Sit at the back of a competition compliance session at an Indian company and watch the room. The slides are careful and technically correct. They explain cartels, they define dominance, and then they illustrate both with a European cement case and an American price fixing prosecution from the nineteen nineties. The regional sales managers in the room are polite and entirely unmoved, because nothing on the screen resembles anything they do. They will go back to a distributor meeting that afternoon, discuss the discount everybody is offering in that territory, and never once connect the two.
That failure is not a failure of legal accuracy. The doctrine on those slides is correct. It is a failure of recognition, and recognition is the only thing compliance training is actually for. A person complies when they can see themselves in the example. The argument of this article is simple: competition compliance built on Indian enforcement outcomes lands, and competition compliance built on imported antitrust theory does not, because the Indian orders describe Indian trade associations, Indian distributor networks, Indian tender processes and Indian internal emails, and the person in the room recognises all of them.
This is written for in-house counsel, compliance heads and company secretaries who have to design a programme that changes behaviour rather than one that survives an audit. It covers how to read an order for compliance value rather than for legal analysis, the evidential patterns that recur, where the substantive risk lines actually sit under the Competition Act, 2002, the mechanisms introduced by the 2023 amendment, dawn raid readiness, and how to keep a standing watch on outcomes in your own sector.
The point of reading orders
A judgment read by a lawyer is read for the reasoning. An order read for compliance value is read for something else entirely: what did these people actually do, what did they write down, and how did anyone find out. The legal test is the least transferable part of the document. The conduct and the evidence are the parts your sales team needs to see.
Why Imported Training Does Not Change Behaviour
Three specific things go wrong when a compliance programme is built on foreign material.
The first is that the commercial context does not map. Distribution in India runs through layers of stockists, distributors and dealers, with territory practices, credit arrangements and rebate structures that have no clean equivalent in the examples the slides are drawn from. A rule illustrated in a context nobody recognises becomes an abstraction, and abstractions are complied with abstractly.
The second is that the institutional furniture is different. Indian commercial life is organised, to a degree that surprises foreign advisers, around industry and trade associations. They set standards, they publish data, they run committees, and they meet regularly. That is not inherently unlawful and much of it is valuable. But it means the single most common physical setting for the conduct that produces enforcement in India is a room that most employees regard as entirely respectable and attend without a second thought.
The third is procedural. What actually happens when the Commission takes an interest, how an investigation proceeds, what appears on the record, where an appeal goes and what the exposure looks like: none of that is transferable. Appeals from the Commission lie to the National Company Law Appellate Tribunal, and from there to the Supreme Court. An employee who has been told about a foreign appellate structure has been told something irrelevant to their own risk.
People do not comply with rules they understand. They comply with rules they can picture themselves breaking.
How to Read an Order for Compliance Value
Reading an order as a lawyer and reading it as a compliance designer are different exercises using the same document. The lawyer wants the test applied, the treatment of precedent, the analytical structure. The compliance designer wants three narrower things, and can safely skim most of the rest.
The conduct, described concretely
Not the legal characterisation but the behaviour: who met whom, how often, what was agreed, how the agreement was policed, and what commercial outcome it was meant to produce. This is the part you can put in front of a sales team, because it is written in the language of their working week rather than the language of the statute.
The evidence relied on
What did the decision maker actually have. Attendance records, circulars, price lists, messaging group content, internal email, notes of meetings, parallel movements in pricing. The evidence section is the most useful part of any order for training purposes, because it tells your people exactly what leaves a trace.
The internal documents that sank the party
Almost every enforcement outcome features a document created by an employee who did not think they were creating evidence. A note recording what was agreed at a meeting. An email explaining the reason for a price move. A message asking a competitor to confirm. Find these and you have your training material.
There is a fourth thing worth extracting, and it is the one compliance teams usually skip: what the party said in its defence and why it did not work. Employees find the defence more persuasive than the allegation, because the defence is usually the exact rationalisation they would have reached for. Showing that the rationalisation was considered and rejected does more work than showing that the conduct was unlawful.
One discipline is essential. An order is a decision on the facts before that decision maker at that time, and it may have been appealed. Before you build training on any particular outcome, check whether it was carried to the appellate tribunal and what happened there, and check whether the position it rests on has moved. Training built on an order that was subsequently set aside is worse than no training, because someone in the room will know.
The Evidential Pattern That Repeats
Read enough Indian enforcement material and the same shapes recur. They are worth stating as a pattern, because a pattern is teachable in a way a case list is not.
- The trade association meeting. A legitimate forum, with an agenda, minutes and an attendance sheet, at which the discussion moves from industry data to something narrower. The attendance sheet is a gift to an investigator: it establishes who was in the room, on what date, without anyone having to prove it.
- The circular or the shared price communication. A document issued to members recording what the industry is doing about pricing, supply or terms. It exists because somebody thought it was helpful administration.
- The messaging group. Informal, fast, and preserved on every participant's device. People write things in a group chat that they would never put in an email, on the assumption that it is conversation rather than record. It is record.
- The internal email explaining why. The most damaging document is frequently not the agreement but the internal note explaining the reason for a decision, written by a manager reporting up the line and being candid about what was discussed with whom.
- The bidding pattern. In procurement, the arithmetic itself carries information: sequences of bids, rotation of winners, unsuccessful bids that are strikingly close or strikingly implausible, and identical documentation defects across supposedly independent bidders.
- The distributor communication. In vertical arrangements, the paper trail sits in the instructions actually sent down the chain, about where to sell, at what price, and what happens if the instruction is ignored.
The sentence to put on the slide
Almost nobody in an Indian enforcement matter set out to commit a competition offence. They attended a meeting they attend every quarter, they were candid in an internal email because candour is what their manager expects, and they wrote in a group chat because that is how the industry talks. The conduct that creates exposure is overwhelmingly ordinary, and it is documented by the people doing it. That is the entire compliance message, and it is far more persuasive than any statement of the legal test.
Where the Risk Lines Actually Sit
Horizontal conduct and the presumption
Section 3 of the Competition Act, 2002 deals with anti-competitive agreements. Within it, the Act treats agreements between competitors differently from agreements between parties at different levels of the supply chain. Certain categories of horizontal agreement, of the kinds that determine prices, limit or control production or supply, share markets or sources of supply, or rig bids, carry a statutory presumption of an appreciable adverse effect on competition. The practical consequence for a compliance programme is enormous and is rarely explained in the room: for this category, the argument that the arrangement was harmless, or was a response to genuine market conditions, starts from a much worse position than employees assume. Most people believe intent and effect are open questions in every case. For this category, that belief is dangerous.
Vertical conduct
Vertical arrangements between a supplier and its distributors, including tie-in arrangements, exclusive supply and exclusive distribution, refusal to deal and resale price maintenance, are assessed differently and turn on effects. For an Indian company this is where a great deal of everyday commercial practice actually lives, and it is the area where a compliance team most needs to give usable guidance rather than a prohibition. Blanket rules fail here because the business will simply ignore a rule that makes ordinary distribution impossible. Guidance keyed to what the enforcement record has actually treated as problematic is more likely to be followed.
Abuse of dominance, and why market definition decides it
Section 4 addresses abuse of a dominant position. The word that matters is dominant, and the reason abuse cases are won and lost on market definition is arithmetic. A wide relevant market makes almost nobody dominant. A narrow one can make a company dominant in a segment it never thought of as a market at all. This is why conduct that seems obviously acceptable to a business, because it is a small player in the industry as it understands the industry, can attract scrutiny once the relevant market is defined by product and by geography in a way the business never contemplated. For platform and digital businesses in particular, market definition is the whole battlefield, because the question of what the product actually is admits of several serious answers.
Section 27 sets out the orders and penalties the Commission may pass on finding a contravention, which include directions to discontinue the conduct and monetary penalties. The Competition (Amendment) Act, 2023 made changes to how penalties are computed. Because the 2023 amendment's provisions were brought into force in stages, the operative position on penalty computation, and on the other mechanisms discussed below, should be confirmed as it stands when you act rather than assumed from any summary, including this one.
Combinations and gun-jumping
Sections 5 and 6 govern combinations, meaning acquisitions, mergers and amalgamations that cross the thresholds the Act prescribes. Two compliance points matter more than the substantive assessment. The first is gun-jumping: consummating a notifiable transaction, or parts of it, before the required approval, and equally, taking steps that amount to exercising control ahead of approval. Deal teams under commercial pressure create this risk without legal input, through integration planning, early operational control, or exchanging competitively sensitive information during diligence. The second is disclosure: what is said in a notification, and what is omitted, carries its own consequences. The Competition (Amendment) Act, 2023 introduced a deal value threshold for combinations, which expands the range of transactions that can require notification beyond those caught by asset and turnover tests alone; again, confirm the operative position and the applicable criteria as they currently stand.
| Risk area | What the enforcement record tends to show | What the programme should actually do |
|---|---|---|
| Horizontal arrangements under Section 3, including price coordination, market sharing and bid rigging | Trade association settings, attendance records, circulars and messaging groups. Conduct documented by participants who believed the forum was legitimate. | Rules for association attendance: what may be discussed, what must be objected to, and the requirement to leave and report. Train on the meeting, not on the doctrine. |
| Bid rigging in public and private procurement | Bidding arithmetic that speaks for itself, near-identical documentation across supposedly independent bidders, and rotation patterns visible across tenders. | A tender protocol: no contact with competing bidders, independent preparation, a record of who prepared what, and a bid file that can be produced later. |
| Vertical arrangements under Section 3, including exclusivity, tie-ins and resale price maintenance | The instructions actually sent down the distribution chain, and the consequences applied when they were not followed. | Reviewed distribution templates plus guidance on what may be communicated to a distributor about pricing and territory, written for the person who sends the message. |
| Abuse of dominance under Section 4 | Cases turn on how the relevant market is defined by product and geography, which can be far narrower than the business's own view of its market. | Identify the segments where the company could plausibly be found dominant and apply heightened guidance there, rather than assuming Section 4 is somebody else's problem. |
| Combinations under Sections 5 and 6 | Gun-jumping through early integration or operational control, and information exchange during diligence, both created by deal teams without legal input. | A clean-team protocol for diligence, a hard rule that integration steps wait for approval, and legal sign-off built into the deal calendar rather than requested late. |
| Investigation exposure, including the Director General's search and seizure powers | Devices and internal documents are the evidence. Deletion and obstruction create a second and separate problem on top of the first. | A rehearsed dawn raid protocol with named responders, and an absolute rule against deleting anything, communicated before it is needed. |
Leniency, Settlement and Commitments: Why the Internal Calculus Changed
Section 46 provides for lesser penalty, which is the Indian leniency regime. Its structure rewards a participant who comes forward with disclosure that assists the Commission, with the benefit typically tied to the order in which parties approach and the value of what they bring. The detailed conditions and the applicable procedure are set out in the Act and in the regulations made under it, and they should be checked in their current form before anybody relies on them.
What matters here is the behavioural consequence, which compliance teams consistently under-explain. Leniency makes a cartel structurally unstable, because every participant knows that every other participant has an incentive to move first. That is a point of enormous force with an employee who knows something and has been sitting on it. The internal message is not the legal one. It is that the arrangement your colleague is comfortable with is one that another company's employee can end tomorrow, on terms that benefit them and not you, and that your own organisation's ability to act depends on hearing about it before that happens.
The internal reporting channel is the whole game
A leniency regime only helps an organisation that knows what its own employees have done. If the person who attended the meeting has no safe route to report it, the first the company hears of it will be from an investigator. The most valuable competition compliance investment is usually not more training, it is a reporting channel people believe is safe to use, together with a stated position on how the company treats someone who comes forward.
The Competition (Amendment) Act, 2023 introduced settlement and commitment mechanisms, which give parties routes to resolve certain proceedings other than by fighting them to a final order. Because the amendment's provisions were brought into force in stages and are supported by regulations, the availability of these routes, the categories of proceeding they apply to, and the procedure and timing for invoking them should be confirmed against the position in force at the relevant time. The compliance significance is directional rather than technical: the existence of resolution routes changes the calculus of how a company responds to an investigation, and a legal team that has not thought about that in advance will be thinking about it under time pressure.
Dawn Raid Readiness as a Practical Topic
The Director General has search and seizure powers exercisable in the course of an investigation, and the first hour of a search is disproportionately important. Almost everything that goes badly in that hour goes badly because nobody had thought about it before.
The remedy is a one page protocol that reception, facilities and the senior team have all seen: who is called immediately, who greets the officials, who accompanies them, who records what is examined and taken, the absolute rule that nothing is deleted or removed, and the instruction that employees are courteous and do not speculate. Rehearse it once. The rehearsal is what makes it a capability rather than a document.
Building a Programme Keyed to Your Own Sector
The final move is to stop treating competition compliance as a general subject. Enforcement is not evenly distributed across the economy, and the risk profile of a cement business, a pharmaceutical distributor, a shipping line, a platform and a component supplier are not the same. A programme built on your own sector's enforcement history is shorter, more specific and far more credible to the people it is aimed at.
Assemble your sector's enforcement record
Search for Commission and appellate outcomes involving your industry, your product category, and the associations your people belong to. You are building a reading list, not a memorandum. Include appellate outcomes, because an order that was modified or set aside on appeal teaches a different lesson from the one it appears to teach.
Extract the conduct and the evidence, and discard the rest
For each outcome, write two short paragraphs: what these people did, and how it was proved. Leave out the analytical reasoning. This is the corpus your training will be built on, and it should read like a description of a working week rather than like a case note.
Map the conduct onto your own processes
Go through the extracted conduct and ask, for each item, where in this company could that happen. Which committee, which distributor communication, which tender, which pricing meeting. The output is a list of specific internal touchpoints, and it is more valuable than any risk matrix.
Write rules for the touchpoints, not for the statute
Convert each touchpoint into a rule someone can follow without legal training. What you may and may not discuss at an association meeting, and what to do when the discussion turns. Who prepares a bid and who they may not speak to. What may be said to a distributor about price. The test of a good rule is that the person can apply it in the room.
Train by function, using your own sector's material
Sales, procurement, pricing, business development and the deal team have different exposures and should get different sessions. Use the extracted Indian examples. The moment of recognition, when someone realises the described meeting is a meeting they attend, is the entire return on the exercise.
Build the reporting channel and say what happens next
Give people a route to raise something, tell them who sees it, and state the company's position on how someone who reports is treated. Without this, the programme generates awareness and no information, which leaves the company aware of a risk it cannot see.
Keep a standing watch on new outcomes in your sector
Enforcement moves, and so does the applicable framework, particularly while the 2023 amendment's provisions and the regulations under them settle into practice. A programme refreshed once and left alone becomes stale in the specific way that matters most: it keeps teaching last year's risk lines.
Where CourtMesh Fits
Two parts of this exercise are research problems rather than compliance problems, and they are the parts that stall.
The first is assembling the corpus. Finding the outcomes that involve your sector, and then following each through to what happened on appeal, means searching across the tribunal and court layers rather than in one place. CourtMesh offers unified search across the Supreme Court, 25 High Courts, District Courts and Tribunals over a corpus of roughly 310 million cases drawn from official government portals, which is how you find the appellate history rather than stopping at the first document you retrieve. AI case analysis helps you get from a long order to the conduct and the evidence quickly, which is the extraction step that otherwise consumes a lawyer's week.
What we index, and what we do not
Being exact about this matters, because the two halves of this research live in different places. CourtMesh does not index the CCI's own orders. The CCI publishes them on its own site, and that site is where you read a Section 27 order, a Section 26(1) direction or a combination approval. What CourtMesh does index is the layer above: the Supreme Court and all 25 High Courts, which is where writ challenges to CCI process and appeals from the appellate tribunal are decided, plus the district judiciary and tribunals, all drawn from official government portals. So the appellate and constitutional history of a competition matter is searchable here in one query; the Commission's own file is not, and you should go to the CCI website for it. What the corpus covers is set out in full.
The second is keeping the watch current. A private organisation-scoped watchlist lets you monitor the parties that matter to your sector, including your own group entities, your significant competitors where they are already public respondents, and the counterparties whose conduct affects you, so that new matters surface as they appear in the source registries. Only your own team sees whom you are monitoring, which is not a trivial point when the list includes competitors. Counterparty litigation and insolvency screening covers the related question that arises whenever you take on a new distributor or supplier, which is what that party is already involved in.
The limits, stated plainly
A search result is not a compliance opinion. Court and tribunal publication in India is neither immediate nor complete, so finding nothing means nothing was found, not that nothing exists, and where the answer matters it should be verified against the official record, which prevails. CourtMesh does not give legal advice. Whether particular conduct contravenes Section 3 or Section 4, whether a transaction is notifiable under Sections 5 and 6, whether to approach the Commission under Section 46, and what the operative position is on the mechanisms introduced by the Competition (Amendment) Act, 2023: every one of those is a question for competition counsel on your specific facts and on the law as it stands when you act. This article is general commentary written to help you design a programme, not to tell you what the law requires of your company.
Teach the meeting, not the doctrine
Competition compliance fails in Indian companies for a reason that has nothing to do with legal accuracy: the examples come from somewhere else, and nobody in the room recognises themselves. The enforcement record here describes trade association meetings people actually attend, distributor instructions people actually send, tender processes people actually run, and internal emails written by managers who thought they were being helpful. Build the programme on that material, key it to your own sector, give people a reporting channel they believe in, and rehearse the first hour of a search before you need it. On the research side, read the CCI's own orders on the CCI's website, because we do not index them, and use CourtMesh for the layer we do: the Supreme Court, all 25 High Courts, the district judiciary and tribunals, where the writ challenges and the appellate history of a competition matter are decided, with AI case analysis to get from a long order to what it actually shows and a private watchlist to keep the answer current. The judgement about what any of it means for your company stays with your counsel, where it belongs.
Explore CourtMesh
