Competition Law and Real-Time Competition Monitoring Systems

1. Introduction

Real-time competition monitoring systems are technological systems that continuously collect, process and analyse market information—such as prices, discounts, inventory, rankings, bids, output, market shares, consumer demand and competitor behaviour—to identify possible changes in competitive conditions.

They can be used by:

• competition authorities;
• regulated industries;
• businesses for compliance;
• digital platforms;
• procurement agencies; and
• independent market-monitoring bodies.

The technology can improve enforcement because competition authorities can detect suspicious pricing or coordination much faster than through traditional periodic investigations. At the same time, continuous monitoring creates competition-law risks where the system enables competitors to observe one another's commercially sensitive information or automatically react to competitors' conduct.

The OECD has specifically identified automated monitoring systems, algorithmic pricing tools and data intermediaries as technologies capable of making information flows more continuous, granular and actionable.

2. Meaning of Real-Time Competition Monitoring

A real-time competition monitoring system may contain five principal components:

1. Data collection
   • prices;
   • discounts;
   • tenders;
   • quantities;
   • inventory;
   • search rankings;
   • consumer behaviour;
   • delivery charges.
2. Automated processing
   • comparison of competitors' prices;
   • market-share calculations;
   • detection of unusual patterns;
   • identification of sudden price movements.
3. Algorithmic analysis
   • anomaly detection;
   • price correlation;
   • cartel-risk scoring;
   • market-power indicators.
4. Alerts
   • notification of potentially suspicious conduct;
   • sudden parallel price increases;
   • bid rotation;
   • exclusionary conduct;
   • unusual market concentration.
5. Enforcement or compliance response
   • investigation;
   • request for information;
   • dawn raid;
   • internal compliance review;
   • corrective measures.

Thus, real-time monitoring does not itself constitute an antitrust violation. Its legality depends principally on what information is collected, who can access it, how it is processed, and whether it facilitates coordination or exclusion.

3. Why Competition Authorities Use Real-Time Monitoring

Traditional competition enforcement is often retrospective. An authority may discover a cartel months or years after the conduct began.

Real-time monitoring potentially allows authorities to detect:

• coordinated price movements;
• suspicious tender patterns;
• market-sharing arrangements;
• exclusionary discounts;
• discriminatory access;
• algorithmic price matching;
• sudden foreclosure of competitors;
• excessive concentration;
• manipulation of online rankings;
• discriminatory platform treatment.

The CCI has itself highlighted algorithmic collusion and real-time/dynamic pricing as emerging issues in digital markets.

4. Legal Framework

A. Anti-competitive agreements

Real-time monitoring can create problems under provisions dealing with:

• price fixing;
• output restrictions;
• market allocation;
• bid rigging;
• exchange of competitively sensitive information;
• concerted practices.

The important distinction is between monitoring for enforcement and monitoring by competitors.

For example:

An antitrust authority collecting publicly available prices to identify cartel behaviour is fundamentally different from competing retailers subscribing to a common system that continuously displays each other's future pricing intentions.

The latter can create a hub-and-spoke or information-exchange concern.

The OECD's 2025 work identifies shared algorithms, common pricing systems and commercially sensitive information as important potential mechanisms for algorithmic coordination.

5. Information Exchange Through Monitoring Systems

Competition law traditionally distinguishes between:

Legitimate information

Examples include:

• historical public prices;
• publicly available market statistics;
• aggregated industry data;
• government statistics.

Potentially problematic information

Examples include:

• future prices;
• individual competitor discounts;
• production plans;
• customer-specific information;
• strategic inventory information;
• confidential tender information;
• planned capacity;
• commercially sensitive algorithms.

A monitoring system can make information significantly more competitively sensitive because it may transform information from occasional and difficult-to-use data into continuous and immediately actionable intelligence.

6. Algorithmic Collusion

One of the most important issues is algorithmic collusion.

Suppose competing companies use algorithms that continuously monitor competitors' prices.

The algorithm may:

1. observe Competitor A;
2. detect a price increase;
3. immediately increase Competitor B's price;
4. observe whether A follows;
5. punish deviations by lowering prices;
6. eventually stabilise prices at a higher level.

There may be no telephone call or traditional cartel meeting.

Competition law therefore increasingly focuses on economic coordination and the role of technology, rather than simply searching for traditional communications.

The OECD describes three important settings:

• algorithm-facilitated explicit collusion;
• hub-and-spoke coordination;
• autonomous algorithmic coordination. 

7. Real-Time Monitoring and Hub-and-Spoke Coordination

A particularly important structure is:

Competitor A → Common Algorithm/Data Provider ← Competitor B

The common provider may:

• collect competitor data;
• process the information;
• recommend prices;
• monitor compliance;
• punish deviations.

If competing firms knowingly rely on such a system to coordinate behaviour, competition authorities may examine whether the arrangement constitutes an indirect information exchange or concerted practice.

The European Commission's horizontal-cooperation guidance and recent OECD analysis recognise the potential competition concerns where a common pricing tool uses competitively sensitive information belonging to multiple competitors.

8. Six Major Case Laws

1. Eturas UAB and Others v Lietuvos Respublikos konkurencijos taryba — CJEU, 2016

Facts

Eturas operated an online travel-booking platform used by multiple travel agencies.

The platform administrator sent a system message concerning a limitation on discounts. Technical modifications were then implemented so that discounts above the specified level could not be freely offered.

Competition issue

The issue was whether participating travel agencies could be treated as engaging in a concerted practice through the common digital platform.

Decision

The CJEU considered knowledge and participation critical. Agencies receiving the communication and aware of the coordinated practice could face a presumption of participation, subject to circumstances such as publicly distancing themselves.

Relevance to real-time monitoring

The case demonstrates that a common digital infrastructure can become a mechanism for coordinating competitors' commercial behaviour.

It is particularly relevant where monitoring systems:

• communicate competitive information;
• automatically impose parameters;
• transmit common pricing signals.

The OECD identifies Eturas as an important example of digitally mediated coordination.

2. Trod Ltd / GB Eye Ltd — UK Competition and Markets Authority, 2016

Facts

Online sellers of posters agreed not to undercut each other's prices on Amazon Marketplace.

Automated repricing software was used to implement the arrangement.

Competition issue

The important issue was whether automation altered the traditional cartel analysis.

Decision

The CMA treated the arrangement as an infringement involving price coordination.

Principle

An algorithm does not legalise an otherwise unlawful cartel.

If businesses agree on prices and use software to implement that agreement, the technological mechanism is simply an instrument for implementing the anti-competitive arrangement.

Relevance

Real-time monitoring systems can similarly become unlawful if they are used to:

• monitor competitors' compliance with a cartel;
• detect deviations;
• automatically retaliate;
• maintain agreed prices.

The OECD identifies the Trod/GBE case as a key example of algorithms being used to implement collusion.

3. United States v Topkins — U.S. Department of Justice, 2015

Facts

Online sellers of posters agreed to maintain minimum prices for their products sold through Amazon Marketplace.

They used automated pricing software to implement the agreement.

Competition issue

The case demonstrated that algorithmic pricing does not prevent traditional antitrust liability.

Significance

The conduct illustrated a relatively straightforward model:

Human agreement → algorithmic implementation → automated price maintenance

Relevance

For real-time competition monitoring systems, the case establishes an important principle:

Automation is not a defence where technology is being used to implement an underlying anti-competitive agreement.

The OECD specifically identifies the U.S. online-posters case as an example of automated pricing software being used to implement an agreed pricing arrangement.

4. United States v RealPage — algorithmic rental-pricing litigation

Facts and legal controversy

RealPage's software became the subject of U.S. antitrust scrutiny concerning the use of algorithmic pricing systems in rental housing.

The central competition concern involves the possibility that a common pricing system can use information relating to multiple competitors and facilitate coordinated pricing.

Competition issue

The case illustrates the transition from traditional cartel analysis toward:

• common algorithm providers;
• data aggregation;
• pricing recommendations;
• competitor information;
• algorithmic coordination.

Relevance

A real-time monitoring system can potentially become a competitive coordination infrastructure when:

multiple competitors supply information to a common system → the system processes that information → the system recommends or determines commercially significant conduct.

This is closely related to the modern hub-and-spoke theory.

5. FTC v Amazon / Amazon Marketplace-related enforcement concerns

Amazon's marketplace has generated significant competition-law scrutiny concerning the use of data, pricing, seller information and platform mechanisms.

The competition concerns are broader than algorithmic collusion and include questions concerning:

• use of seller data;
• platform self-preferencing;
• Buy Box allocation;
• marketplace competition;
• treatment of third-party sellers.

European and UK authorities have similarly examined how platform access to seller data can affect competition. For example, the UK's Amazon commitments addressed the use of non-public seller data and Buy Box allocation.

Relevance

Real-time monitoring can create a dual-use problem:

The same information system can help regulators identify competition problems while potentially giving a dominant platform powerful information advantages over competitors.

6. Uber / algorithmic pricing jurisprudence

Uber-related competition analysis has generated important discussion concerning algorithmic pricing and the relationship between:

• platform rules;
• driver pricing;
• algorithmic recommendations;
• independent service providers;
• platform control.

The central legal question is whether independent actors using a common platform and algorithm may, in particular circumstances, be participating in coordinated conduct.

Relevance

Real-time systems are particularly important in platform markets because algorithms can continuously:

• observe supply;
• observe demand;
• modify prices;
• allocate customers;
• change incentives;
• respond to competitor conduct.

Consequently, authorities may examine not merely the algorithm's output but also the architecture, data inputs, instructions and degree of human control.

9. Indian Competition-Law Perspective

In India, the principal framework is the Competition Act, 2002.

Real-time competition monitoring can interact particularly with:

Section 3

Potentially covers:

• price fixing;
• information exchange;
• concerted practices;
• bid rigging;
• market allocation.

Section 4

Potential concerns include:

• discriminatory conditions;
• denial of market access;
• unfair pricing;
• exclusionary conduct;
• leveraging dominance.

Combination regulation

Real-time monitoring can also assist authorities in identifying rapidly changing market structures and possible effects of digital acquisitions and combinations.

The CCI's contemporary digital-market work recognises algorithmic pricing, algorithmic collusion and discriminatory digital practices as emerging competition issues.

10. Real-Time Monitoring and Abuse of Dominance

A dominant platform may operate a monitoring system that observes:

• competitors' prices;
• seller margins;
• inventory;
• customer conversion;
• transaction volumes.

If the platform uses such information to disadvantage competitors, questions may arise concerning:

A. Self-preferencing

The dominant platform could favour its own products after obtaining competitors' information.

B. Margin squeezing

The platform could use real-time data to manipulate upstream and downstream pricing.

C. Predatory pricing

Algorithms could continuously adjust prices to eliminate competitors.

D. Discriminatory treatment

Different businesses could receive different:

• rankings;
• commissions;
• visibility;
• access conditions.

E. Exclusionary conduct

The monitoring system could identify emerging competitors and allow rapid strategic responses intended to foreclose them.

11. Dynamic Pricing and Competition

Real-time monitoring is closely connected with dynamic pricing.

Dynamic pricing itself is not inherently anti-competitive.

It can produce legitimate efficiencies through:

• demand forecasting;
• inventory optimisation;
• congestion management;
• personalised offers;
• capacity utilisation.

However, competition concerns arise where dynamic pricing becomes a mechanism for:

• price coordination;
• discriminatory exclusion;
• retaliation against competitors;
• signalling future prices;
• enforcement of cartel arrangements.

The CCI has recognised that dynamic pricing can improve efficiency while simultaneously raising concerns about consumer exploitation, price discrimination and algorithmic coordination.

12. Real-Time Monitoring in Procurement and Bid Rigging

Real-time monitoring can be especially valuable in public procurement.

An authority can monitor:

• identical bid prices;
• suspicious bid rotation;
• repeated winning patterns;
• geographic allocation;
• unusually low bids;
• identical tender errors;
• withdrawal patterns;
• coordinated subcontracting.

Example

If Companies A, B and C repeatedly participate in government tenders and:

• A wins Tender 1;
• B wins Tender 2;
• C wins Tender 3;
• losing bidders submit suspiciously similar prices,

an automated monitoring system can flag the pattern for investigation.

Importantly, an alert is evidence for investigation, not proof of infringement.

This distinction is fundamental because parallel conduct can arise from legitimate market conditions.

13. Evidence Generated by Monitoring Systems

Real-time monitoring can create extensive evidence:

Digital evidence

• timestamped price records;
• API logs;
• algorithmic outputs;
• source-code versions;
• server records;
• database entries;
• communications;
• audit trails.

Analytical evidence

• price correlation;
• deviation patterns;
• market-share changes;
• bidding sequences;
• reaction-time analysis.

Forensic evidence

Authorities may examine:

• algorithm instructions;
• training data;
• model parameters;
• logs;
• human overrides;
• software updates.

The evidentiary significance of such material depends upon whether it establishes an agreement, communication, knowledge, intent where legally relevant, market effects or another element of the applicable infringement.

14. Risk of False Positives

A major problem with real-time monitoring is that parallel conduct does not necessarily equal collusion.

Prices may move together because:

• input costs changed;
• taxes changed;
• exchange rates moved;
• demand changed;
• a common supplier increased prices;
• market-wide shortages occurred;
• competitors independently responded to the same public information.

Therefore:

Algorithmic alert ≠ infringement

A responsible competition-monitoring system should distinguish:

detection → investigation → evidence → legal assessment → enforcement.

It should not automatically convert an anomaly into a finding of liability.

15. Data Governance and Privacy

Real-time monitoring may involve enormous quantities of data.

Competition authorities and businesses must therefore consider:

• data minimisation;
• confidentiality;
• cybersecurity;
• access controls;
• personal-data protection;
• retention periods;
• purpose limitation;
• secure transmission.

A competition authority's monitoring system should ideally separate:

competitively sensitive information
from
information legitimately required for market surveillance.

16. Compliance Monitoring by Businesses

Businesses can themselves use real-time systems to prevent competition violations.

For example, an internal compliance system could alert management when:

• an employee contacts a competitor concerning future prices;
• a pricing algorithm uses competitor-confidential data;
• a sales employee receives sensitive competitor information;
• a tender pattern becomes unusual;
• a common pricing vendor is used by multiple competitors.

This transforms technology from merely an enforcement risk into a compliance mechanism.

17. Competition Concerns Associated With Monitoring Systems

18. Safeguards for Lawful Monitoring

A competition-compliant monitoring system should incorporate:

1. Data aggregation

Use aggregated information wherever possible.

2. Time lag

Historical rather than real-time data may reduce coordination risks.

3. Access restrictions

Competitors should not receive each other's confidential information.

4. Independent governance

The system should have clear controls preventing competitors from accessing sensitive information.

5. Audit trails

Every access and modification should be recorded.

6. Algorithmic transparency

The system should be capable of explaining:

• what data it uses;
• how it processes information;
• what recommendations it generates.

7. Human review

Automated alerts should normally trigger investigation rather than automatic enforcement.

8. Competition-law compliance testing

Algorithms should be periodically tested for:

• coordinated pricing;
• discriminatory treatment;
• exclusionary effects;
• unlawful information exchange.

19. Emerging Issue: Autonomous Algorithmic Collusion

The most difficult future question is whether competing AI systems can independently learn to coordinate without an express human agreement.

For example:

Firm A AI → observes Firm B → changes price → Firm B AI observes A → responds → repeated interaction → stable supra-competitive prices

This creates a difficult legal distinction between:

• independent parallel behaviour;
• conscious parallelism;
• algorithm-facilitated coordination;
• explicit agreement;
• concerted practice.

The OECD notes that AI can increase transparency and reaction speed, potentially making coordination more stable because algorithms can rapidly observe and respond to deviations.

20. Role of Competition Authorities

Modern competition authorities may increasingly operate a continuous competition intelligence system involving:

Market data

↓

Automated collection

↓

Real-time analytics

↓

Anomaly detection

↓

Competition-risk alert

↓

Human investigation

↓

Evidence gathering

↓

Legal assessment

↓

Enforcement / remedy

This is likely to become increasingly important in:

• e-commerce;
• digital advertising;
• ride-hailing;
• airline pricing;
• online accommodation;
• energy markets;
• financial markets;
• telecommunications;
• public procurement;
• logistics.

21. Key Legal Principles

The principal principles can be summarised as follows:

1. Technology does not change the fundamental prohibition on cartel conduct.
2. Automated monitoring can be lawful and beneficial when used for independent compliance or enforcement.
3. A common monitoring or pricing system can create information-exchange risks when competitors receive competitively sensitive information.
4. A platform or software provider may potentially function as a hub facilitating coordination.
5. Parallel algorithmic pricing alone should not automatically be treated as proof of collusion.
6. Knowledge, participation, communications, system design and market circumstances may be important in determining liability.
7. Real-time data can increase both enforcement capability and competition risk.
8. Human oversight remains important because algorithmic correlation can produce false positives.
9. Dominant platforms must consider whether monitoring data is being used to disadvantage rivals.
10. Competition authorities increasingly need technological capability to investigate algorithms, databases and automated decision-making systems.

22. Conclusion

Real-time competition monitoring systems represent a significant development in modern competition law. They can allow authorities and businesses to identify suspicious market behaviour far more rapidly than traditional monitoring methods.

At the same time, the same technological infrastructure can create serious competition concerns when competitors receive continuous access to commercially sensitive information or rely upon a common algorithm to determine pricing and strategic decisions.

The cases of Eturas, Trod/GB Eye and Topkins, together with modern algorithmic-pricing and platform investigations, demonstrate an important principle: competition law is concerned with the competitive effect and legal character of the conduct, not merely whether humans or algorithms performed the relevant action.

The emerging regulatory challenge is therefore to maintain a distinction between:

legitimate real-time competition surveillance
and
real-time technological facilitation of coordination or exclusion.