Competitive Intelligence Vs Unlawful Coordination Boundary

1. Introduction

Competitive intelligence is the legitimate collection and analysis of information about competitors, customers, technologies, market trends, prices, capacity, products, regulation and commercial developments. It is an ordinary part of competitive strategy.

The legal boundary is crossed when competitors move from independent observation of the market to communication, receipt, signalling or use of commercially sensitive information in a manner that reduces strategic uncertainty and facilitates coordination.

The central competition-law principle is therefore:

A firm may generally learn about competitors; it must not replace independent competitive decision-making with coordination with competitors.

Under EU competition law, this boundary principally arises under Article 101 TFEU; in the UK, the corresponding domestic framework is principally Chapter I of the Competition Act 1998. The same economic concern appears in US antitrust law concerning information exchanges between competitors. The FTC notes that information concerning price, cost, output, customers and strategic plans is particularly sensitive, while historical, aggregated and sufficiently anonymised information generally presents less risk.

2. Meaning of Competitive Intelligence

Competitive intelligence involves obtaining information through legitimate means, such as:

• public financial statements;
• annual reports;
• regulatory filings;
• published price lists;
• public tenders and procurement databases;
• competitors' websites;
• product catalogues;
• patents;
• public speeches and presentations;
• customer reviews;
• industry publications;
• publicly available market statistics;
• independent market research;
• lawful benchmarking;
• reverse engineering where legally permissible;
• information voluntarily supplied by customers or suppliers.

The important characteristic is independence.

For example, if Company A examines Company B's publicly advertised prices and independently decides its own pricing strategy, this is normally ordinary competitive intelligence.

By contrast, if Company A contacts Company B and obtains its future price increase, intended discount, customer allocation strategy or planned output, and then adjusts its own conduct accordingly, the competition-law risk becomes substantially greater.

3. The Fundamental Boundary

A useful conceptual formula is:

Lawful competitive intelligence

Observe → analyse → independently decide → compete

Potentially unlawful coordination

Communicate → disclose sensitive information → obtain strategic certainty → align conduct

The distinction is not simply whether information was exchanged.

The legal analysis considers:

1. What information was obtained?
2. Who provided it?
3. Was it public or private?
4. Was it historical, current or future?
5. Was it aggregated or firm-specific?
6. Was the recipient a competitor?
7. Was there a legitimate business justification?
8. Could the information reduce strategic uncertainty?
9. Did the recipient modify its conduct?
10. Was there an accompanying agreement, understanding or concerted practice?

4. Information Categories

The FTC similarly distinguishes information that is historical, aggregated and anonymised from current or future firm-specific information.

5. Why Information Exchange Can Become Coordination

Competition normally involves uncertainty.

Company A does not know with certainty:

• what price B will charge;
• whether B will expand capacity;
• whether B will discount;
• which customers B will target;
• whether B will enter a new geographic market.

That uncertainty creates competitive pressure.

An exchange of strategic information can eliminate that uncertainty.

For example:

A tells B: "We intend to increase our price by 10% next month."

B now knows that aggressive price competition may be less necessary.

If B communicates its own intentions in return, both firms may become capable of coordinating their behaviour without expressly saying:

"Let us fix prices."

This is one reason competition law recognises concerted practices, not merely formal written agreements.

6. Case Law

Case 1 — Suiker Unie and Others v Commission, Joined Cases 40–48, 50, 54–56, 111, 113 & 114/73

The Court of Justice explained the fundamental distinction between independent adaptation to market conditions and coordination.

A competitor may observe another firm's conduct and independently react to it. That is not automatically unlawful.

The problem arises where competitors communicate their intended conduct to one another in a manner that removes the uncertainty inherent in competition.

The case is therefore important for the independent-conduct principle.

Principle

Competition law does not require firms to ignore information available in the market.

It requires them to determine their market conduct independently.

This remains one of the foundational principles for distinguishing intelligence gathering from coordination.

Case 2 — Ahlström Osakeyhtiö and Others v Commission (Wood Pulp), Joined Cases C-89/85 etc.

The Wood Pulp litigation is particularly important because it demonstrates that parallel behaviour and market transparency are not automatically proof of unlawful coordination.

The Commission had considered price announcements and other information concerning pulp prices. The Court distinguished between conduct that could be explained by normal market conditions and genuine artificial coordination.

At the same time, the judgment illustrates the danger of organised information exchange accompanying price coordination.

The Court's reasoning therefore prevents an important overreach:

Similar prices + public information ≠ automatically a cartel.

But a structured system through which competitors exchange information about future conduct can provide evidence of concertation.

Importance

This case is particularly relevant to modern algorithmic markets because competitors can observe the same publicly available signals and independently arrive at similar decisions.

Similarity alone should therefore not be confused with communication or coordination.

Case 3 — John Deere Ltd v Commission, Case C-7/95 P

This is one of the leading information-exchange cases.

The Court considered an information-exchange system in the agricultural machinery sector.

The system made commercially significant information available among competitors and increased market transparency concerning competitors' activities.

The Court accepted that information exchange can restrict competition where it reduces the normal uncertainty surrounding competitors' behaviour.

Principle

The legality of information exchange cannot be determined merely by asking:

"Is information being exchanged?"

The relevant question is whether the exchange changes the competitive environment by reducing strategic uncertainty between competitors.

Application to competitive intelligence

A company independently researching John Deere's publicly available information would be fundamentally different from competitors establishing a system through which strategically valuable information is systematically exchanged.

7. Case 4 — T-Mobile Netherlands BV and Others v Raad van bestuur van de Nederlandse Mededingingsautoriteit, Case C-8/08

This case demonstrates how little formal coordination may be required.

The Court considered whether a single meeting could constitute a concerted practice.

It held, in substance, that a concerted practice can arise without requiring a long-running series of meetings where the relevant conduct has an anticompetitive object.

Importance for competitive intelligence

A company cannot necessarily defend an information exchange by saying:

"It happened only once."

Nor is a formal written agreement necessarily required.

A single meeting, call, conference discussion or electronic communication can become legally significant depending on:

• what was discussed;
• the competitive context;
• the nature of the information;
• the parties' subsequent conduct.

This is particularly relevant to industry conferences and trade associations.

8. Case 5 — Asnef-Equifax v Asociación de Usuarios de Servicios Bancarios, Case C-238/05

This case demonstrates the opposite side of the boundary: information sharing can be legitimate and potentially procompetitive.

The case concerned a credit-information register.

The Court identified circumstances in which an information-sharing system would not, in principle, restrict competition, including safeguards concerning:

• market concentration;
• identification of individual lenders;
• non-discriminatory access;
• the structure and operation of the information system.

Principle

Competition law does not prohibit every information-sharing mechanism.

Information can improve competition by:

• reducing information asymmetry;
• improving credit assessment;
• reducing risk;
• facilitating market entry;
• improving allocation of resources.

Thus, the question is not:

"Was information shared?"

but:

"What competitive function does the information-sharing system perform, and does it facilitate coordination?"

9. Case 6 — Eturas UAB and Others v Lietuvos Respublikos konkurencijos taryba, Case C-74/14

This is especially significant for digital platforms and algorithmic coordination.

Travel agencies used a common electronic booking system. The system administrator sent a message concerning restrictions on discounts, and the system subsequently implemented a maximum discount.

The Court considered whether this could constitute a concerted practice.

Principle

A competitor does not escape competition law merely because coordination occurs through:

• software;
• a platform;
• an automated message;
• a common IT system;
• an intermediary.

The important issue is whether competitors became aware of information or instructions capable of influencing their market conduct and whether the circumstances support the existence of concertation.

Modern significance

This case is particularly relevant to:

• pricing algorithms;
• common optimisation platforms;
• industry software;
• marketplace algorithms;
• shared procurement platforms;
• automated discount systems;
• AI-enabled pricing tools.

The technology is not a legal safe harbour.

10. Case 7 — Commission v Anic Partecipazioni, Case C-49/92 P

Anic is important for understanding the broader concept of agreement/concerted practice.

The Court treated agreements and concerted practices as potentially forming part of a broader continuing infringement where firms participate in coordinated conduct.

Principle

A company cannot necessarily isolate each individual communication and argue:

"This particular communication was harmless."

Competition authorities can examine the overall pattern of interactions.

Thus, competitive intelligence activities should be assessed cumulatively where there are:

• repeated competitor contacts;
• repeated sensitive information exchanges;
• reciprocal disclosures;
• aligned commercial responses;
• common monitoring arrangements.

11. UK Example — UK Government Bond Information Exchange

A particularly contemporary UK example is the CMA's investigation into information exchanges concerning UK government bonds.

In February 2025, the CMA issued five infringement decisions concerning Citi, Deutsche Bank, HSBC, Morgan Stanley and Royal Bank of Canada. The conduct involved one-to-one Bloomberg chatroom exchanges concerning competitively sensitive information about trading in gilts and gilt asset swaps. Four banks agreed to fines totalling £104.46 million, while Deutsche Bank received immunity after reporting its participation under the CMA's leniency policy.

The significance is that the communications were not merely broad industry intelligence.

They involved bilateral exchanges between competitors concerning commercially sensitive pricing and trading information.

This provides a useful practical illustration of the boundary under the UK Competition Act 1998.

12. Public Information Versus Private Information

One of the clearest distinctions is:

Public information

Example:

"Company X publicly announced that its product will cost £100."

Company Y can ordinarily observe this and decide how to respond.

Private information

Example:

"Company X privately tells Company Y that it intends to increase its price to £100 next month."

That information is much more problematic.

Particularly dangerous

Information concerning:

• future prices;
• future discounts;
• bids;
• output;
• capacity;
• customer allocation;
• strategic expansion;
• intended market exit;
• production restrictions;
• future commercial strategy.

13. Direct Intelligence Versus Indirect Intelligence

Direct intelligence

A firm asks its competitor:

"What price are you going to charge next quarter?"

This creates obvious risk.

Indirect intelligence

A company obtains competitor information through:

• trade associations;
• consultants;
• common software;
• suppliers;
• customers;
• industry analysts;
• data brokers.

Indirect acquisition does not automatically make the conduct lawful.

The authority can examine whether the intermediary was effectively being used as a communication channel between competitors.

14. The "Third-Party Shield" Is Not Absolute

Using a third party does not automatically solve the competition problem.

For example:

Lower risk

An independent research firm collects historical prices from 50 firms and publishes an anonymised market index.

Higher risk

A consultant collects each competitor's confidential future pricing plans and distributes sufficiently detailed information back to the competitors.

The second arrangement may perform essentially the same economic function as direct communication.

15. Aggregation and Anonymisation

Aggregation can materially reduce competition concerns.

Consider:

High risk:

Company A — £101
Company B — £104
Company C — £107

versus:

Lower risk:

Average industry price — £104.

But aggregation must be genuine.

If the market contains only three firms and the data can easily be reverse-engineered, the apparent anonymity may be superficial.

The FTC's information-exchange guidance similarly identifies aggregation, historical data and independent third-party administration as factors reducing risk.

16. Historical Versus Future Information

This is one of the most important distinctions.

Historical

"Our average production cost two years ago was £70."

Generally less competitively sensitive.

Current

"Our current average production cost is £70."

More sensitive.

Future

"We will reduce our price to £75 next month."

Potentially highly sensitive.

Future information is particularly problematic because it can communicate intentions rather than merely describe past market conditions.

17. Competitive Intelligence in Digital Markets

The boundary becomes more difficult where firms use:

• web scraping;
• pricing APIs;
• AI monitoring;
• algorithmic benchmarking;
• automated competitor tracking;
• machine-readable price feeds;
• shared data platforms;
• common optimisation software.

Suppose an AI system automatically monitors public prices of 10 competitors.

That can constitute legitimate competitive intelligence.

But suppose the same system:

1. communicates with competitors' systems;
2. receives non-public future pricing information;
3. recommends coordinated price movements;
4. monitors deviations;
5. automatically punishes deviations.

The legal character of the system can change dramatically.

The important question is not whether AI made the decision.

It is whether the system facilitated conduct that would be unlawful if performed by humans.

Eturas is particularly relevant because the Court examined coordination occurring through a common electronic booking infrastructure.

18. Algorithms and the "Communication Through Code" Problem

Modern competitive intelligence can become especially problematic when algorithms communicate through:

• APIs;
• automated bidding systems;
• shared data feeds;
• common pricing engines;
• machine-readable signals;
• dynamic pricing platforms.

Consider:

Firm A's algorithm:
"We will increase prices by 8%."

Firm B's algorithm:
"We observe the signal and increase by 8%."

If the system is deliberately designed to transmit strategic intentions between competitors, describing the process as "automated intelligence" does not eliminate competition-law concerns.

The key issue remains coordination and independence.

19. Trade Associations

Trade associations are a frequent boundary area.

Potentially legitimate

• technical standards;
• safety information;
• regulatory compliance;
• aggregated historical statistics;
• general industry trends;
• cybersecurity information;
• non-commercial technical cooperation.

Higher risk

• future prices;
• current individual prices;
• customer allocation;
• production plans;
• bidding intentions;
• capacity restrictions;
• individual discount policies.

The FTC expressly recognises that competitor collaboration through trade associations can be legitimate while warning against information exchanges that undermine independent competition.

20. Procurement and Bid Intelligence

The boundary is especially strict in procurement.

Lawful

A bidder researches:

• previous publicly available tender prices;
• procurement authority specifications;
• published award information;
• competitors' publicly announced capabilities.

Potentially unlawful

Competitors communicate:

• "You bid £10 million and I will bid £11 million."
• "You take Region A and I will take Region B."
• "I will not compete for this contract if you do not compete for the next one."

This moves from intelligence into bid coordination or market allocation.

The information itself can become evidence of a broader agreement.

21. Competitive Intelligence and Market Transparency

An important distinction is:

Natural transparency

Everyone can observe the same public information.

Artificial transparency

Competitors establish a mechanism that gives them unusually precise information about one another's confidential strategies.

Wood Pulp illustrates why ordinary market transparency should not automatically be treated as collusion, while John Deere demonstrates that structured information systems can nevertheless create competition concerns.

22. A Practical Legal Test

A useful six-stage test is:

Stage 1 — Source

Where did the information come from?

Public source → lower risk.

Competitor/private source → higher risk.

Stage 2 — Sensitivity

What does it reveal?

General market trend → lower risk.

Future price/bid/customer strategy → high risk.

Stage 3 — Specificity

Can a particular competitor be identified?

Aggregated → generally safer.

Firm-specific → more dangerous.

Stage 4 — Timing

Is it historical or prospective?

Historical → generally lower risk.

Future intentions → substantially higher risk.

Stage 5 — Purpose and effect

Why was the information exchanged?

Regulatory compliance or legitimate benchmarking → potentially defensible.

Reducing uncertainty between competitors → serious concern.

Stage 6 — Conduct

Did competitors subsequently align their behaviour?

Independent responses → potentially lawful.

Coordinated responses → significant risk.

23. A Simple Boundary Matrix

24. Compliance Safeguards

Businesses conducting competitive intelligence should consider:

1. Public-source preference

Prefer:

• public filings;
• public databases;
• published prices;
• public tender records;
• public technical information.

2. Information classification

Classify information as:

• public;
• historical;
• aggregated;
• confidential;
• competitively sensitive;
• highly sensitive/future-oriented.

3. Competitor-contact protocols

Employees should know what cannot be discussed with competitors.

4. Trade-association controls

Use agendas and compliance protocols for competitor meetings.

5. Third-party safeguards

Where market research is commissioned, ensure that the methodology does not unnecessarily disclose firm-specific confidential information to competitors.

6. Data minimisation

Collect only information genuinely necessary for the legitimate commercial purpose.

7. Audit trails

Maintain records explaining:

• source;
• date;
• purpose;
• methodology;
• recipients;
• aggregation method.

8. Escalation procedure

Employees should immediately escalate unexpected competitor disclosures of:

• future pricing;
• bids;
• customer allocation;
• production plans;
• strategic intentions.

25. A Useful "Red Flag" Rule

An employee should become particularly cautious when a competitor says:

"I'll tell you what we're going to do if you tell me what you're going to do."

That is fundamentally different from:

"Our published price is £100."

The first involves reciprocal strategic disclosure.

The second merely communicates public information.

26. Competitive Intelligence vs Unlawful Coordination — Core Distinction

27. Overall Legal Principle

The boundary can be reduced to one proposition:

Competition law protects the right of firms to know the market, but not the right of competitors to know each other's confidential future strategies through coordination.

The leading cases collectively establish several important propositions:

1. Independent adaptation is not itself collusion — Suiker Unie.
2. Parallel behaviour does not automatically prove coordination — Wood Pulp.
3. Structured information systems can reduce competitive uncertainty — John Deere.
4. A single contact can potentially constitute concerted practice — T-Mobile Netherlands.
5. Information-sharing can be legitimate and procompetitive where properly structured — Asnef-Equifax.
6. Digital systems can facilitate concerted practices just as human communications can — Eturas.
7. Repeated communications may be assessed as part of an overall infringement — Anic.

The modern challenge is therefore not to prohibit competitive intelligence, but to maintain a legally meaningful distinction between learning from the market and learning directly from competitors in a way that substitutes coordination for independent competition. The UK's 2025 government-bond information-exchange decisions demonstrate that this distinction remains practically significant under Chapter I of the Competition Act 1998.