After outlining the relevant clauses of India’s Competition Act and its associated implementing regulation on predatory pricing, this article recounts how the Competition Commission of India has substantially modernised its perspective in just over a decade of enforcement. In an early case, it condemned zero-pricing by a financially powerful firm as an abuse of a dominant position, even though two others had successfully entered and reduced its market share to a third. In more recent cases, the Commission has acknowledged the legitimacy of below-cost pricing strategies to exploit network effects in platform competition, recognizing the resulting efficiencies and the possibility of vigorous competition even with a small number of competitors. This more permissive approach may, however, be reversed if a novel interpretation of the law by India’s Supreme Court proves decisive in several pending appeals.
By Aditya Bhattacharjea1
I. THE LEGAL FRAMEWORK
India’s Competition Act, 20022 explicitly includes predatory pricing under Section 4, which deals with the “Abuse of a Dominant Position.” Definitions of dominance and predatory pricing, which evolved through judicial interpretation in European and U.S. competition law, are carved into the Indian law itself.
In wording that seems to have been inspired by the United Brands judgment of the European Court of Justice, an “Explanation” appended to Section 4 defines a dominant position as “a position of strength, enjoyed by an enterprise, in the relevant market, in India, which enables it to (i) operate independently of competitive forces prevailing in the relevant market; or (ii) affect its competitors in the relevant market in its favour.” We shall see that this definition has recently been decisive in a landmark verdict by India’s Supreme Court.
Section 4 goes on to list some specific types of abuse, including imposition of an “unfair or discriminatory” price “in purchase or sale (including predatory price) of goods or service.” A further “Explanation” defines predatory price as “a price, which is below the cost, as may be determined by regulations, of production of the goods or provision of services, with a view to reduce competition or eliminate the competitors.”
The associated Regulations specify that the cost of production would be “average variable cost, as a proxy for marginal cost,” but with a proviso “that in specific cases, for reasons to be recorded in writing, the Commission may, depending on the nature of the industry, market and technology used, consider any other relevant cost concept such as avoidable cost, long run average incremental cost, market value,” each of which is defined in the Regulations.
In essence, then, India’s Competition Act requires a three-pronged test for predatory pricing: the firm must be dominant in the relevant market, the price must be below an appropriate measure of costs, and it must be set with a view to reduce or eliminate competition. The Act also codifies other criteria that are usually left to courts in mature jurisdictions. Section 19 specifies multiple “factors,” any or all of which the Commission “shall have due regard to” while investigating whether an enterprise is dominant, and while determining the relevant geographic and product markets.
If it veers towards excessive codification in some respects, the Act is egregiously deficient in another: unlike the adjacent sections on anti-competitive agreements and mergers, it does not require an effects-based test for any of the prohibited behaviors specified in Section 4. This allowed the Competition Commission of India (“CCI”) to apply a form-based assessment and find them abusive per se.3 Fortunately, it has not always gone down that road, and has followed the EU in increasingly adopting an effects-based approach.4 This is illustrated by the jurisprudence on predatory pricing reviewed below
But first, a brief outline of procedures under the Competition Act might be helpful. The Commission can inquire into an alleged contravention of Section 4 either “on its own motion” or based on complaints from private parties (known as “Informants”), government departments or statutory bodies. If it believes that there is a prima facie case, it instructs its Director General (“DG”) to undertake an investigation, otherwise it closes the matter.
The DG’s report, containing both findings of fact as well as conclusions of law, is shared with the parties, and after reviewing their comments, the Commission can either close the matter (if the report found no contravention), or proceed with the inquiry. If the inquiry ultimately finds that Section 4 has been violated, then the concerned enterprise can be ordered to discontinue the abusive conduct, and pay a monetary penalty of up to ten per cent of its average turnover of the preceding three financial yearsfiscal years. Orders of the Commission (except those that initiate an investigation) can be appealed, first to an Appellate Tribunal and then to the Supreme Court. (Most of the CCI decisions reviewed below are at various stages in the appeals process.)
II. THE EARLY CASE LAW
The CCI’s very first predatory pricing case, filed soon after it began accepting complaints in 2009, immediately confronted the young agency with the complexities of assessing predation in a digital platform market.5 The informant, MCX Stock Exchange (“MCX”), alleged that the National Stock Exchange (“NSE”) was abusing its dominant position by not charging fees for its recently-launched trading platform for currency derivatives (“CDs”). This forced MCX, which entered the CD segment a few months later, to incur losses by likewise offering free services. The Commission found a prima facie case for an investigation by its DG.
The investigation confirmed NSE’s dominance in the broadly defined market of “stock market services in India,” and zero-pricing of its services in the CD segment, cross-subsidized by its earnings from other segments (debt, equity and futures and options) in which it was the overwhelmingly dominant platform and did charge fees. The DG’s report mentioned that the operation of a securities exchange involved network effects, as well as scale economies due to high fixed costs and low marginal costs. But the report argued that these characteristics had contributed to NSE’s dominant position by making “consumers” dependent on it, and that such effects would enhance the anti-competitive harm of its conduct.
The six-member Commission handed down a split decision. The four-member majority defined the relevant market more narrowly, as that for exchange-traded CDs. It held that, even though there were by then three exchanges with almost equal shares of this market, NSE occupied a dominant position by virtue of its accumulated overall financial strength, which gave it the ability to incur sustained losses. MCX, whose only earnings came from its CD platform, could not match it in terms of resources, experience, nationwide presence, and scale of operations. The majority did not buy the argument that NSE’s fee waivers amounted to promotional pricing in a “nascent” market, because the waivers had been continuing for nearly three years.
NSE had also argued that its variable costs were zero, and so zero prices were not predatory. The majority rejected this argument. It reiterated an argument in the DG’s report that some operating costs, including those on “advertising, promotional activities, clearing and settlement, conveyance, communication and insurance expenses” could constitute variable costs, and the report’s demonstration of how NSE’s total costs could be allocated pro-rata to the CD segment. Average costs could not therefore be zero.
However, even after this detailed discussion, the majority order brushed aside the issue of what the appropriate measure of costs could be in this context, and shied away from a finding of predatory pricing. Instead, it concluded that “the zero price policy of NSE in the relevant market is unfair. In this case, the conduct of zero pricing by the NSE is beyond the parameters of promotional or penetrative pricing. It can, in fact, be termed as annihilating or destructive pricing.”6 The majority order ventured deeper into dangerous territory when it scoffed at NSE’s argument that the mandate of the Commission was to protect competition rather than competitors:
While this may be an interesting and oft-used phrase, it is shorn of the practicality of competition regulation. Similarly, it is not possible to protect competition without in some way protecting the weaker competitor…. Competition in a market is afforded by competitors and harm to competition has to be assessed by evaluating harm to competitors or its consequential impact on consumers.7
Even more ominous was the way in which the majority dismissed NSE’s argument that there had been no harm to consumers. They pointed out that the Act required only that dominance and one or more of the conducts proscribed by Section 4 be established. “Once both are established, there is no statutory requirement to examine any other additional impact on competitors or consumers or the market.”8 NSE was ordered to desist from zero-pricing and to pay a penalty of 5 percent of the average of its turnover of the three preceding years.
The majority order mentioned network effects only while summarizing the DG’s report, but not in its own assessment. In contrast, the dissenting order by the two-member minority (including the only professional economist on the Commission) set out a detailed exposition of such effects, and their implications for pricing behavior. The dissenters also held that NSE was not dominant, because within two years of its entry into the CD segment, the entry of MCX and another exchange had reduced its market share from 100 percent to less than a third.
The dissenters asserted that allegations of predatory pricing must be analyzed carefully, in order to avoid false positives and chilling of competition. Significantly, they insisted on evidence that the alleged predator’s foregone profits could be recouped — a concept that the majority had completely ignored. Based on their analysis of pricing in a network industry, the infeasibility of recoupment in an industry which had seen entry despite zero pricing, and lack of any evidence of intent to reduce competition, the dissenters held that NSE’s pricing was not predatory.
NSE appealed against the Commission’s order to the Competition Appellate Tribunal (“COMPAT”), which defined the relevant market as that of all securities exchanges in India. NSE was indubitably dominant in this broader market as well as in the CD market, in which it could leverage the financial strength derived from its dominance of the other segments. The strength of the promoters of MCX was not relevant, because they constituted a distinct corporate entity.
The COMPAT held that NSE had failed to provide any evidence that its variable costs in the CD segment were zero, so its zero-pricing clearly satisfied the definition of a predatory price. The COMPAT did not address the issue of recoupment, nor did it pay heed to the NSE counsel’s pointing out that its imposition of fees in compliance with the Commission’s order had harmed consumers and halved trading volumes. NSE’s appeal was dismissed.9 It appealed to the Supreme Court, which stayed the penalty, but has not yet given its judgment.
The next three predatory pricing cases were dismissed at the threshold, with the Commission holding that there was no prima facie case. These decisions indicated that the dominance and pricing prongs could be applied in either order, and failing either one would lead to dismissal. In the first case, the Commission did not go into the issue of dominance, because the informant could not provide evidence of below-cost pricing.10 In another case, evidence was reviewed to find that the respondent was not dominant, but the Commission went on to hold that, even presuming dominance, the informant had provided no evidence of the respondent’s costs.11
The third case deserves slightly more elaboration. Reliance Jio, a part of the huge Reliance conglomerate, disrupted the Indian telecommunications sector by offering free voice, data, and messaging on its new 4G network, at a time when all incumbents offered mainly 2G or 3G services. In response to Jio’s rapid penetration of the sector, Bharti Airtel (the largest incumbent) filed a case of predatory pricing.
The Commission pointed out that Jio had a share of less than 7 percent of all wireless telecommunications subscribers, with the rest divided amongst several other established players, many of them large firms — including multinationals — with comparable reputations and financial strength. Jio was therefore not dominant, and its zero-pricing was not anti-competitive, but a short-term market penetration strategy.12
III. THE TAXI WARS
The final set of cases is perhaps the most important, for two reasons. First, they involved platforms which cater to two distinct sets of users, with the attendant problem of assessing two-sided pricing and platform dynamics. Second, an appeal to the Supreme Court resulted in an interpretation of Section 4 that turned earlier jurisprudence on its head.
Traditionally, as in other countries, taxis in India were owned by individuals and could be hired from a stand or on the road. From around 2007, a number of radio taxi companies began operating in the major Indian cities. They set up call centers which dispatched their own taxis to riders on the basis of bookings made via telephone or internet. From 2011, this business model was severely disrupted by the entry of taxi aggregators Ola and Uber, who do not own any vehicles but only provide a platform to connect owner-drivers with passengers, using smartphone-based apps to identify their locations.
Beginning in 2015, three radio taxi operators (Meru, Mega Cabs and Fast Track) whose business had been severely affected, filed multiple cases alleging predatory pricing by Uber and ANI Technologies (owners of the Ola brand) in different cities. Three auto-rickshaw drivers also filed individual cases against Ola, which also provided aggregation services for auto-rickshaws. The CCI clubbed some of these cases for consideration, resulting in five separate verdicts.13
Thus, dominance was not established, and the Commission did not have to decide on the question of predatory pricing. In four of the five cases, the Commission held that prima facie there had been no contravention of the Act; in the fifth, it ordered an investigation by its DG, and then gave a much more detailed order which arrived at the same conclusion, neatly summarized in the following paragraph:
[B]ased on collective consideration of the facts that the competitive process in the relevant market is unfolding, market is growing rapidly, effective entry has taken place thereby leading to gradual decline in [Ola’s] market share, entry barriers are not insurmountable, there exist countervailing market forces that constrain [its] behaviour… and the nature of competition in dynamic, innovation-driven markets, the Commission is of the considered view that [Ola’s] dominance in the relevant market remains unsubstantiated.14
In its earlier decisions, the Commission had required dominance to be established before addressing the allegation of predatory pricing, so the matter could have ended there. Nevertheless, the Commission concluded its order with a discussion of pricing strategy in platform markets characterized by network externalities, benefits to consumers, and equal access to venture capital funding for potential rivals. This perspective seems to suggest that, contrary to the Commission’s earlier MCX order, anti-competitive intent and effects should not be readily inferred in a platform market even if a dominant incumbent is shown to be pricing below costs.
This position ran into difficulties on appeal, leading to a completely new interpretation of the test for predatory pricing. Deciding on Meru’s appeal against one of the earlier CCI orders, the COMPAT questioned the Commission’s definition of the relevant geographical market, its reliance solely on market shares to determine dominance, and the contradictory evidence on market shares which should have called for further investigation. COMPAT remained agnostic on whether Uber’s pricing reflected “phenomenal efficiencies” or “an anti-competitive stance.”15 It ordered an investigation by the CCI’s Director General. This would probably not have fundamentally changed the jurisprudence, but the Supreme Court’s reasoning for upholding the COMPAT order when Uber appealed against it certainly did.
In a very brief verdict, the Court cited a single piece of evidence proffered by Meru: data on Uber’s dealings with four cars for less than a month. It showed that Uber was on average losing a substantial amount per trip by paying incentives to drivers. The Court referred to the definitions provided in the “Explanation” in Section 4 of the Act, according to which a dominant position would include “a position of strength enjoyed by an enterprise…, which enables it to … affect its competitors or consumers or the relevant market in its favour.” Losing money on each trip, the Court opined, would prima facie satisfy this definition, as well as the definition of predatory pricing.16
The immediate applicability of this order was limited to affirmation of the COMPAT’s finding of a prima facie case for investigation by the DG. But if the Supreme Court’s interpretation determines the outcome of this and other similar cases, the law on predatory pricing will be transformed beyond recognition. Thus far, the accepted position in all such cases has been that the dominance and pricing tests were to be applied independently, based on separate criteria laid down in the Act. Failing either test would result in dismissal. If complainants can now satisfy both tests by providing evidence of an isolated instance of below-cost pricing, the chilling effect on competition would be obvious.17
This denouement is not inevitable, however. The DG’s investigation went over much the same evidence as in the earlier case, looked at more recent developments, and arrived at the same conclusion: Ola was providing effective competition to Uber in the relevant market. He also reported that both had followed below-cost pricing strategies for many years, but absent a finding of dominance, this could not be regarded as a violation of Section 4. After hearing the parties, the Commission has very recently reiterated its earlier decision, restating the same arguments.
To reinforce its focus on the “strength of competitive constraints faced by players in a relevant market,” it even quoted its earlier order in a similar case, which is worth quoting here as well for its forthright stance:
[As] long as there is competition in and for the market satisfying these outcomes, regulatory intervention is not warranted to either protect the existing players or to increase the number of players in the market. Towards that end, Competition and competition law is not about counting the number of firms in a particular relevant market to determine whether or not that market is competitive [….] There can be markets which may not be competitive even with large number of players and equally possibly there can be markets which can work perfectly well with fewer players, constraining the conduct of each other. What is significant is that the existing firms are effective enough to constrain the behaviour of one another so as to dissuade independent abusive conduct by any of them.18
Therefore, the Commission concluded, Uber was not dominant. The matter could have ended there, but once again the Commission quoted its earlier order, this time on the importance of early-stage penetrative pricing to generate network effects. It went on to refer to the DG’s more recent evidence of below-cost pricing by both Ola and Uber initially, but their margins turning positive in the last few years—a pattern that the Commission accurately recognized as one “typical to a platform market operator seeking to reap on [sic] network effects i.e. initial low-cost pricing to create network/scale and once a viable scale is achieved, per unit profit.”19
Furthermore, even though Section 4 of the Act does not provide for an efficiency justification or any effects-based analysis, the Commission identified efficiencies in the reduced downtime for drivers and waiting times for consumers. With this commendably modern treatment of pricing in platform markets, the Commission closed the case. Clearly, it treated the Supreme Court’s judgment as a limited directive to investigate, and not the basis for adjudication on the merits. It will be interesting to see how the Court reacts when Uber’s appeal, and appeals on several cases now pending before the CCI that involve deep discounting by e-commerce platforms, inevitably reach it. Other issues that may get resolved include: whether consumer welfare trumps other considerations; whether predatory intent needs to be established independently; whether the possibility of “winner takes all” and recoupment of profits sacrificed during the predatory campaign is relevant; and the role of accumulated data as an entry barrier.
1 Delhi School of Economics, University of Delhi; firstname.lastname@example.org.
2 The enforcement sections of the Act were brought into effect only in 2009, along with the constitution of the full Competition Commission of India (“CCI”). The CCI is both an investigative and adjudicatory agency, so to avoid confusion I refer to adjudicatory decisions as those of the “Commission.”
3 This was presciently pointed out before the Act came into force, in Subhadip Ghosh & Thomas Ross, India’s New Competition Law: A Canadian Perspective 23 Canadian Competition Record 23 (2008).
4 See Payal Malik et al, Legal Treatment of Abuse of Dominance in Indian Competition Law: Adopting an Effects-Based Approach, 54 Rev Industrial Organization 435 (2019).
5 MCX Stock Exchange Ltd. v. National Stock Exchange of India Ltd. and Anr., Case No. 13/2009, order dated June 23, 2011. (CCI decisions — known as “orders” — are accessible from the official website, www.cci.gov.in.).
6 Id. paras 10.76-10.77, emphasis in the original.
7 Id. paras 24.3-23.4.
8 Id. para 25.1.
9 National Stock Exchange of India Ltd. v. Competition Commission of India and Anr., Appeal No. 15 of 2011 with IA Nos. 25/2011, 26/2011, 27/2011, 10/2012, 27/2012, order dated August 5, 2014.
10 Case No. 80/2012, HLS Asia Limited v. Schlumberger Asia Services Ltd., order dated February 6, 2013.
11 Case No. 09/2013, Transparent Energy Systems Pvt. Ltd. v. TECPRO Systems Ltd, Case 09/2013, order dated June 11, 2013.
12 Case No. 03/17, Bharti Airtel Ltd v. Reliance Industries Ltd and Reliance Jio Infocomm Ltd, order dated June 9, 2017.
13 Cases 81/2015, Meru Travel Solutions Pvt Ltd v. Uber India Systems Ltd, order dated December 22, 2015; 82/2015, Mega Cabs Pvt Ltd v. ANI Technologies Pvt Ltd, order dated February 9, 2016; 96/2105, Meru Travel Solutions Pvt Ltd v. Uber India Systems Pvt Ltd and Others, order dated February 10, 2016; 21/2016, Mr Vilakshan Kumar Yadav and Ors. v. ANI Technologies Pvt Ltd, order dated August 31, 2016; 6&74/2015, In Re: Fast Track Call Pvt Ltd and Meru Travel Solutions Pvt Ltd v. ANI Technologies Pvt Ltd., order dated July 19, 2017.
14 Fast Track Call (supra note 11), para 105.
15 Meru Travel Solutions Pvt Ltd v. Competition Commission of India, Uber India Systems Pvt Ltd and Ors., Appeal 31/2016, order dated December 7, 2016, at para 18.
16 Civil Appeal No. 64/2017, Uber (India) Systems (P) Ltd. v. Competition Commission of India and Ors. (2019) 8 SCC 697, order dated September 3, 2018.
17 A more recent appellate decision points in the same direction. For reasons that are not relevant here, COMPAT was abolished in May 2017, and the National Company Law Appellate Tribunal (“NCLAT”) was empowered to decide on appeals against CCI orders. In November 2018, the CCI held that the e-commerce giants Amazon and Flipkart were not dominant in the market for services provided by marketplace platforms. The reasoning was similar to that in the taxi cases. But here too, NCLAT remanded the case for a DG investigation, solely on the basis of evidence (adduced in an unrelated income tax proceeding) that Flipkart was selling goods at a loss on its platform. Competition Appeal (AT) No.16 of 2019, All India Online Vendors Association v. CCI and Ors., order dated March 4, 2020. This order was subsequently stayed by the Supreme Court.
18 Case No. 96/2015, Meru Travel Solutions Pvt. Ltd. (Meru) v. Uber India Systems Pvt. Ltd., order dated July 14, 2021, para 128 (emphasis in the original), citing Fast Track Call, as note 11 supra. The Commission also dismissed other claims by Meru, including joint domination by Ola and Uber; the possibility of coordination between them because of common minority shareholding by institutional investors; and incentive schemes for drivers amounting to exclusivity agreements.
19 Id. para 147.