances concluded that the proof of consumer harm is a necessary element for cases bought under Section 1 and Section 2 of the Sherman Act. ⁴⁹ The rule or reason standard in the U.S. requires the commission to weigh the procompetitive justifications against the anticompetitive effects. Mere evidence of competitors being harmed is not sufficient. Hence the burden of proof lies on the plaintiff (Clark, 2017). This is because, as per the Chicago school of thought, consumer welfare is the primary reason for the enactment and enforcement of competition law (Pyatt, 2023).

The decision given by the Department of Justice (DOJ) in the 1998 Microsoft judgement was highly criticized and later overturned by the Court of Appeals as it did not adhere to the law laid down regarding the requirement of analysing consumer harm. In the Microsoft case; the government had contended that, “the presumption of Antitrust policy is that competition itself brings consumer benefits, and the lessening of competition brings consumer harm. Hence, plaintiffs are required to show injury to competition rather than immediate harm to consumers” (Evans & Schmalensee, 2000). Consequently, the DOJ had concluded that Microsoft has tried to monopolize its position by bundling the Microsoft OS with its web browser i.e. internet explorer. This had led to the exclusion of players like Netscape and Opera from the market and hence hindered competition. The court did recognize several tangible benefits of Microsoft’s conduct like improvement in quality of the web-browser software, reduction in costs and increase in availability. The evidence of consumer benefit was much higher than the evidence which indicated any harm to consumer. No evidence was shown to as to how many people objected the presence of Internet explorer on their devices and how much it would have cost the consumers to buy a browser less version of Windows (Evans & Schmalensee, 2000).

Similarly, in 2007, the E.U. General Court had affirmed the commission’s finding that, “Microsoft has abused its dominant position by bundling Windows OS with Windows Media Player. Here, the court reasoned that the bundling business model of Microsoft led to foreclosure of competition because other media players could not offer their services on Windows OS. Further, the court concurred with the view of the commission that the refusal to offer a version of Microsoft OS without the media player restricted the manufacturers and users to install other media players of their choice and also foreclosed competition for other competing players in the market. Hence, the decision was based on harm to competitors and the consumer choice.

The U.S. Supreme court has emphasized that competition law exists for the benefit of consumers and hence the requirement of proving any harm to them is necessary. There is a possibility that the pro-competitive benefits may lead to consumer welfare and any restriction imposed on the dominant player may cause harm to consumers. Hence, the consumer welfare and harm analysis is necessary.

The Need for Analysing Consumer Harm in the Google Android Case

The Google case can be said to be a testament of exclusionary abuse and hence the NCLAT and the EU general court did conclude that an effect on the competition needs to be analysed and a “per se approach” was not sufficient. However, this analysis was merely limited to the effect on the consumer choice due to foreclosure of competition. Though while considering the objective justifications the EU courts did consider the effect on prices, the claims of the entity were dismissed without an in-depth analysis.

Firstly, the courts should have considered Google’s contentions that their conduct does not hamper consumer choice as the consumers are free to download other apps from the Play Store. An analysis of why consumers used Google and no other search engines or application which could be easily downloaded should have been made. Also, the OEMs are not coerced to install these applications. The apps also take negligible screen space and minimal storage space. Further, as reflected from the NCLAT decision consumers also have the option to disable these applications if they are pre-installed on the devices.

Even if it is argued that consumer choice is affected due to foreclosure of competition there is no actual consumer harm, rather the procompetitive benefits outweigh any harm that maybe caused to competition. Google offers all its services free of cost, promotes innovation by offering the android OS as an open-source platform and preserves consumer sovereignty. The current model provide these applications free of cost to the consumers as it increases their revenue from advertising. ⁵⁰ It allows the company to provide more efficient services, if the products are bundled together. Also, by restricting the creation of Android Forks through the AFAs it preserves the technical integrity and attractiveness of the Android ecosystem for the app developers (Bose, 2018). Hence, the procompetitive benefits should have been considered by the EU as well as the Indian courts.

Resultantly, this decision may have a negative impact on the consumers. If there is a decrease in revenue due to reduced traffic, then the company will start charging a license fee from the OEMs for providing the Android OS. This would increase the cost of the mobile devices and tablets (Bose, 2018). After the 2018 EU decision, the company did start charging a license fee from the OEMs for the provision of their applications (Kastrenakes & Patel, 2018).

Dimensions of Consumer Harm in Digital Markets

The traditional dimensions of consumers and consumer needs to be analysed differently for digital markets. Consumers may include both direct and indirect consumers i.e. wholesalers, retailers and final consumers. Other factors such as, data privacy, user profiling, tracking services, etc. would also directly affect consumer welfare will also have to be accounted for (Ezrachi, 2018). In all the above cases analysis on data privacy, tacking services and user profiling is insufficient. It can be argued that the services provided by Google are not actually free (Bet et al., 2022). The data provided by the consumers to Google could constitute the price that they must pay. Based on the data extracted, the company constructs an exclusive profile of consumers. It then sells this information to advertisers which enables them to target the consumers who are most likely to buy their products. If the consumer is aware about this extraction, then the price paid is zero. However, as per some reports not all consumers are aware about the nature and amount of data that is being collected (Bet et al., 2022). As a consequence of which they are not aware about the price they are paying, his could have a negative effect on the users and effect the welfare of the consumers who value their private information.

An important example illustrating the intersection between privacy law and competition law is the Facebook case adjudicated by the German Competition Authority (Bundeskartellamt). ⁵¹ In this case, Facebook was found to be dominant, and it was held that it has abused its position by obtaining uniformed consent from the users with respect to their personal data. The Bundeskartellamt treated this as an instance of exploitative abuse of market power, emphasizing that the exploitation stemmed from imposing unfair trading conditions on users who lacked meaningful choice.

When the case was referred to the Court of Justice of the European Union (CJEU), it confirmed that competition law authorities may examine infringements of privacy law when assessing abuse of dominance. ⁵² However, the CJEU clarified that privacy law can intervene only when the conduct leads to distortion of competition. Thus, privacy infringements alone, without harm to competition is not enough to invoke competition law enforcement.

Dr. Arletta Goreckac, in her PhD thesis, advocates for encompassing non-price elements like privacy in EU competition law. Referring to the Facebook case mentioned above, she argues that privacy related harms can be read into Article 102 of the TFEU (Gorecka, 2024). The broad and flexible language of Article 102 permits the inclusion of privacy as a dimension of consumer welfare. Provided that, the conduct of the entity should lead to unfair trading conditions or market foreclosure. However, on the other hand she also identifies risks such as regulatory overreach by competition enforcement agencies and the challenge that competition law is not designed to enforce data protection rules. Hence, while privacy can indeed be viewed as a non-price parameter of competition, Dr. Gorecka emphasizes the necessity of maintaining a careful balance to avoid overextending competition law into general privacy law enforcement.

Furthermore, blending privacy law with competition law enforcement can also create unintended consequences. Entities may attempt to invoke privacy justifications to shield potentially anti-competitive conduct (Colangelo, 2023). A prominent example is Apple’s App Tracking Transparency (ATT) policy introduced with iOS 14.5. As per the ATT Policy, to address privacy concerns, applications are required to obtain the consent of the users before tracking their activities across other apps and websites. However, the prompts are differently framed for third party applications vis a vis the apps developed by Apple. Apple’s own apps present consent prompts emphasizing the benefits of personalized services, whereas prompts for third-party apps focus on “tracking,” often with more negative connotations and without explanation of potential benefits (Colangelo, 2023). Even when users grant consent, third party developers face technical and procedural hurdles preventing them from accessing data necessary to personalize ads or measure their effectiveness across apps (Colangelo, 2023). These constraints are not applicable to Apple. This showcases that how differential application of privacy policies can be leveraged by dominant market entities. Apple’s ATT framework is currently under investigation by antitrust authorities in France, Germany, Italy, Poland, and has been scrutinized by the UK Competition and Markets Authority, all of which suspect that privacy concerns might be serving as a pretext for anti-competitive self-preferencing (Colangelo, 2023). Hence, though intersection of competition law and privacy is possible; there are persisting challenges which have to be addressed.

Conclusion

The cases involving Google’s alleged abuse of dominant position, specifically the Google Search Bias case and the Google Android case, have sparked significant debates and resulted in diverse outcomes in different jurisdictions, including the EU, India, and the United States. The EU’s emphasis on the foreclosure theory of harm, without demanding proof of actual anti-competitive effects, reflects a protective stance toward competitors. In contrast, the United States aligns with the Chicago School’s principles, prioritizing consumer welfare and allowing Google to demonstrate pro-competitive benefits through its product design decisions. India's approach in the Google-Android case was similar to the EU approach. However, in terms of self-preferencing (Google Search Bias case), India’s approach appears inconsistent, oscillating between favouring Google and identifying abuse of dominance, incorporating elements of both the Harvard and Chicago schools.

These divergent approaches highlight the challenge of balancing the protection of competitors and the fostering of innovation. To navigate this complexity, it is essential to define a clear harm theory and differentiate between pro-competitive and anti-competitive biases. The paper underscores the potential risk of overshadowing the pro-competitive benefits and hindering innovation and competition. The global conversation surrounding these cases underscores the evolving nature of antitrust law in the digital age and the need for a balanced approach that comprehensively considers the nuances of the tech industry. Ultimately, it emphasizes the importance of safeguarding both consumer welfare and competitive conditions in antitrust enforcement.