A textbook cartel, caught the right way
On July 13, 2026, the Competition Commission of India (CCI) closed the book on two related cartel investigations that had run since 2020. In Suo Motu Case No. 07 of 2020, it found HP India Sales Pvt. Ltd. had coordinated with authorised resellers to rig bids for personal computers and workstations on the Government e-Marketplace (GeM), the Commerce Ministry's centralised public-procurement portal. In Case No. 08 of 2020, it found a parallel "hub-and-spoke" arrangement covering HP printer toner, ink cartridges, and other consumables. Combined, HP India was fined ₹138.85 crore — ₹126.87 crore for the personal-systems cartel and ₹11.98 crore for the supplies cartel — while 21 named resellers were penalised a further ₹3.52 crore, taking the total past ₹142 crore, or roughly $16 million.
The conduct itself is not a close call. Section 3(3)(d) of the Competition Act, 2002 presumes that agreements resulting in bid rigging or collusive bidding have an appreciable adverse effect on competition — one of the few presumptions in Indian antitrust law that shifts the burden onto the accused. The CCI's investigation, drawing on WhatsApp threads, internal emails, and a recorded reseller meeting, found HP India dictating bid prices to resellers and selectively issuing or withholding Manufacturer Authorisation Forms — the credential a reseller needs to legally bid — to decide in advance who would win a given GeM tender and who would submit a "cover bid" to preserve the appearance of competition. Across the two cases, that pattern touched 36 separate tenders between 2017 and 2020.
Steelmanning the aggressive theory
The most defensible criticism of this order is not that CCI over-reached on the facts, but that its "hub-and-spoke" theory — treating a manufacturer as the orchestrating hub of a cartel among its own downstream resellers, who nominally compete with each other but are all bound to the same vertical supply relationship — is a genuinely expansive reading of Section 3. Manufacturer Authorisation Forms, authorised-reseller lists, and price guidance to channel partners are completely ordinary tools of vertical distribution management, used by essentially every OEM that sells through resellers rather than direct sales. A regulator too eager to relabel ordinary vertical coordination as horizontal collusion risks chilling legitimate channel management — telling a manufacturer it can never help decide which of its own resellers pursues which deal without inviting a bid-rigging inquiry.
That is a fair concern in the abstract. It does not survive contact with these facts. The CCI did not penalise HP India for having an authorised-reseller program; it penalised specific, documented coordination — price dictation, pre-allocated winners, cover bids submitted purely to simulate competition — captured in the participants' own communications. GeM exists precisely so that government departments get the benefit of genuine price competition instead of negotiated single-vendor deals; rigging it is not a technical foul, it is the direct diversion of public money that competitive bidding is designed to prevent. A publication that argues for light-touch, evidence-based regulation should also argue that public procurement fraud is exactly the kind of harm competition law exists to catch — proportionate enforcement means matching the response to the harm, not refusing to enforce at all.
The part worth watching: leniency
The more interesting fight is over what the CCI did to HP India's own leniency application. HP India was the first party to approach the Commission under Section 46's lesser-penalty programme, disclosing the cartel's existence in both cases. Ordinarily, a first-in discloser under India's leniency regime is positioned for the largest available penalty reduction, up to full immunity — the same incentive structure that anchors leniency programmes from Washington to Brussels to Delhi, because cartels are hard to detect from the outside and easy to detect from the inside. The CCI instead ruled that HP India could not receive full immunity because it was not merely a participant reporting on peers but the cartel's architect and principal beneficiary — the entity that designed the authorisation-and-allocation scheme in the first place. Cooperation, the Commission held, mitigates the penalty; it does not erase liability for having built the cartel.
As a matter of case-specific equity, that is a defensible line — a company should not get to engineer a cartel for years and then buy full absolution merely by confessing once an investigation becomes likely. But leniency programmes work only because they are predictable: a company deciding whether to self-report is running a probability calculation, and if the payoff for coming forward first can still be substantially discounted based on an after-the-fact judgment about how central its role was, the expected value of confessing early drops. Cartel orchestrators — the parties with the most complete evidence and the most to gain from defecting early — are precisely the ones a leniency regime most needs walking through the door. A ruling that makes them less certain of the reward for doing so may improve the optics of this one case while making the next cartel marginally harder to detect.
What should follow
CCI's order is right on the merits and proportionate to the conduct: real evidence, a real harm to public procurement, and a penalty that runs into the tens of millions of dollars rather than an existential fine. The compliance-programme mandate and cease-and-desist order are the correct remedy for a channel-management practice that crossed into collusion. Where the Commission should move carefully is in clarifying, in guidance or in the next leniency-adjacent order, exactly how "architect versus participant" will be assessed going forward — otherwise the lesson every general counsel in India takes from this case will be that self-reporting is a bet on the CCI's later characterisation of your role, not a predictable trade.