What Counts as a Cartel Under Competition Law?

Cartels are competition law’s core prohibition — the offence every regime agrees on and punishes hardest. Yet real cartel cases rarely look like smoke-filled rooms: they are built from trade-association minutes, pricing signals, WhatsApp groups and “market stabilisation” understandings that participants often did not think of as illegal. This guide defines what counts as a cartel, where the line runs between collusion and lawful cooperation, and what the landmark cases teach — opening the cartel-enforcement pillar of our Global Competition & Antitrust hub.

What exactly makes an arrangement a cartel?

Three elements: competitors (actual or potential), coordination (agreement or concerted practice — form is irrelevant), and an object of restricting competition — price, output, customers, territories or tenders. Once those align, the infringement is complete in “by object” regimes: no market power threshold, no effects analysis, no efficiency defence worth the name.

The breadth surprises businesspeople. “Agreement” includes gentlemen’s understandings, nods at trade dinners and acquiescence in a WhatsApp group one never answered. “Price fixing” includes coordinating surcharges, discounts, credit terms, price announcements timing — any element of the price. “Market sharing” includes respecting each other’s “home” customers without a word being exchanged about prices. And duration aggregates: EU law treats years of varied contacts as a “single and continuous infringement”, so attending three meetings across a decade can mean liability for the whole decade’s cartel.

How do real cartels actually operate?

Modern cases show recurring machinery: regular “technical committee” meetings hosted inside legitimate trade associations; rotating secretariats keeping quota spreadsheets; code names and private email accounts; compensation schemes settling deviations from agreed shares; and — increasingly — algorithmic implementation, where agreed pricing rules are encoded into repricing software.

The vitamins cartel of the 1990s ran like a corporation, with annual “budget” meetings allocating world market shares. The trucks cartel coordinated gross price lists and the timing of emissions-technology cost pass-through for fourteen years, partly through subsidiary-level employees at industry gatherings. The air-cargo and forex cartels lived in chat rooms (“The Cartel” was a real Bloomberg chat name). The operational lesson for compliance teams: cartels are discovered in calendars, expense reports, chat logs and spreadsheets — which is exactly where dawn raids look first.

Where is the line between a cartel and lawful cooperation?

Competition law does not condemn cooperation as such — joint R&D, standardisation, purchasing consortia, genuine joint ventures and sustainability initiatives can all be lawful, and the EU’s 2023 horizontal guidelines map the safe zones. The line runs at coordination on the parameters of competition: prices, outputs, customers, bids, strategy.

The gray zones are where liability is actually made. Information exchange among competitors — even without any agreement to act on it — can itself infringe when it removes strategic uncertainty about future prices or volumes; we treat this fully in the information-exchange guide. Benchmarking clubs, trade-association statistics and “market colour” conversations sit on the same slope. Joint negotiations are lawful inside genuine purchasing groups but become buyer cartels outside them — and employer coordination on wages or hiring (“no-poach”) is now prosecuted as labour-market cartel conduct in the US, EU member states and Türkiye alike, a frontier many HR departments still have not internalised.

What are the landmark cartel cases every executive should know?

Beyond trucks: the vitamins cartel (global market allocation; record fines of their era and the template for leniency-driven discovery); air cargo (fuel-surcharge coordination among two dozen airlines, litigated worldwide for fifteen years); Libor and forex (benchmark and trading coordination that converted competition law into a financial-markets enforcement tool, with billions in fines across the EU, US and UK); and the power cables and auto-parts networks, where Japanese, Korean and European suppliers rigged OEM tenders — the auto-parts programme remains the US DOJ’s largest criminal campaign, with dozens of executives imprisoned.

Türkiye’s docket makes the same points locally: the Rekabet Kurulu’s banking cartel decision (twelve banks fined over deposit and loan rate coordination), cement cases across regions, and labour-market investigations against dozens of employers for wage and no-poach coordination. The pattern across jurisdictions is uniform: mature industries, homogeneous products, trade-association infrastructure and procurement concentration are the recurring risk factors.

Who is liable — companies, parents, individuals?

All three, in widening circles. The infringing entity is liable; its parent company is presumed jointly liable in the EU for wholly owned subsidiaries (the “single economic unit” doctrine), which is how group turnover — not subsidiary turnover — sets the 10% fine cap. Successor liability follows the business through restructurings and acquisitions, making cartel exposure a standard M&A diligence item: buy a company, inherit its cartel.

Individuals face prison in the US (Sherman Act felonies), the UK (cartel offence), Canada, Australia, Japan and a growing list; the EU fines only companies, but member states add individual sanctions, and Türkiye can impose personal administrative fines on employees and managers who played a determining role. Directors face disqualification in the UK; and everywhere, careers end quietly — the executive who signed the quota spreadsheet is unemployable in the industry once the decision publishes names.

How are cartels discovered — and why do they keep collapsing?

Mostly from inside. Leniency — full immunity for the first participant to confess — remains the largest single source of cases, having converted the cartel’s own instability into the enforcer’s weapon; the mechanics and the race dynamics are detailed in our leniency guide. The rest come from screening (bid data analytics flagging suspicious tender patterns), complainants and customers, whistle-blower reward programmes, dawn raids in adjacent cases, and — increasingly — cooperation between authorities exchanging evidence across borders.

The game theory does the enforcer’s work: every cartel member holds a get-out-of-jail card that appreciates as detection risk rises, so any hint of investigation triggers a race to confess. Add damages exposure, and stable equilibrium becomes nearly impossible: the trucks cartel ended not because it stopped being profitable, but because MAN ran to Brussels first.

What should you do if you discover cartel contact inside your company?

Move on four tracks within days, all under legal privilege. Preserve: suspend deletion policies for relevant custodians immediately — destruction after discovery is obstruction, separately punishable and fatal to leniency. Investigate: external counsel interviews the individuals and reviews the documents to establish scope — products, geographies, duration, participants — because every later decision depends on knowing what you actually have.

Stop: end the conduct without signalling to co-cartelists (an abrupt public exit can tip the ring into evidence-destruction mode; counsel choreographs the exit). Decide: board-level leniency decision on the full multi-jurisdiction ledger, fast — the option is wasting. In parallel, quietly assess the employees involved: discipline too early can create hostile witnesses and independent leniency applicants, too late looks like ratification. This sequence — preserve, investigate, stop, decide — is rehearsable in advance, and companies that have rehearsed it consistently reach the marker queue first; the mechanics continue in our leniency guide.

How does cartel risk enter through M&A and joint ventures?

Two doors. Acquisitions import liability: successor doctrine attaches the target’s cartel history to the buyer, and post-closing discovery of a cartel file is a classic integration crisis — price the risk through diligence interviews, document sampling in high-risk sectors, and indemnities with survival periods matched to limitation rules, as our antitrust diligence guide details.

Joint ventures create standing contact: JV boards seat competitors together for years, and information legitimately needed to govern the venture (its prices, capacities, plans) borders information illegal to share about the parents’ competing businesses. The controls are structural — clean governance charters, information barriers between JV representatives and parent commercial teams, agenda and minute discipline — and they decay without maintenance. A material share of information-exchange cases begin life as sloppy JV governance.

Frequently Asked Questions