Definitions — Enterprise, Relevant Market, Cartel

Every contest under the Competition Act, 2002 is won or lost on definitions. Before the Commission can ask whether an agreement is anti-competitive under Section 3 or whether a firm has abused dominance under Section 4, it must first answer three threshold questions: Is the entity an enterprise within Section 2(h)? What is the relevant market under Section 2(r) read with the factors in Sections 19(6) and 19(7)? And, where collusion is alleged, does the conduct fit the inclusive definition of a cartel in Section 2(c)? Get these wrong and the substantive analysis collapses — as the Supreme Court repeatedly demonstrated in CCI v. Coordination Committee of Artists and Technicians, Excel Crop Care and Rajasthan Cylinders. This chapter dissects each definition, the statutory factors that flesh them out, and the case law that has given them shape.

Why the definitions are jurisdictional, not academic

The Competition Act is a regulatory statute whose operative prohibitions — Section 3 (anti-competitive agreements) and Section 4 (abuse of dominant position) — are switched on only when the actors and the arena fit the statutory definitions in Section 2. An entity that is not an "enterprise" is outside Section 4 altogether; an alleged abuse that cannot be located in a properly delineated "relevant market" cannot be tested for dominance under Section 19(4); and a coordination that does not answer the description of a "cartel" in Section 2(c) loses the presumption of appreciable adverse effect on competition (AAEC) that Section 3(3) attaches to horizontal arrangements. Definitions, in other words, are the gatekeepers of jurisdiction.

This is why the Supreme Court in Competition Commission of India v. Coordination Committee of Artists and Technicians of W.B. Film and Television, (2017) 5 SCC 17, spent the bulk of its reasoning not on the boycott itself but on whether the trade unions were "enterprises" and what the "relevant market" was — questions on which the Commission, the COMPAT and the Supreme Court each took a different view. The order in which the Act asks its questions is deliberate: first identify the player, then map the field, then characterise the conduct. This chapter follows that sequence and connects to the deeper treatment in our introduction to the Competition Act and the dominance-focused chapter on determining dominance through relevant market analysis.

Enterprise — Section 2(h): the functional test

Section 2(h) defines an "enterprise" expansively as a person or a department of the Government who or which is, or has been, engaged in any activity relating to the production, storage, supply, distribution, acquisition or control of articles or goods, or the provision of services of any kind, or in investment, or in the business of acquiring, holding, underwriting or dealing with shares, debentures or other securities of any other body corporate, either directly or through one or more of its units or divisions or subsidiaries. The definition reaches units, divisions and subsidiaries wherever located.

Two features dominate. First, the test is functional, not formal — what matters is the economic activity engaged in, not the legal label the entity wears. A trade union, a statutory body or a government department can be an enterprise to the extent it carries on economic activity. Second, the proviso carves out activities of the Government "relatable to the sovereign functions", expressly including the departments of the Central Government dealing with atomic energy, currency, defence and space. The carve-out is narrow: only primary, inalienable and non-delegable sovereign functions escape; commercial or welfare activities do not.

The locus classicus is Coordination Committee of Artists (2017), where the Supreme Court held that the film and television artists' associations, though styled as trade unions, were acting as "enterprises" when they coordinated a boycott of dubbed serials because they were engaged in economic activity affecting the market. The functional approach means the same body can be an enterprise for one activity and not for another — the Commission examines activities individually. The breadth of Section 2(h) also explains why public-sector undertakings are squarely covered, a point the Supreme Court reinforced in its competitive-neutrality jurisprudence.

"Person" and the outer reach of the enterprise concept

The word "person" inside Section 2(h) is itself defined in Section 2(l) to include an individual, a Hindu undivided family, a company, a firm, an association of persons or body of individuals (whether incorporated or not), any corporation established by or under any Central, State or Provincial Act, a body corporate incorporated outside India, a co-operative society, a local authority and every artificial juridical person. The combined effect of Sections 2(h) and 2(l) is that almost any economic actor — domestic or foreign, incorporated or not, public or private — can be an enterprise.

This breadth has practical consequences. Foreign entities are caught where their conduct has effects in India, dovetailing with the extraterritorial reach in Section 32. Associations of persons — trade bodies and industry associations — are routinely treated as enterprises or as facilitators of agreements between enterprises, which is precisely how the Cement Manufacturers' Association was implicated in the cement cartel matter discussed below. The expansive definition is the doctrinal foundation on which the substantive prohibitions in our chapters on anti-competitive agreements and abuse of dominant position rest.

Relevant market — Section 2(r): the indispensable backdrop

Section 2(r) defines "relevant market" to mean the market which may be determined by the Commission with reference to the relevant product market or the relevant geographic market or with reference to both the markets. The definition is deliberately framed as a power conferred on the Commission, signalling that market delineation is a fact-sensitive exercise, not a mechanical formula.

The relevant market is the analytical canvas for both dominance and effects. Under Section 4 read with Section 19(4), dominance is assessed within a relevant market; under Section 3, the AAEC inquiry under Section 19(3) presupposes a market in which the effect is felt. Get the market wrong — too narrow and dominance is inflated, too wide and it disappears — and the entire substantive conclusion is unsafe. That is exactly why the relevant-market finding has been the decisive battleground in MCX Stock Exchange v. National Stock Exchange, Coordination Committee of Artists, Uber India and CCI v. Schott Glass India. The two components — product and geographic — are defined separately in Sections 2(t) and 2(s) and are filled out by the statutory factor lists in Sections 19(7) and 19(6). We treat the mechanics of dominance in detail in the chapter on determining dominance and relevant market analysis.

Relevant product market — Section 2(t) and the Section 19(7) factors

Section 2(t) defines "relevant product market" to mean a market comprising all those products or services which are regarded as interchangeable or substitutable by the consumer, by reason of characteristics of the products or services, their prices and intended use. Substitutability — chiefly demand-side substitutability — is the organising idea: two products belong in the same market if buyers would readily switch between them.

Section 19(7) supplies the factors the Commission must have due regard to in delineating the product market: (a) physical characteristics or end-use of goods; (b) price of goods or service; (c) consumer preferences; (d) exclusion of in-house production; (e) existence of specialised producers; and (f) classification of industrial products. The list is illustrative, not cumulative — the Commission may rely on "all or any" of these factors.

The product market can be drawn remarkably narrowly where substitutes are absent. In Competition Commission of India v. Schott Glass India Pvt. Ltd., 2025 INSC 668 (Civil Appeal No. 5843 of 2014), the Supreme Court accepted two distinct upstream product markets — neutral glass clear (NGC) tubing and neutral glass amber (NGA) tubing — because they were not interchangeable for pharmaceutical packaging users, and found Schott dominant in both; even so, it held there was no abuse, underscoring that a correct market finding does not by itself establish liability. The same substitutability logic separates horizontal from vertical concerns, as explained in our chapters on horizontal agreements and vertical agreements and the rule of reason.

Relevant geographic market — Section 2(s) and the Section 19(6) factors

Section 2(s) defines "relevant geographic market" to mean a market comprising the area in which the conditions of competition for supply of goods or provision of services or demand of goods or services are distinctly homogenous and can be distinguished from the conditions prevailing in the neighbouring areas. Homogeneity of competitive conditions — and a clear boundary from neighbouring areas — is the touchstone.

Section 19(6) lists the geographic factors: (a) regulatory trade barriers; (b) local specification requirements; (c) national procurement policies; (d) adequate distribution facilities; (e) transport costs; (f) language; (g) consumer preferences; and (h) need for secure or regular supplies or rapid after-sales services. As with the product list, the Commission may rely on all or any of them.

Geographic delineation can be national or intensely local. In MCX Stock Exchange Ltd. v. National Stock Exchange of India Ltd. (CCI, Case No. 13/2009, order dated 23.06.2011), the Commission settled on the market for stock exchange services for the currency derivatives segment in India and found NSE dominant, penalising its zero-price policy as predatory. By contrast, in Uber India Systems Pvt. Ltd. v. Competition Commission of India (Supreme Court, decided 3 September 2019), the relevant market was the far narrower "radio taxi services in the Delhi-NCR", within which Uber's per-trip losses were treated as prima facie evidence of dominance and predatory pricing sufficient to order an investigation under Section 26(1). Geography, like product scope, is outcome-determinative.

Substitutability and the SSNIP test

Although the Act does not name it, the analytical engine behind market delineation is the hypothetical-monopolist or SSNIP test (Small but Significant Non-transitory Increase in Price, conventionally 5–10%). The question asked is whether a hypothetical monopolist of a candidate set of products in a candidate area could profitably impose such a price rise; if customers would switch to other products or sources so as to defeat the rise, those substitutes must be folded into the market until a price rise would be profitable. The narrowest set for which it would be profitable is the relevant market.

Indian practice treats SSNIP as a useful conceptual aid rather than a rigid arithmetical test, and the Section 19(6)–(7) factors are the statutory route by which substitutability is assessed. The test fits awkwardly where price is zero — a recurring problem in digital and platform markets, where free-to-user services cannot be tested by a price increase. The Commission has accordingly leaned on qualitative substitutability, consumer preferences and end-use under Section 19(7) rather than a literal SSNIP in such cases. The interplay of substitutability and dominance is developed further in the dominance chapter.

Cartel — Section 2(c): the inclusive definition

Section 2(c) provides that "cartel" includes an association of producers, sellers, distributors, traders or service providers who, by agreement amongst themselves, limit, control or attempt to control the production, distribution, sale or price of, or, trade in goods or provision of services. Three points of construction matter for exams and practice alike.

First, the definition is inclusive ("includes"), so it is not exhaustive — informal and tacit arrangements can qualify. Second, the gist is an agreement within the wide meaning of Section 2(b), which includes any arrangement, understanding or action in concert whether or not formal, in writing or intended to be enforceable; a cartel need not be a contract. Third, even an attempt to control is caught — the offence does not require that prices were actually fixed, only that the participants agreed to limit, control or attempt to control the relevant parameters.

Because a cartel is a species of horizontal agreement falling under Section 3(3), it attracts the rebuttable presumption of AAEC. Section 3(3) lists the classic cartel conduct — price-fixing under 3(3)(a), output or supply limitation under 3(3)(b), market or source sharing under 3(3)(c), and bid-rigging or collusive bidding under 3(3)(d). The presumption shifts the evidential burden to the accused enterprises once the agreement is established, a feature explored in the chapter on horizontal agreements.

Bid-rigging, the AAEC presumption and burden of proof

Bid-rigging — defined in the Explanation to Section 3(3)(d) as any agreement between enterprises engaged in identical or similar production or trading of goods or provision of services which has the effect of eliminating or reducing competition for bids or adversely affecting or manipulating the process for bidding — is the cartel form most often litigated, because public tenders generate documentary trails. Once such an agreement is shown, Section 3(3) presumes an appreciable adverse effect on competition; the enterprises must then displace the presumption with credible evidence.

In Excel Crop Care Ltd. v. Competition Commission of India, (2017) 8 SCC 47 (AIR 2017 SC 2734, decided 8 May 2017), three manufacturers of aluminium phosphide tablets quoted identical prices in tenders floated by the Food Corporation of India. The Supreme Court upheld the cartel finding but made the landmark ruling that the penalty under Section 27(b) must be calculated on "relevant turnover" — the turnover of the infringing product — rather than the enterprise's total turnover, importing proportionality into the penalty regime. This relevant-turnover principle has since governed cartel penalties, though it has been the subject of subsequent legislative attention.

Parallel conduct, oligopoly and the limits of inference

The hardest cartel question is when parallel behaviour — identical prices, simultaneous moves — proves an agreement and when it merely reflects rational adaptation in a concentrated market. The Act's inclusive cartel definition and the Section 3(3) presumption tempt the Commission to infer agreement from parallelism, but the courts have insisted on "plus factors".

In Rajasthan Cylinders and Containers Ltd. v. Union of India, (2020) 16 SCC 615 (2018 INSC 916, decided 1 October 2018), forty-odd manufacturers of LPG cylinders submitted near-identical bids in a tender by Indian Oil Corporation. The Commission and COMPAT found a cartel, but the Supreme Court set the finding aside. Drawing on its own reasoning in Excel Crop Care and on the market's oligopsonistic structure — a single dominant buyer (IOC) dictating quantities, price formulae and territories — the Court held that parallel pricing alone could not establish collusion and that the conditions of the market provided an innocent explanation for the parallelism. Suspicion was not enough; there was "not enough material" to conclude a cartel had in fact formed.

The lesson is that the inclusive cartel definition does not relieve the Commission of proof. Parallel conduct must be accompanied by plus factors — pre-bid meetings, common agents, information exchange, market-sharing — before an agreement is inferred, particularly in oligopolistic or oligopsonistic markets where price uniformity may be structurally inevitable.

The cement cartel: definitions in combination

The definitions interlock in the most consequential cartel decision to date. In Builders Association of India v. Cement Manufacturers' Association (CCI, Case No. 29/2010, order dated 20.06.2012), the Commission found eleven cement manufacturers and their trade association liable under Sections 3(3)(a) and 3(3)(b) read with Section 3(1) for coordinating prices and limiting production and supply, and imposed a penalty exceeding Rs 6,300 crore — at the time the largest in Indian antitrust history.

Every threshold definition was in play: the manufacturers and the Association were "enterprises" within Section 2(h) read with 2(l); the conduct fitted the inclusive "cartel" description in Section 2(c) because the producers, by agreement, attempted to control production, supply and price; and the analysis was anchored in a relevant market for grey cement. The Association functioned as the coordinating platform — illustrating why trade bodies are routinely roped in. The matter's long appellate history (including remand on natural-justice grounds and re-imposition by the Commission in 2016) shows that even a well-founded cartel finding must survive procedural and evidentiary scrutiny, but the substantive characterisation turned entirely on the Section 2 definitions.

Common confusions and exam traps

Cartel is not the only horizontal infringement. Section 3(3) covers cartels but also any agreement among competitors that fixes prices, limits output, shares markets or rigs bids; the cartel label in Section 2(c) is a subset, primarily relevant to enforcement vocabulary and the leniency regime. Treat "cartel" and "horizontal agreement under Section 3(3)" as overlapping, not identical.

Relevant market is not relevant turnover. Candidates routinely conflate the market-delineation exercise (Sections 2(r)–(t), 19(6)–(7)) with the penalty base. Excel Crop Care concerns relevant turnover under Section 27 — a penalty concept — not market definition under Section 2(r).

Sovereign-function exclusion is narrow. The Section 2(h) proviso shields only atomic energy, currency, defence and space and other inalienable sovereign functions; commercial activity by any government department or PSU is fully within the Act.

An "agreement" needs no writing. Section 2(b) covers arrangements, understandings and concerted action; the absence of a signed contract is no defence to a cartel allegation — though, as Rajasthan Cylinders shows, the Commission must still prove concert beyond mere parallelism.

Putting the definitions to work

A disciplined competition analysis proceeds definition by definition. Begin with Section 2(h): is each actor an enterprise engaged in economic activity, and does any sovereign-function carve-out apply? Then delineate the relevant market under Section 2(r), using Section 2(t) and the Section 19(7) factors for the product dimension and Section 2(s) with the Section 19(6) factors for the geographic dimension, informed by substitutability and, where useful, the SSNIP logic. Only then characterise the conduct: if competitors are coordinating, ask whether it answers the inclusive cartel description in Section 2(c) and engages the Section 3(3) presumption — and whether parallel behaviour is backed by plus factors per Rajasthan Cylinders.

The case law maps neatly onto this sequence: Coordination Committee of Artists on enterprise and market scope; MCX/NSE, Uber and Schott Glass on product and geographic delineation; Excel Crop Care, the cement cartel and Rajasthan Cylinders on the proof and consequences of cartelisation. Master the definitions and the rest of the Act — Sections 3, 4 and the penalty regime — falls into place. For the foundational framework, return to the Competition Act notes hub, and for the substantive prohibitions see anti-competitive agreements and abuse of dominant position.

Frequently asked questions

Is the definition of "cartel" in Section 2(c) exhaustive?

No. Section 2(c) says a cartel "includes" an association of producers, sellers, distributors, traders or service providers who by agreement limit, control or attempt to control production, distribution, sale, price or trade. The word "includes" makes it inclusive and non-exhaustive, so tacit and informal arrangements can qualify; the underlying "agreement" is read with the wide definition in Section 2(b), which needs no writing or enforceability.

Can a Government department or PSU be an "enterprise" under Section 2(h)?

Yes, to the extent it carries on economic activity. The test is functional. The proviso to Section 2(h) excludes only activities relatable to sovereign functions — expressly atomic energy, currency, defence and space — and that carve-out is read narrowly to cover only primary, inalienable and non-delegable functions. Commercial and welfare activities of departments and public-sector undertakings are fully within the Act, consistent with the competitive-neutrality principle.

How does the CCI determine the relevant product market?

Under Section 2(t), the relevant product market comprises products or services regarded as interchangeable or substitutable by the consumer by reason of their characteristics, prices and intended use. Section 19(7) lists the factors — physical characteristics or end-use, price, consumer preferences, exclusion of in-house production, existence of specialised producers and classification of industrial products. In CCI v. Schott Glass India (2025 INSC 668) the Supreme Court accepted two separate markets for NGC and NGA glass tubing because they were not substitutable.

What is the difference between the relevant product market and the relevant geographic market?

The relevant product market (Section 2(t), factors in Section 19(7)) is about which products are substitutable for each other. The relevant geographic market (Section 2(s), factors in Section 19(6)) is about the area where competitive conditions are distinctly homogenous and distinguishable from neighbouring areas, judged by factors such as regulatory barriers, transport costs, language and consumer preferences. Section 2(r) lets the Commission define the relevant market by reference to product, geography or both.

Does identical or parallel pricing automatically prove a cartel?

No. In Rajasthan Cylinders and Containers Ltd. v. Union of India, (2020) 16 SCC 615 (decided 1 October 2018), the Supreme Court set aside a cartel finding against LPG cylinder manufacturers, holding that parallel pricing in an oligopolistic/oligopsonistic market dominated by a single buyer (IOC) could be explained by market structure. "Plus factors" — such as pre-bid meetings, common agents or market-sharing — are needed before parallelism is treated as evidence of an agreement.

On what turnover is a cartel penalty calculated?

Following Excel Crop Care Ltd. v. Competition Commission of India, (2017) 8 SCC 47 (AIR 2017 SC 2734), the penalty under Section 27(b) is computed on the "relevant turnover" — the turnover of the infringing product or service — rather than the enterprise's total turnover, to keep the penalty proportionate to the harm. This relevant-turnover doctrine has governed cartel penalties since, subject to later legislative refinement of the penalty provisions.