The Competition Act 2002 is the statute that decides what Indian businesses may agree with each other, what a powerful business may do with its power, and which mergers and acquisitions need permission before they close. It is cited in the Gazette as Act 12 of 2003, because although Parliament passed it under the 2002 title, the President assented to it on 13 January 2003 and it was published the following day. That small mismatch between the name and the number is the first of many ways in which this Act refuses to sit still under a casual reading. The second is that almost everything commonly said about it in summary form is wrong in one specific direction: summaries describe it as a law against big business, against cartels and against mergers, when its text is in fact a law against defined conduct. It prohibits particular kinds of agreement, particular listed abuses of a dominant position, and combinations above defined thresholds that are not notified in time. Size, success and market power, without more, contravene nothing in it.

A reader approaching the Act for the first time should begin where its own logic begins, with five definitions that the operative sections presuppose. An enterprise, section 2(h), is a person or a department of government engaged in any activity relating to the production, storage, supply, distribution, acquisition or control of goods, or the provision of services, of the kinds the clause enumerates, and it excludes the sovereign functions of the state. An agreement, section 2(b), is any arrangement, understanding or action in concert, formal or informal, written or unwritten. A cartel, section 2(c), is an association of producers, sellers, distributors, traders or service providers who, by agreement among themselves, limit, control or attempt to control production, distribution, sale, price or trade. Turnover, section 2(y), is the value of goods sold or services rendered. And the relevant market, section 2(r), is the market the Commission determines by reference to a relevant product market and a relevant geographic market. Between them, these five clauses fix who the Act watches, what it counts as cooperation between them, and the arena inside which their power is measured; the prohibitions state what follows.

This article is the statute pillar for the Act. It states the substantive law: the three prohibitions and regulatory schemes in sections 3, 4, 5 and 6, the analytical apparatus of section 19 that the Competition Commission of India must apply, the orders and penalties in sections 27 and 28 and their companions, the leniency provision in section 46, and the extraterritorial reach in section 32. Two sibling articles take the neighbouring ground. The 2023 amendment, which rewrote parts of the penalty base and merger control, is owned by the article on that amendment, and the Commission as an institution, its composition in working detail and its enforcement record, is owned by the article on the Commission. What belongs here is the law a business, a lawyer or an examinee actually has to apply: which agreements are presumed harmful, which are assessed on their effects, why dominance is lawful while its abuse is not, and when a transaction must wait for approval before it closes.

A statute that arrived in instalments

The Competition Act did not begin to operate when it was enacted, and its staggered arrival is part of its substance, because for six years India had a competition statute on the book that prohibited nothing.

The Act replaced a predecessor built on a different theory. The Monopolies and Restrictive Trade Practices Act, 1969, enacted on the recommendation of the Monopolies Inquiry Commission of 1965 and shaped by the Dutt Committee’s work on industrial licensing, treated concentration of economic power as itself suspect and required large undertakings to seek approval for expansion, new undertakings and mergers. The 1991 amendments deleted most of that size-based apparatus, leaving a statute concerned mainly with restrictive and unfair trade practices and with little economic content. The High Level Committee on Competition Policy and Law, chaired by S. V. S. Raghavan and reporting in 2000, recommended repeal and replacement: India, the Committee reasoned, needed a law that policed conduct in markets rather than the dimensions of the firms in them, enforced by a specialist expert body. The Competition Bill followed, and the Act received assent on 13 January 2003 as Act 12 of 2003.

The distance between the two statutes is worth measuring, because the 2002 Act is intelligible mainly as a rejection of its predecessor’s method. The 1969 Act had been built for a planned economy. Its operative anxiety was concentration: undertakings above prescribed asset values were classified as monopolistic or dominant undertakings, and the classification itself attracted prior approval requirements for substantial expansion, for establishing new undertakings, and for mergers, amalgamations and takeovers. Trade practices were separately sorted into monopolistic, restrictive and, after the 1984 amendment on the Sachar Committee’s recommendation, unfair trade practices, and the MRTP Commission inquired into them with largely recommendatory powers in respect of the monopolistic class. The system’s signature was that the state decided in advance which growth was permissible. Its defect, visible by the 1990s, was that it regulated the corporate map rather than competition: a firm could be entirely lawful under it while charging what it liked in a market it had been licensed to occupy alone, and a merger of two small firms could require the same approval as the creation of a giant.

The Raghavan Committee’s recommendations inverted the logic, and the enacted Act follows the inversion closely. The Committee proposed that the MRTP Act be repealed; that restrictive trade practices be folded into a modern competition law concerned with conduct and its effects; that the new authority be an expert, quasi-judicial body rather than a department of government; that merger regulation be prospective and threshold-based rather than discretionary and size-based; and that the statute concern itself with the promotion of competition as an organising value, including through an advocacy function by which the regulator would speak to government policy itself. Sections 3, 4 and 5 enact the conduct core of that design. Section 18 writes the promotion of competition into the Commission’s charter, and section 49 gives the advocacy limb its form: on a reference from the Central Government or a State Government, the Commission renders an opinion on the possible effect of a policy on competition, and the governments are to consider its views in framing policy, without being bound by them. The pillar articles of this series take the provisions in order; what matters at the outset is that the Act’s narrowness is inherited from a committee that had examined the alternative and rejected it.

Section 1(3) left commencement to notification, and the notifications came in pieces. A first group of provisions, including section 1, parts of the definitions in section 2 and the sections on appointment and terms of the Commission’s members, was brought into force on 31 March 2003 by notification S.O. 340(E). The majority of the remaining sections, including section 7 on establishment of the Commission, followed on 19 June 2003 by notification S.O. 715(E). The Competition Commission of India was established on 14 October 2003. Yet sections 3, 4, 5 and 6, the entire operative core, remained unnotified. A statute book entry existed; a working prohibition did not.

The delay was not only administrative. The Commission as first designed was to combine advisory, regulatory and adjudicatory functions in one body whose members need not be judges, and that design was challenged in Brahm Dutt v. Union of India. The Supreme Court recorded the Union’s position that the scheme would be reconsidered, and the outcome was the Competition (Amendment) Act, 2007, Act 39 of 2007, whose principal provisions took effect on 12 October 2007. The amendment split the design in two. The Commission became a seven-member expert body, a Chairperson and not fewer than two and not more than six other members under section 8, functioning as a market regulator. A new Competition Appellate Tribunal was created under section 53A to hear appeals. The amendment also converted merger notification from the voluntary scheme of the original text into a mandatory one with a standstill obligation, a change whose consequences run through the whole of the combinations chapter.

The intervening years left a peculiar institutional residue. Between establishment in October 2003 and the commencement of the substantive provisions in 2009, the Commission existed, recruited, published and advocated, but prohibited nothing, because the sections that prohibit had not been brought into force and, for much of the period, because the Commission’s own composition remained incomplete while the design litigation ran its course. The advocacy half of the Raghavan design was thus the only half in operation for the first six years of the institution’s life, a sequence that shaped the Commission’s self-understanding: it arrived at enforcement in 2009 as a body that had been explaining competition to the government for years and had to learn, in public and under appeal, how to prove it against unwilling parties.

With a full Commission appointed in 2009, the substantive law finally switched on. Sections 3 and 4 and the enforcement provisions around them came into force on 20 May 2009. On 1 September 2009, section 66 took effect and repealed the Monopolies and Restrictive Trade Practices Act, 1969, with a two-year transitional tail under which the old Commission finished pending cases before dissolution while pending investigations moved to the new Commission; the mechanics of that succession are the subject of the companion article on the repeal. The combination provisions, sections 5 and 6 together with the inquiry and order provisions in sections 20, 29, 30 and 31, came last, appointed for 1 June 2011 by notifications issued in March of that year. An Act assented to in January 2003 therefore became fully operative only in June 2011, eight years and five months later. The appellate half of the 2007 design had a shorter life than the delay that produced it: the Finance Act, 2017 substituted the National Company Law Appellate Tribunal for the Competition Appellate Tribunal as the Appellate Tribunal under the Act with effect from 26 May 2017, and appeals under section 53B have since lain to that forum.

The One Test: three pillars, three different logics

Everything in the Act’s operative part answers to one of three provisions, and the discipline of the statute is that each has a different trigger, a different burden and a different consequence. Confusing them is the source of most errors made about this law.

Section 3 prohibits agreements that cause or are likely to cause an appreciable adverse effect on competition within India. Section 4 prohibits the abuse of a dominant position by an enterprise that holds one. Sections 5 and 6 regulate combinations, meaning acquisitions, mergers and amalgamations above prescribed financial thresholds, by requiring prior notice to the Commission and prohibiting the parties from closing until the Commission approves or the statutory waiting period expires. The first two provisions are prohibitions addressed to conduct. The third is a screening scheme addressed to transactions, most of which it clears.

The table below is the findable artifact of this article, and it states the distinctions on which everything else depends.

Category of conduct	Legal test	Who bears the burden	What evidence matters	Consequence	Defence available
Horizontal agreement between competitors (section 3(3))	Presumed to have an appreciable adverse effect on competition	The parties, to displace the presumption	The agreement itself and its terms; market context if the presumption is contested	Agreement void; cease and desist direction; penalty	Joint venture that increases efficiency; the section 19(3) factors in rebuttal
Vertical agreement along a supply chain (section 3(4))	Unlawful only if it causes or is likely to cause an appreciable adverse effect	The Commission and the informant, to establish the effect	Market shares, foreclosure, entry barriers, consumer benefit, the section 19(3) factors	Agreement void if the effect is shown; penalty	Showing of no appreciable adverse effect; intellectual property and export exceptions in section 3(5)
Unilateral conduct of a dominant enterprise (section 4)	Dominance must be proved, then conduct matched to the exhaustive list of abuses in section 4(2)	The Commission, on both dominance and abuse	Relevant market definition, the section 19(4) factors, the conduct and its terms	Cease and desist direction; penalty; possible recommendation for division under section 28	Conduct outside the listed abuses is not prohibited; dominance itself is no offence
Combination above the thresholds (sections 5 and 6)	Notification is mandatory; approval turns on whether the combination causes or is likely to cause an appreciable adverse effect	The parties must notify and wait; the Commission assesses the effect	The section 20(4) factors: entry barriers, countervailing power, substitutes, innovation, market shares	Clearance, clearance with modification, or a direction that the combination shall not take effect; penalty for closing early under section 43A	Thresholds not crossed; notified exemptions, including the small target exemption; approval after review

The allocation of burdens across the three pillars is itself a policy statement, and it is the deepest structural fact in the Act. Under section 3(3), Parliament found the harm itself, legislatively, and left only rebuttal to the parties; under section 3(4) and section 4, the legislature declined to find harm in advance and set the Commission to prove it, in the one case as effect and in the other as dominance plus listed conduct; under sections 5 and 6, where the transaction has not happened and there is no harm to prove, the burden is reversed in form rather than in substance, the parties bearing the duty to come forward, disclose and wait, while the Commission retains the burden of the adverse finding that alone can stop or condition the deal. The gradient is rational. The closer the conduct stands to the hard core of collusion, the less the state must prove; the closer it stands to ordinary commerce, the more it must. Most errors in argument under this Act are errors about where on that gradient a given case sits.

Two features of this scheme deserve emphasis before the detail begins. The first is that the Act regulates agreements and transactions prospectively and conduct retrospectively, but in no case does it penalise a state of affairs. A market with one supplier is a fact the Act can investigate; it is not, without an agreement, an abuse or an unnotified combination, a contravention. The second is that the phrase that does the real work, appreciable adverse effect on competition, is nowhere reduced to a formula. Section 19 instead gives the Commission lists of factors it must weigh, and the case law and orders under the Act are largely a record of that weighing. The statute is therefore best understood as a set of instructions about where the burden of persuasion sits, followed by a set of factors by which persuasion is attempted.

Section 3: agreements, and the line between presumed and assessed

Section 3(1) states the general prohibition. No enterprise, association of enterprises, person or association of persons shall enter into an agreement in respect of production, supply, distribution, storage, acquisition or control of goods, or provision of services, which causes or is likely to cause an appreciable adverse effect on competition within India. Section 3(2) provides that an agreement entered into in contravention is void. The definition of agreement in section 2(b) is deliberately wide: it includes any arrangement, understanding or action in concert, whether or not formal, in writing, or intended to be enforceable by legal proceedings. A nod between competitors at a trade association meeting is capable of being an agreement under this Act, and the Commission’s cartel orders have repeatedly found agreements in patterns of parallel conduct held together by evidence of communication rather than in signed documents.

Agreements within a group

One boundary of the agreements pillar is fixed by the definitions rather than by section 3 itself. An agreement requires at least the meeting of conduct between enterprises or persons, and the Act’s definition of group, built on voting rights, board appointment and control of management or affairs, identifies the enterprises that function as one economic unit. Arrangements between a parent and the subsidiary it wholly controls, allocating functions, territories or production between them, present the question whether there are two actors capable of agreeing or one actor organising itself. The Commission’s approach in its orders has been functional rather than formal: where the subsidiary enjoys no genuine autonomy of conduct in the market, the arrangement is treated as internal organisation, while arrangements between group entities that do face each other, or face third parties, as independent market actors are examined as agreements in the ordinary way. Parties should not assume the label settles the question; the same control analysis that aggregates a group’s figures for the section 5 thresholds can aggregate its members’ conduct for section 3, and can equally dissolve an apparent agreement into a single enterprise’s internal decision. The safe statement of the law is that group membership is evidence about economic unity, not an exemption from the pillar, and that genuine autonomy, documented in how decisions are actually taken, is what keeps an intra-group arrangement outside section 3.

Proving an agreement that nobody signed

The width of section 2(b) creates the central evidentiary question of the agreements pillar: how is an unwritten understanding proved, and how is it separated from the innocent parallelism of firms that watch each other in a transparent market? Identical prices, standing alone, prove very little, because competitors in a market for a homogeneous product will often arrive at the same price independently, each responding to the same costs and the same published prices of the others. The Commission’s practice, sustained through the appellate process in its broad direction, is to look for the understanding behind the parallelism: meetings, telephone records, messages, association circulars, exchanges of price intentions, identical and otherwise inexplicable timings, departures from each firm’s previous behaviour that make sense only as coordination, and the testimony of a participant obtained, frequently, through the leniency route in section 46. Circumstantial evidence of that kind is not a dilution of the prohibition; it is the only form in which the prohibited thing ordinarily exists. A cartel that generated documents describing itself as one would have failed at its own purpose.

Two consequences follow for readers of the orders. The first is that the reasoning in a section 3(3) case is usually a chain: a finding of concert, drawn from communications and conduct together; the placement of the concert within one of the four listed categories; and then, the presumption doing its work, the contest shifts to rebuttal on the section 19(3) factors. The second is that information exchange sits uneasily at the edge of the pillar. An exchange of current prices or future price intentions between competitors, through an association or directly, can itself be the mechanism by which a price is determined indirectly, and the Commission has treated such exchanges as evidence of, and sometimes as the substance of, the prohibited agreement. Aggregated, historical, genuinely anonymised market statistics collected for a trade body’s legitimate purposes stand at a different point on the line, and the Act’s text draws no bright boundary between the two: the inquiry is whether the exchange, in its market, causes or is likely to cause the prohibited effect, or forms part of an arrangement that determines prices. Businesses that share information with competitors bear the burden of that ambiguity, and the prudent course the orders teach is to keep exchanges old, aggregated and one-directional toward the public.

Within the general prohibition, section 3 draws a line that organises the whole law of agreements in India. On one side stand the four categories of horizontal agreement listed in section 3(3), which Parliament has presumed to have an appreciable adverse effect. On the other side stand the five categories of vertical agreement listed in section 3(4), which are unlawful only when the effect is proved. The placement of an agreement on one side or the other of that line decides who must prove what, and it is the single most useful thing a reader can carry away from this article.

The section 3(3) categories apply to agreements between enterprises engaged in identical or similar trade of goods or provision of services, including cartels, which determine prices, limit production or markets, share markets or sources, or rig bids. The section 3(4) categories apply to agreements between enterprises at different stages or levels of the production chain in different markets: tie-in arrangements, exclusive supply agreements, exclusive distribution agreements, refusal to deal, and resale price maintenance. Each is examined below.

Are all agreements between competitors unlawful?

No. Only the four listed horizontal categories carry the presumption: price determination, limitation of production, supply, markets, technical development or investment, market sharing, and bid rigging. Every other agreement between competitors is tested on its effects against the factors in section 19(3), and one outside the list producing no appreciable adverse effect contravenes nothing.

The presumed category: section 3(3)

Section 3(3) applies where enterprises or their associations, including cartels as defined in section 2(c), engaged in identical or similar trade, agree to do any of four things. The first is to directly or indirectly determine purchase or sale prices. The second is to limit or control production, supply, markets, technical development, investment or provision of services. The third is to share the market or source of production or provision of services by allocation of geographical area, type of goods or services, or number of customers. The fourth is to directly or indirectly engage in bid rigging or collusive bidding, a term section 3(3) itself explains as an agreement that has the effect of eliminating or reducing competition for bids or adversely affecting or manipulating the process for bidding. An explanation to the sub-section adds that a practice carried on, or decision taken by, an association of enterprises or persons is included, which is how trade associations come to be respondents in cartel cases when their platforms are used to align the conduct of members.

The legal consequence of falling within the list is stated in a single clause: such agreements shall be presumed to have an appreciable adverse effect on competition. The Commission need not prove the effect. The parties may attempt to displace the presumption by reference to the factors in section 19(3), and the Commission’s orders record such attempts, but the starting point is against them and the practical burden of the case has shifted before evidence is called.

The presumption has a proviso. It does not apply to an agreement entered into by way of a joint venture if the joint venture increases efficiency in production, supply, distribution, storage, acquisition or control of goods or provision of services. Two limits should be noted. The proviso is attached to the presumption, so it answers the section 3(3) case; a joint venture agreement alleged under the general prohibition would in any event be assessed on effects. And the proviso is narrow on its face: it speaks of joint ventures and of increased efficiency, and parties invoking it must be ready to show both. Genuine joint ventures between competitors, for research, for shared infrastructure, for entry into a market neither could serve alone, are the commercial centre of the proviso, and the Commission’s task in such cases is to separate the efficiency the venture creates from any price or market alignment that travels with it and serves no efficiency purpose.

The case for the presumption, stated at its strongest, is the case made by enforcement practice in most jurisdictions that treat hard core cartels as unlawful by object. Cartels are covert. Their members keep no minutes marked as such, communicate through intermediaries and associations, and design their arrangements to resemble independent behaviour. Direct evidence of effect is exactly what a successful cartel suppresses, because its profit comes from the gap between the price it charges and the price competition would have set, and that counterfactual is expensive to reconstruct years later. If the authority had to prove appreciable adverse effect to an adjudicative standard in every case, most cartels would be unenforceable against, and the prohibition would exist mainly as literature. A presumption against the four classes of conduct that experience shows to be harmful in the overwhelming run of cases is, on this view, an evidentiary rule fitted to the offence.

The case against deserves equal statement, and it has been pressed by counsel in the Commission’s proceedings and by commentators on the Act. The Indian provision does not declare the listed conduct unlawful in itself, as a per se rule would. It presumes an effect, and a presumption in law is rebuttable, yet the Act nowhere specifies by what evidence or to what weight the presumption may be rebutted, so parties face a burden whose dimensions they learn only from the Commission’s orders, case by case. The categories are also drawn by conduct description rather than by market setting: an agreement on prices between two firms with a trivial combined share of a fragmented market is, textually, within section 3(3)(a) in the same way as the same agreement between the only two suppliers in the country, and the work of distinguishing them is left to the rebuttal stage rather than done at the threshold. Defenders of the design answer that the section 19(3) factors give the Commission and the parties a common vocabulary for the rebuttal, and that in practice the Commission does examine market context even in section 3(3) cases. Both accounts are accurate about different things: the first about the text, the second about the practice. The verdict this article records is that the presumption is defensible as an enforcement instrument and should be read, as the better orders read it, as shifting the burden of production on effects while leaving the Commission obliged to engage with the market evidence actually placed before it.

The four categories in practice

Each of the four categories has a characteristic shape in the decided cases, and the shapes are worth separating because the evidence differs between them.

Price determination, section 3(3)(a), is the widest of the four. It reaches direct fixing of the price charged, and it reaches the indirect apparatus by which prices are aligned without a stated figure: common discount structures, agreed margins, uniform delivered-pricing formulas, undertakings communicated through an association to abide by a price circular, and agreements on the components of price, such as transport charges, credit periods or taxes passed through. A recommendation by a trade association that its members adopt a price is capable of falling within the category where the recommendation is in substance a decision of an association of enterprises, a form the Explanation to section 3(3) expressly contemplates. The prosecution’s characteristic exhibits are the circular, the meeting and the sudden uniformity.

Limitation, section 3(3)(b), is the quota agreement: an understanding that production, supply or investment will be held below what each firm’s separate interest would produce, so that scarcity maintains price. Its evidence is arithmetic, capacity records and output data set against demand, together with the communications that organised the restraint. The category also catches agreements limiting technical development, the joint decision not to bring a product or a process to market, a form less visible than a price agreement and correspondingly rarer in the orders.

Market sharing, section 3(3)(c), divides the field instead of the price: territories allotted so that each party sells undisturbed in its own, products allocated by type, customers assigned by name or by class. Sharing by number of customers extends to arrangements under which parties take turns, or abstain in rotation, so that each receives an agreed proportion of the available business. The category’s evidence is the pattern of abstention itself combined with the record of the arrangement that produced it.

Bid rigging, section 3(3)(d), is market sharing applied to tenders: agreements on who will bid, at what price, and who will stand down, cover bidding in which the designated loser submits an inflated or incomplete bid to create the appearance of competition, and rotation schemes across a series of tenders. Public procurement is its natural habitat, which is why the category carries a particular enforcement weight: the victim is the public purchaser, the loss is measurable against the tender record, and detecting authorities increasingly find the pattern in bid data before they find a witness. The definition the sub-section supplies, an agreement that eliminates or reduces competition for bids or manipulates the bidding process, is deliberately functional, catching the arrangement whatever its paperwork.

Horizontal agreements outside the list

The presumption list does not exhaust horizontal cooperation, and the agreements it leaves out are examined under the general prohibition in section 3(1) on the section 19(3) factors, with the burden on the Commission. Three recurring kinds show the range. Joint purchasing, by which smaller buyers combine their orders to win terms none could win alone, restricts each participant’s freedom to buy separately, yet its characteristic effect is countervailing power rather than foreclosure, and it fails the assessment only where the buying group is large enough to depress prices below competitive levels or to close the market to non-members. Research and development cooperation between competitors pools a cost and a risk that the market under-supplies when each firm fears its rivals will copy the result; it is assessed on whether the cooperation is confined to the research or extends to the marketing of its fruits, which is where the restraint stops serving the efficiency that justified it. Standard-setting, the agreement of an industry on common specifications, interfaces or testing methods, is indispensable to interoperable markets and becomes a contravention where the standard is written to exclude a rival’s technology, where the process is captured by interested parties, or where the standard is used as the venue for aligning prices under a technical flag. None of these is presumed harmful; none is immune; and the discipline of the effects test, placing benefit and foreclosure on the two sides of the section 19(3) sheet, is what separates the pillar’s treatment of cooperation from a regime that would license it in advance or forbid it in gross.

The assessed category: section 3(4)

Section 3(4) lists five vertical arrangements: a tie-in arrangement, by which a buyer must take a second product or service as a condition of taking the first; an exclusive supply agreement, restricting the purchaser in the course of trade from acquiring goods other than those of the seller; an exclusive distribution agreement, limiting, restricting or withholding the output of goods or the supply of goods or services to defined markets or customers; a refusal to deal, restricting the persons or classes of persons to whom goods are sold or from whom they are bought; and resale price maintenance, a condition in a contract that the price to be charged on resale shall be a price stipulated by the seller, unless it is clearly stated that a lower price may be charged. Each is prohibited only to the extent that it causes or is likely to cause an appreciable adverse effect on competition in India.

The consequence is that exclusive distribution, selective dealing and stipulated resale prices are not unlawful in themselves in India. They are common instruments of distribution economics: a manufacturer entering a State may rationally appoint one distributor and support that distributor’s investment in service networks with territorial protection; a premium brand may rationally police the conditions of resale to protect the quality of the retail experience on which its price rests. Indian law asks what such arrangements do in their market. Where the supplier has a small share and rival brands compete at every level, the answer will ordinarily be that inter-brand competition disciplines the restraint and no appreciable adverse effect follows. Where the supplier is large, the restraint covers most of the market, and its effect is to foreclose rival suppliers from distribution or to eliminate price competition among retailers of the same brand, the assessment can go the other way. This effects-based treatment of resale price maintenance is more permissive than the treatment the practice receives in some jurisdictions, where stipulated resale prices are condemned with little inquiry, and it is frequently misunderstood by businesses that assume a rule from another legal system applies in India. It does not. The question in India is effect, and the party alleging the contravention must prove it.

The burden under section 3(4) shapes the litigation in a way worth stating. An informant cannot succeed by producing the vertical agreement alone, however restrictive its terms appear. The informant must also construct the market: the supplier’s share, the coverage of the restraint across the market, the position of rival brands, the ease of entry at each level, and the reasons the restraint exists, because the section 19(3) sheet must be filled in on both sides before the Commission can weigh it. Respondents, correspondingly, defend vertical cases with the distribution economics of their trade, the service investments the restraint protects, the free-riding it prevents and the inter-brand rivalry that disciplines the whole arrangement. Many section 3(4) informations close at the prima facie stage precisely because the market construction is absent, and the pillar’s lesson for businesses and complainants alike is that in this half of section 3, the economics is not the decoration of the legal case. It is the legal case.

The five vertical arrangements in practice

The section 3(4) categories likewise reward separate treatment, because the route by which each can harm competition is different, and the assessment follows the route.

A tie-in arrangement compels the buyer of one product to take another. The competitive concern is leverage through contract rather than through dominance: a supplier strong in the first product forecloses the market for the second by attaching it, and rival makers of the second product lose access to buyers who would otherwise have chosen freely. The assessment asks how strong the tying product is, what proportion of the tied market the arrangement covers, and whether buyers can realistically obtain the first product unbundled, from this supplier or another.

An exclusive supply agreement closes the buyer’s side of the market: the purchaser agrees to source a product only from the one supplier. Short exclusivity between small parties is a routine instrument of supply security. The concern grows with duration, coverage and the supplier’s share, because a market in which most substantial buyers are tied for years at a time is a market a new supplier cannot enter at viable scale, whatever the quality of its offer.

An exclusive distribution agreement closes the selling side symmetrically, confining a distributor to a territory or a class of customers, or withholding supply from others. Its effects run in two directions at once: intra-brand competition between distributors of the same product is suppressed by design, while inter-brand competition may be sharpened if the protection induces the distributor to invest in promotion and service. The assessment weighs which effect dominates on the facts, and the supplier’s market share is usually the pivot, since territorial protection granted by a small brand constrains a small part of the market.

A refusal to deal, as a listed agreement, concerns restrictions on the persons or classes of persons to whom goods are sold or from whom they are bought: the supplier that will sell only to distributors who boycott a rival, or the buyers’ arrangement that refuses supply from a discounter. Refusal by one firm acting alone, without agreement, is not within section 3 at all, and reaches the Act, if at all, only as an abuse of dominance under section 4(2)(c); the section 3(4) entry requires the restriction to be the content of an agreement.

Resale price maintenance closes the set. The stipulation may fix the resale price exactly, fix a minimum, or operate through margins and discounts; the Act’s definition is a condition that the price to be charged on resale shall be the price stipulated by the seller, unless it is clearly stated that a lower price may be charged. A recommended price with a genuine liberty to charge less is outside the definition. Where the stipulation binds, the assessment examines why the supplier imposes it, whether it suppresses discounting that consumers would otherwise enjoy, whether dealers themselves procured it to end price competition among them, in which case the arrangement shades toward the horizontal, and what share of the market the supplier’s network covers. The Indian treatment, assessment rather than condemnation, places the analytical weight where it belongs: a resale price is a fact about a distribution system, and it is the system’s market power, not the stipulation’s existence, that decides the case.

The exceptions: section 3(5)

Section 3(5) provides that nothing in section 3 restricts two classes of right. The first is the right of a person to restrain an infringement of, or to impose reasonable conditions necessary for protecting, rights conferred under the Copyright Act, 1957, the Patents Act, 1970, the Trade Marks Act, 1999, the Geographical Indications of Goods (Registration and Protection) Act, 1999, the Designs Act, 2000 and the Semi-conductor Integrated Circuits Layout-Design Act, 2000. A patent licence that confines the licensee to the field the patent covers, or a franchise that protects the licensed marks by quality conditions, is therefore not to be condemned under section 3 for doing what the intellectual property statute entitles the holder to do; the exception is the point at which this Act meets the Patents Act 1970 explained and its siblings, and its limits are set by the words reasonable and necessary, which keep conditions going beyond the protection of the right within the reach of section 3. The second protected class is the right to export goods from India, where the agreement relates exclusively to production, supply, distribution or control of goods, or provision of services, for such export. An export cartel, an arrangement among Indian producers that allocates foreign markets among them, is outside section 3 so long as the arrangement touches only the export trade; the moment its terms govern sales within India, the protection ends.

The intellectual property limb repays a closer reading, because its limits are litigated more often than its existence is noticed. The exception protects the right itself: restraining an infringement, and imposing conditions that are reasonable and, the harder word, necessary for protecting the rights the listed statutes confer. A condition that travels beyond the statutory right is not within the shelter merely because it appears in a licence. A patentee’s licence that fixes the price at which the licensee must resell unpatented goods, or that ties the licence to purchases of products the patent does not cover, protects the patentee’s commercial position rather than the patent, and such conditions return to section 3 to be judged like any other agreement, under section 3(4) if vertical and under the general prohibition otherwise. The dividing question is always the same: is the condition explained by the existence and scope of the intellectual property right, or is the right being used as the occasion for a restraint the market would otherwise assess on its merits? The exception, so read, is not an antitrust immunity for intellectual property. It is a coordination clause between two statutes, telling the reader that conduct one statute affirmatively entitles shall not be condemned by the other, and leaving everything outside the entitlement to the ordinary analysis.

The apparatus of analysis: effect, market, factors

Sections 3 and 4 both turn on judgments the Act refuses to leave unstructured. Section 19 supplies the structure in three places: the factors for appreciable adverse effect, the factors for dominance, and the method for fixing the relevant market inside which both inquiries take place.

Section 19(3) lists the factors the Commission shall have due regard to in determining whether an agreement has an appreciable adverse effect on competition: the creation of barriers to new entrants in the market; the driving of existing competitors out of the market; the foreclosure of competition by hindering entry into the market; the accrual of benefits to consumers; improvements in production or distribution of goods or provision of services; and the promotion of technical, scientific and economic development by means of production or distribution of goods or provision of services. The list is a balance sheet. Its first three entries look for harm, its last three for benefit, and an assessment under section 3(4), or a rebuttal under section 3(3), consists of placing the agreement’s features on the two sides of that sheet. The factors are mandatory considerations rather than a checklist that decides by counting: an agreement that raises a barrier and also delivers a demonstrable improvement in distribution presents a weighing question, not an arithmetical one.

The relevant market is defined in section 2(r) as the market determined by the Commission with reference to the relevant product market, the relevant geographic market, or both. Section 19(5) directs the Commission to have due regard to both components. For the relevant geographic market, section 19(6) lists factors including regulatory trade barriers, local specification requirements, national procurement policies, adequate distribution facilities, transport costs, language, consumer preferences and the need for secure, regular supplies or rapid after-sales services. For the relevant product market, section 19(7) lists factors including the physical characteristics or end-use of goods, their price, consumer preferences, the exclusion of in-house production, the existence of specialised producers and the classification of industrial products. The practical consequence is that market definition is an evidence question answered afresh in each case. A product market drawn too widely will make dominance impossible to find and vertical restraints harmless by construction; drawn too narrowly it will manufacture dominance out of a brand. Much of the contested ground in the Commission’s major orders, and in the appeals from them, is this preliminary-looking question, because everything downstream depends on it.

The working method by which the product market is fixed deserves statement in the pillar, because it is the point at which law and economics meet in this Act. The question is substitutability, asked from the buyer’s side: if the price of the candidate product rose by a small but significant amount, and stayed there, would enough buyers switch to another product to make the rise unprofitable? The products to which they would switch belong in the market; the products they would refuse at any realistic price do not. Supply-side substitution, the ability of producers of neighbouring products to redirect capacity, enters through the same factors. The geographic inquiry is its mirror image, asking how far buyers will travel, ship or connect before the constraint of distance, regulation or preference releases the seller from rivalry. None of this machinery is spelled out in section 19, whose factor lists are deliberately open-textured, and the Commission has built the method in its orders case by case, with the appellate tribunals reviewing its application. The discipline the factors impose is real all the same: a market definition that cannot be explained through the listed factors, or that changes between two stages of the same case to fit the conclusion sought, will not survive appeal, and several of the Act’s best-known reversals have turned on exactly that failure.

Indian conditions give the geographic limb a character of its own. For heavy goods of low value relative to their weight, cement, steel products, aggregates and the like, transport costs confine rivalry within a radius, and the relevant geographic market is frequently a State or a cluster of districts rather than the country; a producer dominant in one State’s market may be a minor supplier nationally, and the section 19(4) inquiry must be run inside the boundary where its buyers actually have alternatives. For services delivered through networks, banking, telecommunications and digital platforms among them, the market may be national in form while particular factors, language, local specification, State-level licensing, the density of physical service networks, pull the effective boundary inward. The section 19(6) factors read like a description of the Indian economy’s frictions: regulatory trade barriers between States, procurement policies that favour local supply, distribution facilities that thin out with distance, and consumer preferences formed by language and habit. A market definition that ignores them, drawing the geographic market at the national border by default, will overstate the competitive constraint on a regional monopolist and understate it on a national platform, erring in both directions in the same case load. The Commission’s better orders work the factors State by State and product by product, and the discipline of doing so is a large part of what distinguishes a finding of dominance that survives appeal from one that does not.

Section 4: dominance is lawful, abuse is not

Section 4 is the shortest of the three pillars and the most misstated. Its opening words are that no enterprise or group shall abuse its dominant position. The prohibition attaches to abuse. Dominance itself, however complete, is nowhere prohibited, and no provision of the Act empowers the Commission to act against an enterprise merely for being large, merely for holding the greater part of a market, or merely for having displaced rivals by better products and lower prices. The namable claim of this article, the sentence around which its reading of the Act is built, is that dominance is lawful and abuse is not; the whole of section 4 is the working out of that distinction.

Is being dominant an offence in India?

No. The Act contains no offence of being dominant, holding a large market share, or driving rivals out of business by lawful competition. Section 4 prohibits only the abuse of a dominant position, and abuse is confined to the five kinds of conduct listed in section 4(2). A dominant enterprise that does none of the five contravenes nothing.

The distinction is the foundation of the pillar, and it has a practical corollary for how cases are pleaded and defended. An informant cannot build a section 4 case out of size plus grievance. The informant must first construct the relevant market, then establish dominance within it on the section 19(4) factors, then match identified conduct to an entry in the section 4(2) list. Failure at any stage ends the case, however unattractive the respondent’s commercial behaviour may appear in the round.

Dominance: the section 19(4) inquiry

Dominance is determined inside a defined relevant market, by the factors in section 19(4), and not by market share alone. The Commission must have due regard to the enterprise’s market share; its size and resources; the size and importance of its competitors; its economic power, including commercial advantages over competitors; its vertical integration and sales or service network; the dependence of consumers on it; whether its position results from statute or from being a government company or public sector undertaking; entry barriers of every kind, regulatory, financial, technical and those arising from economies of scale, scope and the density of the market; countervailing buying power; the structure and size of the market; social obligations and social costs; the relative advantage the enterprise has, by way of contribution to economic development, that has caused or is likely to cause its position; and any other factor the Commission considers relevant. Share is the starting point of the inquiry and the entry in the list that is placed first, but the text makes it one factor among thirteen, and an enterprise with a large share in a market with low entry barriers and powerful buyers may escape a finding of dominance, while one with a smaller share, controlling an indispensable facility in a market ringed by regulatory barriers, may not.

The Act defines the term itself. The Explanation to section 4 states that dominant position means a position of strength, enjoyed by an enterprise, in the relevant market in India, which enables it to operate independently of the competitive forces prevailing in that market, or to affect its competitors or consumers or the relevant market in its favour. The two limbs matter. Independence from competitive forces describes the firm that can raise price, degrade quality or slow innovation without losing enough custom to make the strategy unprofitable. The ability to affect competitors, consumers or the market in its favour describes the firm whose conduct others must accommodate: the platform whose terms suppliers must accept, the producer whose prices rivals treat as the weather. A finding of dominance should identify which limb is satisfied and by what evidence, and the Commission’s orders are at their strongest when they do exactly that, working through the section 19(4) factors against the record rather than inferring strength from a share figure alone.

The five abuses: section 4(2)

Once dominance is established, section 4(2) provides that there shall be an abuse of a dominant position if an enterprise or group directly or indirectly does any of five things. The list is exhaustive in character: conduct of a dominant firm that does not fall within one of the five entries may be sharp practice, may offend another statute, but it is not an abuse under this Act.

The first entry is the imposition, in a contract or otherwise, of an unfair or discriminatory condition in the purchase or sale of goods or services, or an unfair or discriminatory price in the purchase or sale of goods or services, including a predatory price. The second is limiting or restricting production of goods or provision of services or the market for them, or limiting technical or scientific development relating to goods or services to the prejudice of consumers. The third is indulging in a practice or practices resulting in the denial of market access in any manner. The fourth is making the conclusion of contracts subject to acceptance by other parties of supplementary obligations which by their nature, or according to commercial usage, have no connection with the subject of such contracts, the statute’s description of tying imposed by a dominant firm. The fifth is using a dominant position in one relevant market to enter into or protect another relevant market, the leveraging clause.

Predatory pricing, named inside the first entry, is defined by the second Explanation to section 4: a predatory price is the sale of goods or provision of services at a price below the cost, as may be determined by regulations, of production of the goods or provision of the services, with a view to reducing competition or eliminating competitors. Three elements must coincide: a price below cost, cost being measured by the standards the Commission’s cost regulations supply; dominance in the relevant market; and the purpose of reducing competition or eliminating a competitor. Deep discounting by a new entrant, or by any firm that is not dominant, is not predatory pricing under this Act whatever its effect on rivals, because section 4 does not reach the non-dominant at all. Even for a dominant firm, low prices are the thing competition law exists to produce, and the requirement of a view to eliminate competition is the textual barrier that keeps ordinary price competition outside the abuse.

The dominance the Act does not recognise

Section 4’s predicate is a single enterprise, or a group bound by control, holding a position of strength. The Act contains no doctrine of collective dominance, under which two or three firms that are not connected and have not agreed, but whose parallel conduct produces the effects of monopoly, could be treated as jointly dominant and restrained. The omission is significant and, on the text, deliberate. An oligopoly that coordinates through an agreement is caught by section 3; an oligopoly that merely watches itself into parallel prices, each firm independently calculating that a price cut will be matched, presents the state with high prices and no defendant whose conduct the Act condemns, because section 4 demands dominance and section 3 demands agreement, and tacit coordination is neither. Merger control is the Act’s answer at the structural level, since it stops such markets from forming by concentration; after formation, the statute largely leaves the outcome to entry, to countervailing buyers and to time. Commentators have criticised the gap, and legislatures elsewhere have debated market-investigation instruments to fill it; the Indian text, as it stands, contains none, and the pillar states the position because a reader who assumes the Commission can proceed against a comfortable oligopoly as such will search the Act in vain for the power.

Section 4 in application

The abuse entries are abstract nouns until they are set against the situations that recur in the case law, and three situations account for most of the pillar’s litigation.

The first is the refusal, by a dominant firm that controls an input, a facility or a platform, to deal with a rival or a customer on terms the firm extends to others, or at all. Section 4(2)(c), denial of market access in any manner, is the entry under which such conduct is usually examined, with section 4(2)(a) capturing the variant in which access is granted but on unfair or discriminatory conditions. The inquiry concentrates on indispensability and effect: whether the input is genuinely necessary to compete, whether the dominant firm supplies it to its own downstream arm, and whether the refusal eliminates competition in the dependent market rather than merely disadvantaging one competitor. A firm is not the insurer of its rivals’ businesses, and the entry does not oblige a dominant firm to subsidise competition; it obliges the firm not to use control of the gateway as the instrument by which competition beyond the gateway is ended.

The second is leveraging, section 4(2)(e): the use of dominance in a market the firm lawfully dominates to enter or protect a second market in which it does not. The characteristic modern form is the platform or integrated supplier that extends its core position sideways, by tying, by self-preferencing its own adjacent services, or by conditions in the first market that compel adoption in the second. The entry’s distinctive requirement is the causal bridge between the two markets: dominance in the first, conduct capable of distorting competition in the second, and a connection between them. Conduct confined to one market, however aggressive, is tested under the other entries, and the leveraging clause is not a general warrant against conglomerate growth.

The third is the pricing squeeze in its several forms: prices to the dominant firm’s own downstream competitors set so that an efficient rival cannot match the firm’s retail price and survive, which the Commission examines through the discriminatory and unfair pricing entry and the denial-of-access entry together; and exclusionary rebates or loyalty conditions that make it commercially irrational for buyers to test a rival for even a part of their requirement, examined under the unfair conditions entry. The thread through all three situations is the Act’s refusal to condemn the dominant firm’s advantages as such. Scale, integration and incumbency appear in these cases as the source of the firm’s ability to act; liability still waits on the act, matched to the list, and on evidence that the act, rather than the firm’s efficiency, is what competition suffered from.

One absence in section 4 should be recorded, because practitioners trained on other systems look for it. The section contains no general justification clause, no provision excusing an abuse that meets competition, protects a legitimate interest or yields efficiencies sufficient to outweigh it. Justification enters Indian abuse law through the words of the list itself: a price is condemned as unfair or discriminatory, a condition as unfair, a price as predatory only where it is below cost and aimed at elimination, so that the commercial rationale of the conduct is litigated inside the elements of the abuse rather than after them. It enters also through the section 19(4) factors at the dominance stage, where contribution to economic development and countervailing power can deny the predicate. But there is no separate stage at which a proved abuse is forgiven for its benefits, and arguments framed that way have to be re-expressed in the statute’s own vocabulary before the Commission can act on them.

A word on the provision’s own amendment history completes its text. Section 4 as enacted in 2002 was addressed to the abuse of a dominant position by an enterprise. The Competition (Amendment) Act, 2007 substituted the wider frame under which an enterprise or a group may hold and abuse the position, and supplied the definition of group by reference to the control relationships the section 5 Explanation describes. The change closed an evident route of escape, under which the conduct of a dominant business carried on through separately incorporated members of one house could be fragmented for analysis into pieces none of which appeared dominant, and it aligned the unit of analysis throughout the Act: the same group concept that aggregates assets and turnover for the combination thresholds identifies the actor whose dominance is to be measured and whose abuse is to be restrained. Section 4(1) also carries a territorial rider often missed: it speaks of abuse of a dominant position in the relevant market in India, and the Explanation’s definition of the position is correspondingly Indian, so that dominance abroad without an Indian relevant market is outside the section, whatever the enterprise’s world position.

Two structural features of section 4 complete the picture. Section 4 applies to groups as well as single enterprises: the Act’s definition work in section 5’s Explanation and the 2007 amendment extended the provision so that enterprises forming part of the same group, exercising control over each other as there defined, can be assessed together for dominance and its abuse. And section 28 stands behind the whole provision as its ultimate remedy: where the Commission reports that a dominant enterprise has abused its position, the Central Government may direct the division of the enterprise, to ensure that it does not abuse that position, with the order providing for the transfer of property, rights, liabilities and contracts of the divided parts. Division is a remedy of the state on the Commission’s recommendation rather than an order the Commission passes itself, and its presence in the Act marks the outer edge of the status-and-conduct settlement: the Act will not punish dominance, but it reserves a structural instrument for the case where conduct remedies have been judged insufficient against it.

Must a dominant firm charge the same price to every buyer?

No. Section 4(2)(a) reaches an unfair or discriminatory condition or price, and the qualifying word is unfair. Prices that differ because costs, volumes, credit terms or delivery conditions differ are the ordinary texture of trade. The abuse lies in discrimination that lacks a commercial justification and operates to distort competition, not in differential pricing as such.

Division is the pillar’s reserved remedy, and it belongs to the same section of the argument. Under section 28, if the Commission finds abuse of a dominant position, the Central Government may direct the division of the enterprise to ensure that it does not abuse its dominant position. The direction settles the transfer and vesting of property, rights and liabilities and the adjustment of contracts between the resulting parts. The Commission recommends; the government divides. It is the Act’s only structural remedy and it sits in government hands, which is one reason it has remained a background possibility rather than an instrument of ordinary enforcement, directed at the state of the market rather than at any single act of the enterprise.

Sections 5 and 6: combinations, thresholds and the standstill

The third pillar works in the opposite direction from the first two. Sections 3 and 4 are invoked after conduct has occurred. Sections 5 and 6 intervene before a transaction completes, on the theory that a market structure created by merger cannot reliably be unmade afterwards, and that it is cheaper to examine a concentration in advance than to police its consequences for decades. The scheme has three moving parts: a definition of which transactions count as combinations, expressed through financial thresholds; a duty to notify and to wait; and an assessment, against its own list of factors, of whether the combination causes or is likely to cause an appreciable adverse effect on competition.

What counts as a combination

Section 5 defines combinations by transaction type and financial size together. It covers the acquisition of control, shares, voting rights or assets of one or more enterprises by a person or enterprise; the acquiring of control by a person over an enterprise where that person already has direct or indirect control over another enterprise engaged in identical business or a similar trade; and mergers and amalgamations between or among enterprises. Each type counts as a combination only where the parties, or the group to which the resulting enterprise will belong, cross the thresholds the section prescribes, expressed as values of assets and amounts of turnover, in India and worldwide with an India leg.

Three words in that account carry more weight than the thresholds themselves. The first is control. The Explanation to section 5 provides that control includes controlling the affairs or management by one or more enterprises, jointly or singly, over another enterprise or group, or by one or more groups over another group or enterprise. Control under this Act is not confined to majority shareholding; it is the ability to direct the affairs of the enterprise, and the Commission’s orders have examined veto rights, board composition and contractual protections to ask whether what is acquired is influence over strategic decisions or only protection of an investment. The second word is group: the Explanation treats two or more enterprises as a group where one is in a position to exercise twenty-six per cent or more of the voting rights in the other, or to appoint more than half of its board of directors, or to control its management or affairs, and group figures are aggregated for the threshold test, which is why a modest acquisition by a member of a very large group can be notifiable when the same acquisition by a standalone firm would not. The third is value: the Explanation fixes the value of assets by the audited books, the book value as reduced by depreciation, with intangible assets and goodwill included in the manner the provision specifies, so the threshold question is answered from accounts rather than from market estimates. Filing practice under the pillar is, in large part, the disciplined application of these three definitions to a transaction’s documents.

The thresholds operate through two alternative tests, and a transaction is a combination if either is met. The parties test aggregates the enterprise being acquired, taken over or merged with the enterprise of the acquirer or the merging parties: their assets and turnover, in India and worldwide, are placed side by side against the lower set of figures. The group test looks past the immediate parties to the group to which the acquired enterprise will belong after the combination, aggregating the whole group’s assets and turnover against the higher set of figures, four times the parties figures as enacted. The two tests catch different things. The parties test catches the large bilateral deal, two substantial enterprises joining directly. The group test catches the acquisition of a modest target by a very large group, where the competitive significance lies not in the target’s present size but in the resources, distribution and adjacent positions of the group it joins; a small company acquired into a conglomerate can change a market in ways its standalone figures do not predict, and the group test is the Act’s recognition of that fact. Counsel computing notifiability therefore work both columns, on both the asset and the turnover limbs, in India and worldwide, before concluding that a transaction is outside the pillar.

The thresholds as enacted in 2002 stood at the enterprise level at assets of more than one thousand crore rupees in India or turnover of more than three thousand crore rupees in India, or worldwide assets of more than five hundred million United States dollars including at least five hundred crore rupees in India, or worldwide turnover of more than fifteen hundred million dollars including at least fifteen hundred crore rupees in India; at the group level the figures were four times those amounts. Section 5 closes with a power in the Central Government, in consultation with the Commission, to revise the values by notification every two years on the basis of the wholesale price index or fluctuations in the exchange rate.

That power has been used, and the figures a filing party must apply are the notified figures, not the enacted ones. The thresholds were enhanced by fifty per cent by notification dated 4 March 2011, S.O. 480(E). They were doubled by notification dated 4 March 2016, S.O. 675(E), taking the enterprise levels to assets of more than two thousand crore rupees or turnover of more than six thousand crore rupees in India, and worldwide to one billion dollars of assets or three billion dollars of turnover with India legs of one thousand crore and three thousand crore rupees respectively. They were raised again by the notification of 8 March 2024, under which the enterprise thresholds stand at assets of more than two thousand five hundred crore rupees or turnover of more than seven thousand five hundred crore rupees in India; worldwide assets of more than one and a quarter billion dollars, including at least one thousand two hundred and fifty crore rupees in India, or worldwide turnover of more than three and three-quarter billion dollars including at least three thousand seven hundred and fifty crore rupees in India; and the group thresholds at assets of more than ten thousand crore rupees or turnover of more than thirty thousand crore rupees in India, or worldwide at five billion dollars of assets or fifteen billion dollars of turnover, in each case with the same India legs. Alongside the section 5 figures operates the small target exemption, a notification-based exemption for acquisitions whose target enterprise falls below prescribed asset and turnover values; on the 2024 notification figures the exemption protects a target with assets of not more than four hundred and fifty crore rupees or turnover of not more than one thousand two hundred and fifty crore rupees. Transactions counsel should be tested against the notification in force on the date of the trigger document, because the notified figures, having moved three times since enactment, are the operative law and the enacted figures are history.

The 2023 amendment added a further trigger of a different kind, stated here at the headline level the series assigns to this article: a transaction whose value exceeds two thousand crore rupees is treated as a combination where the target enterprise has substantial business operations in India, whether or not the asset and turnover thresholds of section 5 are crossed, and combinations on this deal value basis took effect from 10 September 2024. The amendment also shortened the outer review clock for combinations. The working of the deal value threshold and of substantial business operations belongs to the article on the competition amendment act 2023 explained, and is noticed here only so that a reader of the pillar does not mistake the threshold list for a closed one.

The duty to notify and to wait

Section 6(1) provides that no person or enterprise shall enter into a combination which causes or is likely to cause an appreciable adverse effect on competition within the relevant market in India, and that a combination which does so shall be void. Section 6(2) requires the person or enterprise proposing to enter into a combination to give notice to the Commission disclosing the details of the proposed combination within the period the section prescribes from approval of the proposal by the board of directors, in the case of a merger or amalgamation, or from execution of the agreement or other document, in the case of an acquisition. Section 6(2A) imposes the standstill: no combination shall come into effect until two hundred and ten days have passed from the day on which the notice has been given, or the Commission has passed orders under section 31, whichever is earlier. The notice is a document-heavy exercise, resting on audited accounts, shareholding charts and market data assembled in spreadsheet and word-processed schedules; parties and advisers who need a viewer for opening the Excel, Word and presentation files in the combination filing record will find the practice familiar. The 2023 amendment reduced that outer period to one hundred and fifty days, again a change owned in detail by the amendment’s own article. The scheme’s teeth against early closing is section 43A: where a person or enterprise fails to give notice of a combination, the Commission may impose a penalty which may extend to one per cent of the total turnover or the assets, whichever is higher, of the combination. This article explains the law; it is not advice on any transaction, and a party near these thresholds needs the notifications and the combination regulations as they stand on its own trigger date.

The foreign-to-foreign case follows from the thresholds’ own structure. Because the worldwide limbs of section 5 carry an India leg rather than an Indian-party requirement, a merger between two enterprises neither of which is Indian, neither of which has an Indian subsidiary of consequence in the target’s line of business, can still be notifiable where worldwide assets and turnover are large and the India legs are crossed; and a combination between foreign parties with substantial Indian sales through imports can meet the turnover test on the value of goods sold into India. Global transactions are therefore screened in India as a matter of course, the notice is filed by the acquirer in the prescribed form, and the Commission’s review concerns itself with the Indian markets the combination touches. The corollary is jurisdictional modesty in substance: where the Indian footprint is modest and the markets affected are worldwide, the Commission’s assessment is confined to the Indian effect, and its orders run, at their widest, to the combination’s Indian aspects, a pattern consistent with the effects principle on which section 32 is built.

Certain combinations are outside the notice duty by the Act’s own terms. Section 6(4) exempts the acquisition of shares or voting rights by a public financial institution, foreign institutional investor, bank or venture capital fund pursuant to a covenant of a loan or investment agreement. Section 6(5) keeps out share subscription and acquisitions by such institutions through underwriting and the like. The combination regulations made under section 63, the Competition Commission of India (Procedure in regard to the transaction of business relating to combinations) Regulations, carry a schedule of transaction categories that ordinarily do not need notice, covering among others minority acquisitions made solely as an investment or in the ordinary course of business below the prescribed proportion of voting rights, acquisitions of shares where the acquirer already holds fifty per cent or more, and intragroup transactions that do not change control. The regulations are subordinate legislation and change more often than the Act; their categories are noticed here so that the pillar’s account of the notice duty is not read as wider than the working law.

The scheduled categories repay attention because they govern the transactions businesses actually structure. The minority investment category protects acquisitions of less than twenty-five per cent of the shares or voting rights of a target, made solely as an investment or in the ordinary course of business, not leading to acquisition of control; its two conditions work together, so that a holding below the line accompanied by board rights, vetoes over strategic decisions or an understanding toward control falls outside the protection and must be notified in the ordinary way. The existing-majority category protects an acquirer that already holds fifty per cent or more of the target’s shares or voting rights, acquiring more, except where the joint control the acquirer shared becomes sole control, a change the scheme treats as real because it extinguishes a second centre of decision. The intragroup categories protect mergers and acquisitions between enterprises of the same group, and acquisitions of a group’s own enterprise by another member, where the transaction does not produce a change in control in substance; group reorganisations are the staple of this category and the reason the filing system’s volume is not larger than it is. Two further mechanics complete the picture. Interconnected transactions, a series of steps by which a single combination is achieved in parts, are treated under the regulations as one combination, so that the notice duty cannot be dissolved by slicing the deal. And where a hostile or open-offer acquisition makes prior disclosure of details impossible in the ordinary sequence, the regulations accommodate the filing on the basis of the public announcement, keeping the standstill intact while the information follows. The exemptions, read together, state the scheme’s philosophy a second time: the state reviews changes of control in markets, not movements of paper within a single decision-making unit or below the line of influence.

Assessment and orders: sections 20 and 31

Where notice is given, or where the Commission inquires on its own knowledge or information within the period section 20 allows, the question is the one section 20(4) frames. The Commission must have due regard to the actual and potential level of competition through imports in the market; the extent of barriers to entry; the level of combination in the market; the degree of countervailing power in the market; whether the parties are likely, by the combination, significantly and sustainably to increase prices or profit margins; the extent to which effective competition is likely to remain in the market; whether the combination will leave available substitutes for the goods or services; the market share, individually and combined, of the parties; the likelihood that the combination will result in the removal of a vigorous and effective competitor; the nature and extent of innovation; the relative advantage the combination will bring, by way of contribution to economic development; and whether the benefits of the combination outweigh the adverse impact, where such impact exists. The factors deliberately echo section 19(3) while adding the forward-looking questions that only a merger inquiry poses, about substitutes, innovation and competitors yet to matter.

Section 20 also preserves a residual jurisdiction over combinations that were never notified. The Commission may inquire, on its own knowledge or on information received, whether a combination has caused or is likely to cause an appreciable adverse effect, but the section bars the initiation of such an inquiry after the expiry of one year from the date on which the combination has taken effect. The one-year bar is the price of transactional certainty: parties that never filed face a defined exposure window rather than a perpetual one, and the Commission’s practical answer to the window is the section 43A penalty, which is not time-barred in the same way and which has made unnotified closings a poor bargain even where the substantive review would have been favourable. Turnover, for the threshold and penalty computations alike, bears the meaning section 2(y) gives it, the value of goods sold or services rendered, and the figures are taken from the audited accounts in the manner the section 5 Explanation prescribes.

Section 31 gives the outcomes. The Commission may approve the combination; approve it subject to modification, which the parties may accept, or pursue through the response and amendment procedure the section lays down; or direct that the combination shall not take effect, where it finds that the combination has caused or is likely to cause an appreciable adverse effect. Silence also decides: if the Commission does not pass an order within the outer period from notice, the combination is deemed to have been approved. In practice the great majority of notified combinations are approved without modification, frequently at the first stage of review; a small number are approved with behavioural or structural modifications offered by the parties; and directions that a combination shall not take effect are the rare terminus of the scheme. The regulations also provide a voluntary green channel under which specified combinations with no horizontal, vertical or complementary overlaps receive deemed approval on filing, a procedural route that has become the ordinary path for transactions without overlaps. The institutional detail of review, timelines as amended in 2023 and the Commission’s merger practice belong to the sibling articles; the pillar’s point is the legal structure: thresholds trigger notice, notice triggers the standstill, and the section 20(4) factors decide.

The form of the notice follows the same logic of proportion. The combination regulations provide a short form, Form I, for the ordinary case, and a long form, Form II, where the post-combination market share exceeds fifteen per cent in a market where the parties overlap horizontally, or twenty-five per cent where their relationship is vertical; the long form demands the detailed market and effects evidence on which a contested assessment will turn. Between notice and order, section 29 organises the contested path: the Commission forms a prima facie opinion on whether the combination is likely to cause an appreciable adverse effect; where the opinion is adverse, the parties are called upon to show cause, a report may be called for, objections from persons affected or interested are invited through the publication the Act prescribes, and the matter proceeds to the section 31 order, with section 30 governing the intermediate procedure. The modification route within section 31 is where most contested Indian merger work is done: parties propose divestments of overlapping businesses, licensing arrangements, or behavioural commitments addressed to the identified concern, the Commission tests whether the modification removes the appreciable adverse effect, and approval follows on the modified terms if it does. A modification accepted under section 31 is then binding on the parties as the condition of their clearance, and closing in breach of it re-opens the combination to the Act’s penalties.

The notice clock itself has been rebuilt once and should be stated in its two states. As enacted and first operated, section 6(2) required notice within thirty days of the trigger, the board’s approval of a merger proposal or the execution of the acquisition agreement, and the standstill ran for two hundred and ten days from the giving of notice or until the Commission’s order, whichever was earlier; a filing made late was itself a contravention exposed to section 43A. The 2023 amendment recast the duty: notice is to be given before the combination takes effect, and the outer standstill stands reduced to one hundred and fifty days. The pillar records the original design because the case law and the penalty orders under section 43A were built on it, and the amendment’s article records the transition; parties filing under the scheme must work from the text as amended, and the transitional treatment of notices filed across the change is prescribed in the amending scheme itself.

Does a small acquisition need approval?

Ordinarily no. A transaction must first be a combination within section 5, crossing the notified asset or turnover thresholds at the level of the parties or their group, or the deal value threshold with substantial business operations in India. Beneath those lines there is no filing, no standstill and nothing to approve, whatever the size of the acquirer.

Two qualifications keep the short answer honest. The small target exemption protects acquisitions whose target falls below the notified asset and turnover values even where the acquirer’s side of the account is large, because the thresholds are applied to the combination’s parties in the manner the notifications and regulations prescribe, and a genuinely small target takes the transaction out. And exemption from merger control is not exemption from the Act: the parties to an unnotified, sub-threshold transaction remain fully exposed to section 3 if they implement it through a prohibited agreement and to section 4 if the result is an abused dominance.

Inquiry, orders and penalties

A prohibition without a procedure is a sentiment, and the Act’s enforcement machinery is concentrated in sections 19, 26, 27 and the penalty provisions that follow them. The Commission may inquire into an alleged contravention of section 3 or section 4 on receipt of information from any person, consumer or their association or trade association; on a reference made by the Central Government, a State Government or a statutory authority; or on its own knowledge. Where the Commission is of the opinion that a prima facie case exists, section 26 directs the Director General, the investigation office established under section 16 and headed by an officer appointed under it, to investigate and report. The Supreme Court settled the character of that step in Competition Commission of India v. Steel Authority of India Limited, (2010) 10 SCC 744: a direction to the Director General under section 26(1) is an administrative direction that sets an investigation in motion, not an adjudication of anyone’s rights, and it is not appealable as if it were a finding. The merits are decided later, by the Commission, after the report, hearing and the opportunity section 26 and the regulations provide.

The path of an inquiry

The route from information to order is worth stating as a sequence, because each stage allocates rights that matter to the parties. Section 19(1) opens the inquiry on the three triggers already described, and the first filter is the prima facie opinion: if the Commission forms no prima facie view of contravention, the matter closes at the threshold under section 26(2), the closure is communicated to the informant and to the governments or authorities concerned, and the informant’s remedy lies in the appeal the Act provides against such a closure. The closure power is the Act’s answer to the speculative information: a large share of the informations filed under sections 3 and 4 end at this stage, and the discipline of the prima facie order, reasons recorded in the file and a stated application of the statutory factors, is what makes the closure system administrable rather than arbitrary. An information that survives it has crossed the Act’s first real threshold, and everything after it is machinery for testing whether the prima facie impression was right. If the opinion is formed, section 26(1) sends the matter to the Director General, whose office investigates with the powers section 41 confers, to summon and enforce attendance, to require discovery and production of documents, to receive evidence on affidavit and to requisition records from public authorities, in the manner of a civil court under the Code of Civil Procedure. The Director General reports findings, not conclusions of law; the report recommends, and it is the Commission that decides. Section 26 then works the report through objection and hearing: the report goes to the parties, to the governments or authorities that referred the matter, their responses are taken, further inquiry may be directed where the Commission disagrees with the report’s findings, and the order under section 27 is passed after the parties have been heard. The Commission’s own procedural powers during the inquiry are gathered in section 36, which clothes it with civil court powers for summoning witnesses, examining on oath and receiving evidence, while expressly releasing it from the procedure of the Code, subject to the principles of natural justice and its own regulations.

This architecture explains both of the section’s celebrated limits. Investigation and adjudication are deliberately separated, so that the body that prosecutes the inquiry is not the body that decides it; the cost of the separation is the time the two stages consume. And the whole process is inquisitorial in form: there is no plaintiff and no defendant in the strict sense, an informant is a source of information rather than a party carrying a burden of proof, and the Commission’s obligation is to the state of competition in the market, not to the settlement of the informant’s private grievance, which belongs to the compensation route the Act separately provides in section 53N for persons who have suffered loss from a found contravention. Readers who approach the Commission’s orders expecting the shape of a civil judgment will misread them; they are regulatory determinations, appealable as such, and their authority stands or falls on the adequacy of the investigation record and the reasoning that connects it to the statutory factors.

Where the Commission finds a contravention, section 27 arms it with orders. It may direct the parties to discontinue the agreement or the abuse and not to re-enter into it. It may direct modification of the offending agreement. It may pass such other orders and issue such directions as it may deem fit, including directions about costs. And it may impose penalties, on the scale the provision sets.

The penalty scale is the provision’s sharpest edge. Under section 27(b), the Commission may impose on each producer, seller, distributor, trader or service provider included in the contravention such penalty, not exceeding ten per cent of the average of the turnover for the last three preceding financial years, as it may determine. For cartels, the proviso goes further: the penalty may extend to up to three times the amount of profit made out of the agreement in question, or ten per cent of the turnover for each year of the continuance of the agreement, whichever is higher, imposed on each member of the cartel. The percentage formulation conceals a question that produced one of the Act’s defining appeals: ten per cent of what, for a company that makes many products, where the contravention touches only one of them. In Excel Crop Care Limited v. Competition Commission of India, (2017) 8 SCC 47, the Supreme Court read the word turnover in section 27(b) as relevant turnover, the turnover of the products and services covered by the contravention, holding that a penalty computed on the whole turnover of a multi-product enterprise, for a contravention confined to one product line, would offend proportionality and the doctrine of harmonious construction on which the provision must be read. The 2023 amendment has since displaced that construction by re-basing the penalty on the enterprise’s global turnover derived from all its products and services, a change examined in the amendment’s own article and noticed here because a reader of the penalty section who stops at the judgment will misstate the position the statute has reached. Penalties also attach to the machinery itself: section 43A carries the gun-jumping penalty described above; section 44 penalises false statements and material omissions made to the Commission; section 42 penalises contravention of the Commission’s orders, with daily penalties for continuing default and imprisonment in the last resort for unpaid penalties; and section 48 makes a person in charge of a company’s business answerable for the company’s contravention unless the statutory defences are made out.

Section 48 deserves its own sentence of explanation, because it converts the corporate contravention into personal exposure. Where a contravention is committed by a company, every person who, at the time of the contravention, was in charge of and responsible to the company for the conduct of its business, as well as the company itself, is deemed guilty of the contravention and liable to the penalty, unless that person proves that the contravention was committed without their knowledge or that they exercised all due diligence to prevent it. Directors, managers, secretaries and other officers are additionally liable where the contravention is shown to have taken place with their consent or connivance or to be attributable to their neglect. The provision’s practical effect is felt in cartel cases, where the individuals who attended the meeting or sent the message are proceeded against alongside the enterprise, and where the leniency queue therefore matters to persons as well as to balance sheets. Behind the penalties stands a compensatory route of a different character: under section 53N, a person who has suffered loss or damage from a contravention found by the Commission, or established in appeal, may apply to the Appellate Tribunal for compensation, so that a public finding can be converted into private recovery by those the conduct injured, including the purchasers who paid a cartelised price.

Section 33 adds the interim instrument. Where during an inquiry it is proved that an act in contravention of section 3(1), section 4(1) or section 6 has been committed and continues to be committed, or is about to be committed, the Commission may, if satisfied that appreciable prejudice is likely to be caused to the interest of consumers or the market, temporarily restrain the party from carrying on the act until the conclusion of the inquiry or until further orders, after recording reasons and without the prior notice that the final process requires. The power is exercised sparingly, and its sparingness follows from its conditions: the Commission must be satisfied on material placed before it at a stage when the investigation is incomplete, and an interim restraint against conduct that the final order may exonerate inflicts the very market distortion the Act exists to prevent, in reverse. The grants of interim relief under the Act are correspondingly few, and the provision functions in practice as a reserve for the case of continuing, demonstrable harm that cannot await the inquiry’s end.

The scale in practice is therefore two-tiered. For an ordinary contravention the ceiling in section 27(b) is ten per cent of the average turnover for the last three preceding financial years. For a cartel the proviso raises it to three times the profit made out of the agreement, or ten per cent of turnover for each year of the continuance of the agreement, whichever is higher, for each member of the cartel. Which turnover the percentage bites on was settled as relevant turnover by Excel Crop Care and then re-settled, by the 2023 amendment, on a global turnover base.

Leniency: section 46

Section 46 is the Act’s answer to the secrecy that protects cartels. The Commission may, if satisfied that any producer, seller, distributor, trader or service provider included in an alleged cartel has made a full and true disclosure in respect of the alleged violations and that the disclosure is vital, impose a lesser penalty than leviable under the Act or the rules and regulations. Two conditions mark the provision’s limits. The privilege is withdrawn where the applicant does not continue to cooperate until the proceedings complete, or has not complied with the conditions on which the lesser penalty was granted, or has given false evidence. And the section is addressed to cartels: it does not create a general amnesty for unilateral abuse or for vertical restraints. The Lesser Penalty Regulations, 2009 proceduralise the offer into a queue. The first applicant whose full and true disclosure is vital, and who continues to cooperate, may receive a reduction extending to the whole of the penalty. An applicant marked second in priority may be granted a reduction of up to fifty per cent, and applicants marked third or later a reduction of up to thirty per cent, in each case where the disclosure adds significant value to the evidence the Commission already holds. Priority is marked by the order of application, is held while the applicant perfects its disclosure within the period the regulations allow, and passes down the queue if the applicant ahead is rejected. The arithmetic is the policy: the distance between one hundred and fifty per cent is the reward for breaking first, and it is what makes every cartel meeting after an investigation begins a negotiation among potential applicants rather than a council of war. The 2023 amendment and the regulations that followed it added a leniency plus mechanism, under which an applicant that discloses a second, separate cartel can earn an additional reduction; that development belongs to the amendment’s article. Within this pillar, section 46’s function is structural: it converts the cartel’s own stability problem into the enforcement authority’s instrument, since every member of a price-fixing ring knows that the first to disclose may pay little or nothing while the rest pay on the proviso’s scale.

Two working features of the regulations complete the account. The application may be made by an enterprise or by an individual on its behalf, and the priority contest is frequently run by individuals whose personal exposure under section 48 gives them the sharper incentive; the regulations accordingly extend the lesser penalty to the individuals named in a successful application who were in charge of the cartel’s operation for the applicant. And the applicant’s identity and submissions are treated as confidential in the course of the investigation, protected against disclosure to the other parties except to the extent the inquiry’s fairness requires, because a leniency regime whose applicants are exposed at the threshold would receive no applications. The confidentiality is procedural protection, not anonymity against the final order: the Commission’s order, when it comes, records the grant of the lesser penalty and the reasoning, and the applicant’s cooperation is part of the public record of the case it helped to make.

Conduct abroad: section 32

Section 32 extends the Act’s inquiry power to acts taking place outside India. The Commission may inquire into an agreement of the kind referred to in section 3, an enterprise in a dominant position, or a combination, where the agreement, the position or the combination has taken place outside India but has, or is likely to have, an appreciable adverse effect on competition in the relevant market in India. The provision is the effects doctrine written into an Indian statute: foreign conduct is not regulated for its own sake, and the Commission’s concern begins only where the consequence lands in an Indian market. The practical limits are evidentiary and coercive rather than textual, because the Commission’s summons and its penalties must operate against persons and assets within reach, but the textual grant itself is unqualified once the effect is shown.

The reach of section 32, stated plainly, is this: the Commission may inquire into a foreign agreement, a foreign dominant enterprise or a foreign combination where the conduct, though occurring outside India, has or is likely to have an appreciable adverse effect on competition in the relevant market in India. The connecting factor is the effect in India, not the nationality of the parties or the place where the agreement was made. Global combinations are, in consequence, notified in India as a matter of routine where the section 5 thresholds are met on worldwide figures with an India leg, and a foreign cartel selling into India faces the same proviso scale under section 27 as a domestic one, subject always to the practical reach of the Commission’s processes against the persons concerned.

The judicial spine: four decided propositions

The text of the Act has been worked on by the courts at four points where its meaning was genuinely open, and a pillar account should collect the four propositions in one place, because together they fix the character of the machine: who decides, by what kind of act, on what measure of penalty, and in what order against other regulators.

The first proposition is Brahm Dutt’s legacy about design. The Commission as enacted in 2002 united functions that the constitutional tradition usually separates: it was to advise the government, regulate markets and adjudicate contraventions, imposing penalties of great size, under a selection process controlled by the executive, with members who need not have held judicial office. The challenge in Brahm Dutt v. Union of India fastened on that combination. The Supreme Court did not strike the Act down; the Union instead undertook to amend, and the 2007 amendment’s division of labour, an expert Commission to regulate and investigate and an appellate tribunal with judicial membership to hear appeals, is the settlement the litigation produced. The proposition the case leaves in force is structural: the Act’s adjudicatory output is legitimate because it is appealable to a judicially constituted tribunal, and the Commission’s own character is regulatory and inquisitorial rather than curial. Everything in the case law that resists treating the Commission as a court, or its processes as trials, descends from this settlement.

The second proposition is Steel Authority’s, about the nature of the Commission’s first coercive step. A direction under section 26(1) sends the Director General to investigate, and respondents repeatedly sought to appeal or quash such directions as if they were determinations. The Supreme Court held that the direction is administrative: it decides nothing about the respondent’s rights, records only that a prima facie case exists warranting inquiry, and is not an order of the kind the appeal section reaches. The proposition does double work. It keeps the investigation stage from being litigated to a standstill before evidence exists, which is what an appealable direction would invite; and it locates the respondent’s protections where the design puts them, in the hearing before the final order, in the requirement of reasoned findings, and in the appeal that then lies. Bharti Airtel later reaffirmed the proposition in terms, and it is the settled premise of every section 26 direction given since.

The third proposition is Excel Crop Care’s, about the measure of penalty under section 27(b). The Court’s reasoning proceeded in two movements. Textually, the provision penalises contraventions, and in a multi-product enterprise the contravention touches identified products or services; reading the penalty base as the enterprise’s entire turnover detaches the punishment from the wrong. Constitutionally and doctrinally, proportionality governs penal discretion of this amplitude, and a construction that exposes an enterprise to a penalty computed on turnover unrelated to the contravention would be unreasonable and liable to be struck down; the Court therefore read turnover down to relevant turnover, the aggregate value of the sales of the products and services covered by the contravention. The proposition governed penalty computation from 2017 until the 2023 amendment replaced the base with global turnover by legislation, a legislative reversal that confirms rather than disturbs the judgment’s account of what the unamended text meant.

The fourth proposition is Bharti Airtel’s, about sequence between regulators. The Court did not hold that the telecom regulator’s statute displaces the Competition Act; it expressly preserved the Commission’s jurisdiction and the Act’s overriding clause. What it held is that where the facts on which jurisdiction depends are themselves technical questions committed to a sectoral regulator, arising under that regulator’s own instruments and calling for its expertise, the competition inquiry waits until the regulator has determined them; the Commission then proceeds on the facts as so determined, applying competition law to them. The distinction the Court drew against the patent-licensing case, Telefonaktiebolaget LM Ericsson v. Competition Commission of India, completes the proposition: a controller deciding the rights of individual parties under a statute is not a sectoral regulator performing the jurisdictional-fact function, so no waiting is required there. Together the two cases draw the working map of sections 21 and 21A: reference and consultation between authorities, sequence where a regulator’s technical determination is anterior, and no exclusion of the competition statute anywhere.

Read together, the four propositions describe a coherent judicial attitude. The courts have protected the Act’s enforcement capacity at the points where it was attacked, the unappealable direction, the penalty power, the jurisdictional reach, while disciplining its exercise at the points where power meets the individual case: proportionality in penalty, anterior expertise in regulated sectors, a judicial tribunal above the Commission. A reader who holds the four propositions can predict the shape of most appellate outcomes under the Act, which is why this article collects them as the spine of the pillar.

The Commission, sector regulators and the reach of the Act

The Commission itself is established by section 7 as a body corporate with perpetual succession, consisting under section 8 of a Chairperson and not fewer than two and not more than six whole-time members appointed by the Central Government, chosen through the selection process section 9 prescribes. Its duties under section 18 are stated as the elimination of practices having an adverse effect on competition, the promotion and sustenance of competition, the protection of the interests of consumers and the assurance of freedom of trade carried on by other participants in markets in India. The institutional anatomy of the Commission, the Director General’s office, the appellate route and the record the institution has built since 2009 are the subject of the companion article on the CCI’s antitrust enforcement in India; what this pillar needs from section 7 and its companions is only the legal proposition that the prohibitions are administered by a specialist regulator whose orders are appealable to the Appellate Tribunal under section 53B, and whose civil court rivals are excluded by section 61.

Two boundary questions follow. The first is the boundary with sectoral regulators. Indian markets are thick with them: telecommunications and broadcasting are regulated by the Telecom Regulatory Authority of India, whose establishing Act belongs to an earlier cluster of this series, and securities, insurance, banking and electricity have regulators of their own. Sections 21 and 21A create the traffic rules. Where an issue before a statutory authority involves a determination that may contravene this Act, the authority may refer the issue to the Commission, and the Commission in turn may refer to the statutory authority an issue that falls within that authority’s competence; the opinions exchanged are binding in neither direction under the original design, the receiving body deciding for itself. The Supreme Court tested the arrangement in Competition Commission of India v. Bharti Airtel Limited, decided on 5 December 2018 in Civil Appeal No. 11843 of 2018. The Commission had ordered investigation into whether the major mobile operators and their association had coordinated against a new entrant over points of interconnection, a subject on which the telecom regulator’s regulations and expertise bore directly. The Court set aside the investigation order at that stage, holding that the regulator must first determine the jurisdictional facts within its own technical field, after which the Commission’s inquiry could proceed; the Commission’s jurisdiction was postponed, not ousted, and the Court reaffirmed both the Steel Authority proposition on the administrative character of a section 26(1) direction and the distinction between a sectoral regulator’s role and that of an authority, such as a patent controller, deciding individual rights. The working rule after Bharti Airtel is sequence rather than exclusion: the specialist regulator finishes the technical determination its statute assigns to it, and competition law, including the article on the TRAI Act and telecom regulation in India, then operates on the facts so found. Section 60 gives the Act overriding effect to the extent of any inconsistency with another law, and section 62 states that its provisions are in addition to and not in derogation of other laws; read together with Bharti Airtel, the sections describe a statute that claims the last word on competition questions while leaving other regulators their first word on their own.

The reference machinery deserves a word of its own, because Bharti Airtel is its gloss rather than its source. Under section 21, where in the course of a proceeding before a statutory authority an issue arises involving a determination that might contravene this Act, the authority may make a reference to the Commission, which is to give its opinion within sixty days; the authority then proceeds, taking the opinion into account, and records its reasons where it departs from it. Section 21A runs in the opposite direction: where in the course of a proceeding before the Commission an issue arises that falls within a statutory authority’s competence, the Commission may refer it, receives the opinion within the same period, and decides taking it into account. The opinions bind neither side, and the design is deliberate: each authority keeps its own statute, and coordination is achieved by mandatory consultation in form, and by the appellate discipline of reasoned departure in fact. Bharti Airtel added the sequencing rule for the harder case, where the reference machinery is not used but both jurisdictions are engaged at once: the sectoral regulator’s technical determination comes first, and the Commission applies competition law to the market as that determination establishes it.

The second boundary is the boundary with the state itself. Section 2(h) defines enterprise to include a department of government engaged in an activity relating to production, storage, supply, distribution, acquisition or control of goods, or provision of services, while excluding the sovereign functions of government, among them atomic energy, currency, defence and space. When Coal India Limited resisted the Commission’s jurisdiction, its position rested substantially on its statutory character as the nationalised coal producer; the challenge failed, and the proceedings against it proceeded on the footing that a statutory monopoly carrying on commercial activity is an enterprise within the Act. The Act’s test is functional: what matters is whether the body is carrying on activities of the kind an enterprise carries on, not the source of its charter or the ownership of its shares.

The predecessor at headline level

The repeal that section 66 performed on 1 September 2009 closed a forty-year experiment and needs only its outline here, since the companion article on the MRTP Act’s repeal owns its mechanics. The Monopolies and Restrictive Trade Practices Act, 1969 is repealed by section 66 of this Act; the MRTP Commission ceased to take new cases from the repeal date, continued for two years to dispose of cases pending before it, and then stood dissolved; investigations pending with its Director General passed to the Commission under this Act, monopolistic and restrictive trade practice cases passed to the Appellate Tribunal, and cases that were purely of unfair trade practice passed to the consumer commissions under the Consumer Protection Act, 1986. The Competition (Amendment) Act, 2009, Act 39 of 2009, adjusted the transfer scheme in fine detail. The substance of the succession is a change of question: the old Act asked whether an undertaking was large enough to worry the state, and the new one asks what an undertaking has done in its market.

The advisory half of the design

The Act assigns the Commission one function that prohibits nothing and decides nothing, and its presence in the text completes the Raghavan design. Section 49 provides that where the Central Government or a State Government, in formulating a policy on competition or on any other matter, refers the matter to the Commission for its opinion on the possible effect of the policy on competition, the Commission shall give its opinion within sixty days of receiving the reference, after hearing the parties concerned. The government is to consider the opinion in formulating the policy; the opinion does not bind it. Section 18’s statement of duties places the promotion and sustenance of competition, and the assurance of freedom of trade, on the same footing as the elimination of anti-competitive practices, which is the textual home of the advocacy work the Commission does outside litigation: studies of markets, comments on draft legislation and regulation, and the published articulation of how its enforcement powers will approach a sector. Section 64’s regulation-making and the Commission’s general power to issue guidance belong to the same half of the design.

The advisory function matters to a reading of the prohibitions, because it marks the boundary of the Commission’s legal personality. On policy, the Commission persuades; on contraventions, it commands. An opinion under section 49 creates no obligation and settles no rights, while an order under section 27 does both, and the Act keeps the two instruments in separate chapters with separate procedures and separate legal consequences. Confusion between them, treating advocacy publications as if they were law, or enforcement orders as if they were advice, is a recurring error in commentary on the Act, and the pillar’s discipline answers it in the usual way: ask which provision is being exercised, and what follows from it under the text.

Assessment: what the Act forbids and what it deliberately does not

Read as a whole, the Competition Act is a statute of deliberate abstentions. It abstains from prohibiting dominance, because its framers accepted that market power is frequently the prize that competition awards to efficiency, and that a law punishing the prize would tax the behaviour it wants to produce. It abstains from condemning vertical restraints without proof of effect, because the same contract that forecloses in one market disciplines distribution in another. It abstains from reviewing combinations below notified thresholds, accepting the residual risk of small concentrations as the price of keeping the filing system proportionate to the transactions that matter. And for forty years of accumulated habit under the predecessor statute, it substitutes a single organising distinction, between status and conduct, that the Raghavan Committee recommended and Parliament enacted: the state does not care how big a firm is; it cares what the firm agrees, what it does with dominance, and what it buys.

The penalty history of the Act, gathered in one view, is the clearest record of its growing pains. The text of 2002 fixed ceilings as percentages of an undefined word, turnover, and said nothing about the products to which the percentage should be applied. Excel Crop Care supplied the omission judicially in 2017, fourteen years after assent, tying the base to the contravention. The 2023 amendment then reversed the tie by legislation and substituted a global base, while the Commission acquired, in the same amendment, the settlement and commitment machinery it had lacked, and merger control acquired the deal value threshold that the digital economy’s acquisitions had made necessary. Each change was defensible in itself. Together they mean that the operative law of penalties and thresholds at any moment is a compound of the enacted text, the notifications in force, the judgments unreversed and the amendments commenced, and that any account of the Act, including this one, must be read with its date in mind: the pillar states the architecture, and the architecture has proved more durable than any of the instruments hung on it.

The abstentions have a cost that honest assessment should record. An effects-based law is an expensive law to administer, because every prohibition except the section 3(3) core requires economic evidence about markets that must first be defined and then weighed, and the Commission’s resources and the appellate system’s time are spent substantially on preliminary questions of market and dominance before the merits are reached. The staged commencement left India without operative competition prohibitions through most of a decade of liberalised growth. And the penalty architecture has been rebuilt twice, by judgment and then by amendment, which tells its own story about the difficulty of expressing deterrence as a percentage of a word, turnover, that the original text left undefined. None of this disturbs the settlement the article set out to state. The Act’s prohibitions are narrow because they were meant to be, and the correct way to test any claim made about the Competition Act 2002 is to return to the three pillars and ask which of them the claim belongs to, and whether the claim respects the burden and the test that pillar carries.

Frequently Asked Questions

Q: What does the Competition Act 2002 prohibit?

The Act prohibits three classes of conduct. Agreements that cause or are likely to cause an appreciable adverse effect on competition in India are prohibited by section 3, with four categories of horizontal agreement presumed to have that effect. Abuse of a dominant position is prohibited by section 4, dominance itself being lawful. Combinations above the prescribed thresholds that are not notified, or that take effect without approval, are regulated by sections 5 and 6. Nothing else is prohibited: size, market share and commercial success, without a listed agreement, abuse or unnotified combination, contravene no provision of the Act.

Q: Which agreements are presumed anti-competitive?

Section 3(3) presumes an appreciable adverse effect for agreements between enterprises in identical or similar trade, including cartels, that directly or indirectly determine purchase or sale prices; limit or control production, supply, markets, technical development, investment or provision of services; share the market or source of production by area, type of goods or services, or number of customers; or engage in bid rigging or collusive bidding. The presumption shifts the burden to the parties to displace it, and it does not apply to an efficiency-increasing joint venture between the parties.

Q: How is dominance determined?

Dominance is determined within a relevant market, defined by its product and geographic dimensions under sections 19(5), 19(6) and 19(7). The Commission then applies the factors in section 19(4): market share, the size and resources of the enterprise, the size and importance of competitors, economic power and commercial advantages, vertical integration, dependence of consumers, statutory or public sector sources of the position, entry barriers, countervailing buying power, market structure, social obligations and costs, relative advantage by way of contribution to economic development, and any other relevant factor. Market share opens the inquiry; it does not close it.

Q: What counts as abuse of a dominant position?

Section 4(2) lists five abuses: imposing unfair or discriminatory conditions or prices, including predatory prices; limiting or restricting production, services or the market, or technical and scientific development to the prejudice of consumers; practices resulting in denial of market access; making contracts subject to supplementary obligations unconnected with their subject; and using dominance in one relevant market to enter into or protect another relevant market. Conduct by a dominant enterprise outside this list is not an abuse under the Act, and conduct within the list by an enterprise that is not dominant is equally outside section 4.

Q: When must a merger be notified to the regulator?

Notice must be given where the transaction is a combination under section 5: an acquisition of control, shares, voting rights or assets, an acquisition of control over an enterprise by a person already controlling a similar business, or a merger or amalgamation, crossing the asset and turnover thresholds in force by notification, or, under the 2023 amendment, the deal value threshold of two thousand crore rupees where the target has substantial business operations in India. Notice follows the board approval or the executed agreement, and the combination cannot take effect until the Commission orders under section 31 or the statutory waiting period runs out.

Q: What penalties can be imposed under the Competition Act?

For contraventions of sections 3 and 4, section 27(b) permits a penalty of up to ten per cent of the average turnover for the last three preceding financial years. For cartels, the proviso permits up to three times the profit from the agreement or ten per cent of turnover for each year of the agreement’s continuance, whichever is higher. Failure to notify a combination attracts up to one per cent of the combination’s total turnover or assets under section 43A. Separate penalties attach to false statements, contravention of orders and offences by persons in charge of companies.

Q: What is the leniency programme for cartels?

Section 46 permits the Commission to impose a lesser penalty on a member of an alleged cartel that has made a full and true disclosure of the alleged violations, where the disclosure is vital to the case. The reduction can extend to the whole penalty for the first applicant that qualifies, with lesser reductions for later applicants whose disclosures add significant value under the Lesser Penalty Regulations, 2009. The protection is forfeited if the applicant stops cooperating, breaches the conditions imposed, or gives false evidence. The programme exists because cartels are secret, and it rewards the member that breaks the secrecy first.

Q: Does the Competition Act apply to conduct outside India?

Yes. Section 32 empowers the Commission to inquire into an agreement, a dominant position or a combination situated outside India where that agreement, position or combination has, or is likely to have, an appreciable adverse effect on competition in the relevant market in India. The inquiry is not into foreign conduct as such but into its Indian effect. A cartel organised abroad that fixes the price of goods sold into India, or a foreign merger that removes the principal competitive constraint in an Indian market, is within the Commission’s reach on that footing.

Q: What does appreciable adverse effect on competition mean?

It is the Act’s central standard, the effect that sections 3 and 6 prohibit and that combination review exists to prevent. The Act does not define it by a formula; section 19(3) supplies the factors by which it is judged: the creation of barriers to new entrants, the driving of existing competitors out of the market, the foreclosure of competition by hindering entry, set against benefits to consumers, improvements in production or distribution, and the promotion of technical, scientific and economic development. An appreciable adverse effect is found where, weighing those factors, the agreement or combination can be expected to leave competition in the market materially worse than it would otherwise have been.

Q: How is the relevant market determined?

Under section 2(r) a relevant market has a product dimension and a geographic dimension. For the product market, section 19(7) directs regard to the physical characteristics and end-use of the goods, price, consumer preferences, the exclusion of in-house production, the existence of specialised producers and the classification of industrial products; the working question is which products buyers treat as substitutes. For the geographic market, section 19(6) directs regard to regulatory trade barriers, local specification requirements, procurement policies, distribution facilities, transport costs, language, consumer preferences and the need for secure and regular supply. Both dimensions are fixed on evidence in each case, because the boundaries decide dominance and effect.

Q: Are vertical agreements such as resale price maintenance assessed or presumed unlawful?

They are assessed. The five categories in section 3(4), tie-in arrangements, exclusive supply agreements, exclusive distribution agreements, refusal to deal and resale price maintenance, contravene the Act only where they cause or are likely to cause an appreciable adverse effect on competition, judged on the section 19(3) factors. Resale price maintenance is therefore not banned as such in India; a stipulated resale price imposed by a supplier with a modest market share and vigorous rivals will ordinarily pass, while the same stipulation imposed by a supplier whose arrangements cover most of a market, foreclosing rivals or suppressing price competition between its own dealers, can fail the assessment.

Q: When does the joint venture proviso protect an agreement between competitors?

The proviso to section 3(3) lifts the statutory presumption from an agreement entered into by way of a joint venture if the venture increases efficiency in production, supply, distribution, storage, acquisition or control of goods, or provision of services. The parties invoking it should be organised as a genuine joint venture and be able to show the efficiency: shared production that neither could finance alone, common infrastructure, combined research. The proviso answers the presumption, not the general prohibition, so a venture whose coordination extends to prices or markets beyond what the efficiency requires remains exposed to section 3 on its effects.

Q: Does section 3 restrict conditions imposed to protect intellectual property rights?

No, within the limits of section 3(5). Section 3 does not restrict the right of any person to restrain an infringement of, or to impose reasonable conditions necessary for protecting, rights under the Copyright Act, 1957, the Patents Act, 1970, the Trade Marks Act, 1999, the Geographical Indications of Goods (Registration and Protection) Act, 1999, the Designs Act, 2000 and the Semi-conductor Integrated Circuits Layout-Design Act, 2000. Conditions that go beyond what protection of the right requires lose the shelter, because the exception is confined to conditions that are reasonable and necessary for that protection.

Q: What counts as a cartel under the Act?

Section 2(c) defines a cartel to include 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 the provision of services. Cartels are the core of the section 3(3) categories: price determination, market sharing and bid rigging are the characteristic cartel behaviours. The definition’s inclusion of an attempt to control matters, because an arrangement that fails to hold still falls within the class the presumption and the cartel penalty proviso address.

Q: What is predatory pricing under the Act?

The Explanation to section 4 defines a predatory price as the sale of goods or provision of services at a price below the cost, as may be determined by regulations, of producing the goods or providing the services, made with a view to reducing competition or eliminating competitors. It is actionable only as an abuse under section 4(2)(a), which means the seller must first be dominant in the relevant market. Price cutting by an entrant or by any non-dominant firm is competition, not predation, under this Act, and even a dominant firm’s low price requires the below-cost measure and the eliminative purpose to be shown.

Q: What happens if a combination is closed without the Commission’s approval?

Closing early breaks the standstill in section 6(2A) and, where notice itself was not given, the duty in section 6(2). Section 43A empowers the Commission to impose a penalty of up to one per cent of the total turnover or the assets, whichever is higher, of the combination, on the person or enterprise that failed to notify. The transaction is also exposed on its substance: a combination that causes an appreciable adverse effect is void under section 6(1), and the Commission retains its power under section 20 to inquire into a combination after the event within the period the section allows.

Q: Can the Commission order the division of a dominant enterprise?

The Commission cannot divide an enterprise by its own order, but section 28 gives the Act a structural remedy one step removed. Where the Commission reports a case of abuse of dominant position, the Central Government may direct the division of the enterprise enjoying the dominant position to ensure that it does not abuse that position, and the direction settles the transfer of property, rights and liabilities and the adjustment of contracts between the parts. Division is the reserved sanction behind the conduct remedies in section 27, and its placement in government hands reflects its character as an act of economic policy rather than of adjudication.

Q: What factors decide whether a combination harms competition?

Section 20(4) lists the factors: the actual and potential level of competition through imports; the extent of barriers to entry; the level of combination in the market; the degree of countervailing power; whether the combination is likely significantly and sustainably to increase prices or profit margins; the extent to which effective competition is likely to remain; the availability of substitutes; the market shares of the parties, individually and combined; the likelihood of removing a vigorous and effective competitor; the nature and extent of innovation; the relative advantage by way of contribution to economic development; and whether the benefits of the combination outweigh its adverse impact, where an adverse impact is found.

Q: Which orders can the Commission pass besides a penalty?

Section 27 is primarily a remedial provision. The Commission may direct the parties to discontinue the offending agreement or the abuse of dominance and not to re-enter into it; direct the modification of an agreement found to contravene section 3; and pass such other orders, including directions as to costs, as it considers fit. Under section 31 it may approve a combination, approve it subject to modifications the parties accept, or direct that it shall not take effect. Under section 33 it may restrain a party by interim order during an inquiry where appreciable prejudice to consumers or the market is likely. Penalties accompany these orders; they do not replace them.

Q: Why did the Competition Act take effect in stages?

Because section 1(3) left the commencement of every provision to separate notification, and the government used that power across eight years. The institutional and definitional provisions began on 31 March and 19 June 2003 and the Commission was established on 14 October 2003, but the original design of the Commission was challenged in Brahm Dutt v. Union of India and recast by the Competition (Amendment) Act, 2007, which separated the regulator from a new appellate tribunal. Only with a reconstituted Commission in place did sections 3 and 4 begin on 20 May 2009, section 66 followed on 1 September 2009, and the combination provisions came last, on 1 June 2011.