[By Anirud Raghav]
The author is a student at NLSIU, Bangalore.
Introduction
Even after two decades of competition jurisprudence, questions regarding the scope and applicability of s.3 of the Competition Act (hereinafter, “the Act”) persist. Briefly put, s.3 prohibits anti-competitive agreements, and is an indispensable feature of global competition jurisprudence (see Art.101 TFEU, s.2 of UK Competition Act, 1998 inter alia). A key unresolved issue is whether s.3(1) can apply independently, or if it must be read in conjunction with s.3(3) and 3(4) dealing specifically with horizontal and vertical agreements respectively. This piece argues that allowing Section 3(1) to have standalone applicability is necessary to address increasingly complex business arrangements, especially in digital markets, that defy simple categorization as horizontal or vertical.
This post will consider the objections levelled against an expansive interpretation of s.3(1) and reject these objections as lacking any merit. It proposes that s.3(1) should be expansively interpreted and proposes potential safeguards while doing so to ensure that a balance is maintained between ease of doing business and preserving healthy competition.
CCI Decisional Practice: What is the Law?
While previous blogposts have provided a broad overview of the jurisprudence on the applicability of s.3(1), we will examine two seminal judgments proposing contrasting approaches to the question.
In Ramakant Kini v. Hiranandani Hospital, there was an exclusive supply agreement between Hiranandani Hospital and Cryobank Ltd. in respect of stem-cell banking services. Effectively, the agreement meant that customers of Hiranandani Hospital could only avail of stem cell banking services from Cryobank; it could not be independently provided by Hiranandani Hospital. This was challenged as being anti-competitive, as a vertical exclusive supply agreement under s.3(4)(b). The relevant holding can be found in paragraph 11 of the decision: “Section 3(3) and section 3(4) are expansion of section 3(1) but are not exhaustive of the scope of section 3(1)…Section 3(3) carves out only an area of section 3(1). The scope of section 3(1) is thus vast and has to be considered keeping in view the aims and objects of the Act…This makes it abundantly clear that scope of section 3(1) is independent of provision of section 3(3) & 3(4).”
From the above, it can be inferred that s.3(1) should be expansively interpreted considering the objectives of the Act, including consumer welfare and freedom of trade (see the Preamble of the Competition Act, 2002).
The other case is Alkem Laboratories Ltd. v. CCIi in which the Competition Appellate Tribunal holds the following: “Section 3(1) is the general sub-section which legally intended to cover the overall principles of breach of either Section 3(3) or 3(4) of the Competition Act. Conclusion of breach of Section 3(1) in the absence of findings relatable to breach of Section 3(4)(d) in this case against Alkem is bad in law.” The decision reasons that if indeed s.3(1) were legislatively intended to be independently applied, then there would be no need for s.3(3) and s.3(4), since the former has such a catch-all phrasing that it would in any case include agreements contemplated under s.3(3) and s.3(4) rendering them superfluous. This is an objection often reiterated in existing literature.
At first glance, both cases acknowledge that s.3(1) is general and vast in scope. They only differ in their theories of why s.3(1) is so broad. On the one hand, Ramakant’s take is that the breadth of s.3(1)’s scope is evidence that the legislative intent was to contemplate agreements beyond the horizontal-vertical dichotomy. On the other, Alkem Laboratories opines that s.3(1)’s breadth only means that it is a general section “legally intended to cover the overall principles” of s.3 – it is not meant to apply independently of s.3(3) and s.3(4), but rather in conjunction with them.
I argue that Alkem’s reasoning is suspect for two reasons. First, if all s.3 was ever meant to contemplate was horizontal and vertical agreements, then there would be no reason for having s.3(1) in the first place. This is because s.3 would remain internally cohesive even if s.3(1) were removed, or never existed to begin with. Second, CCI’s characterization of s.3(1) as a provision that merely mentions “overall principles” related to s.3 breach is unprecedented both in statute as well as CCI’s decisional practice. In any case, it is unclear why these principles would be required to begin with – all the potentially ambiguous terms used in s.3(3) and s.3(4) are defined or otherwise clarified in the statute. For instance, “agreement” is defined under s.2(b), “cartels” under s.2(c), “appreciable adverse effect on competition” (AAEC) under s.19(3) and so on. In other words, s.3(1) does not enumerate any principles as such, without which s.3(3) and s.3(4) would become impossible or otherwise difficult to interpret. This further renders Alkem’s reasoning untenable. In the following section, I will consider the same line of objection in greater detail.
Is the Alkem Laboratories Objection Sound?
The objection apparent from Alkem is that if we allow a standalone interpretation of s.3(1), the provision is so broad and catch-all that it will render s.3(3) and s.3(4) redundant. Proponents of this view might also rely on an argument from parliamentary intent to argue that the intention was to categorize anti-competitive agreements into horizontal and vertical agreements only. In this section, it is argued that it is incorrect to say that s.3(3) and s.3(4) become infructuous altogether.
S.3(3) and s.3(4) do not become altogether infructuous
It is undisputed that s.3(1) is expansive in its phrasing and sets out a framework that broadly covers horizontal and vertical agreements contemplated under s.3(3) and s.3(4). However, we must not hasten to say that this would, per se, render the s.3(3) and s.3(4) infructuous, for the latter two provisions differ from s.3(1) in two salient ways. They may be considered independently.
S.3(1) and s.3(3)
The difference between s.3(1) and s.3(3) may be considered first. To prove a case under s.3(1), the CCI has to show that an agreement causes or is likely to cause AAEC. Typically, AAEC is established with reference to factors enumerated under s.19(3). So, under s.3(1), the CCI has to prove that an anti-competitive agreement will, on a balance of factors mentioned under s.19(3), have a net-anti-competitive effect (this balancing approach was prescribed by the CCI in Shamsher Kataria).
Unlike this approach, if the CCI wants to argue that an agreement is anti-competitive under s.3(3), then it must merely establish one of the four factors mentioned from s.3(3)(a)-(d). Establishing this will suffice to raise a presumption of AAEC – albeit rebuttable. In other words, there is no need to independently establish AAEC under s.19(3) factors. More importantly, it is noteworthy that s.3(3)(c) & (d), which speak about market sharing and bid rigging respectively, are not a part of the 19(3) factors, even if 3(3)(a) & (b) broadly resemble 19(3) factors. This makes it very clear that s.3(3) was meant to prohibit very specific types of horizontal arrangements, which is unlike the broad and general nature of the AAEC factors mentioned under s.19(3).
S.3(1) and s.3(4)
As for the difference between s.3(1) and s.3(4), we may note that s.3(4) similarly outlines a list of specific forms of vertical arrangements that CCI needs to prove to make out a violation. It must then independently prove that these arrangements cause or is likely to cause AAEC. Once again, this is a two-step process that requires specifically identifying the character of the conduct, then establishing that it can lead to AAEC.
The above discussion shows that s.3(3), s.3(4) are not rendered pointless by the mere existence of s.3(1). Granted, s.3(1) is a broad provision which reflects the spirit of the below two provisions – to prevent adverse effect on competition. But s.3(1) is a far more general provision than the latter two, which stipulate specific circumstances under which horizontal and vertical agreements must be proven. Given that these provisions are, by design as well as operation distinct, it is incorrect to argue that they are rendered moot because of s.3(1).
Clarifications and Proposals
I would like to dispel a few concerns about my position. First, the idea is not that the horizontal-vertical framework be altogether abandoned. This article is merely advocating for widening the ambit of s.3(1), so that agreements having anti-competitive effects, but escaping the rigid technical dichotomy are not excused altogether. Secondly, one might indeed presage threat of overregulation if s.3(1) were subjected to a broader interpretation. Thus, if an expansive interpretation were to be preferred, safeguards are necessary to strike a balance between ease of doing business and protecting competition. Traditionally, competition regulation penalizes certain types of economic relationships – which is why the vertical-horizontal dimensions exist.
One potential safeguard could be to hold allegations of violation of merely s.3(1) to a higher standard of proving AAEC. This effectively renders s.3(1) as a fall-back provision, in the event that s.3(3) or s.3(4) could not be proved due to technicalities. Moreover, there is already precedent of this sort of differential standards of proof: for instance, s.3(3) and s.3(4) are already held to different standards of proof. As mentioned above, if merely one of the acts under s.3(3)(a)(d) is proven, AAEC is presumed; however, under s.3(4), AAEC must be established independently in addition to proving conduct under s.3(4)(a)-(e). Thus, by holding independent s.3(1) violation allegations to a higher standard of proof of AAEC would not necessarily deviate from the existing statutory scheme, and would help secure the objectives of the Act. To clarify further, “higher standard of proof of AAEC” would illustratively mean requiring to concretely establish any or a combination of factors under s.19(3), and/or an instatement of a presumption of non-AAEC for independent s.3(1) allegations. The modalities of implementing the higher standard of proof may vary, but the underlying rationale for it is sound and consistent with the existing scheme. This promises to help us overcome the stranglehold of horizontal-vertical agreements, and grants CCI broader, yet restrained, powers to effectively regulate the rapidly changing business landscape today. This will only become more important in the nascent yet protean digital markets, where economic relationships are not as straightforward as traditional markets.
Conclusion: Digital Markets and the Way Forward
Emerging digital markets do not have straightforward economic relationships classifiable strictly as horizontal and vertical. Consider the example of an AI company incorporating cloud services from another company in its service delivery. Consumers, especially big businesses, would be naturally drawn to such a novel and innovative combination, given the critical importance of data protection and data storage. Thus, such an agreement, especially if it is the first of its kind, could considerably impose barriers to entry into the market, thereby producing AAEC as contemplated under s.19(3). However, the issue is that the agreement between the two companies is neither horizontal, because they are not between competitors, and nor are they vertical, since AI software delivery and cloud production are not strictly at different levels in the same production chain. Thus, such an agreement would straddle the rigid lines between horizontal and vertical and is unlikely to be clearly classified into one of the two. This would create problems for the competition regulator, considering that AAEC results, but the agreement would evade liability on account of not being strictly classifiable into either the horizontal or the vertical category. Furthermore, the recent 2023 amendment does not remedy this issue, since it is limited to plugging the void with regards to hub and spoke cartels. It merely adds a presumption that parties that are not directly engaged in similar trade may also be presumed to violate s.3(3) if they participate in furtherance of the agreement. It does not amend the framework to contemplate a situation where the agreement is neither horizontal nor vertical – rather, such a provision only applies if a prior horizontal agreement is established and seeks to also impose liability on the “spokes”. Thus, the foregoing issues continue to persist and prompt a reconsideration of the existing framework.