Competition Law and Competition Governance in Computational Legal Systems

Introduction

Computational legal systems are legal, regulatory, and commercial environments in which software, artificial intelligence, algorithms, automated decision-making systems, smart contracts, legal-tech platforms, data infrastructures, and machine-readable rules participate directly in market governance.

Competition law in such systems must address a fundamental transformation: market conduct may no longer be determined exclusively by human decision-makers. Pricing, ranking, matching, access, contracting, enforcement, compliance, and even dispute resolution can increasingly be performed through computational systems.

The principal competition-law question is therefore not merely whether a company has violated traditional antitrust rules, but also how competition should be governed when economically significant decisions are embedded in code, algorithms, data architectures, and automated institutional systems.

I. Meaning of Computational Legal Systems

A computational legal system may be understood as a system in which legal rules, contractual obligations, regulatory requirements, or market decisions are:

1. represented digitally;
2. processed algorithmically;
3. automatically applied or enforced;
4. connected with economic transactions; and
5. capable of affecting market access or competitive conditions.

Examples include:

• algorithmic pricing systems;
• AI-driven procurement;
• automated compliance platforms;
• smart contracts;
• blockchain-based marketplaces;
• digital courts and dispute-resolution platforms;
• automated licensing systems;
• algorithmic credit allocation;
• digital public infrastructure;
• regulatory technology platforms;
• automated merger-monitoring systems;
• platform ranking and recommendation systems; and
• machine-readable contractual ecosystems.

II. Competition-Law Issues

1. Algorithmic Collusion

One of the most significant concerns is that algorithms may facilitate coordination between competitors.

Traditional cartel law normally looks for:

• an agreement;
• concerted practice;
• communication;
• coordination; or
• conscious parallelism.

Computational markets complicate this analysis because algorithms may independently observe competitors and repeatedly adjust prices.

Problem

Two competing firms may use similar pricing algorithms. Neither may explicitly communicate with the other, yet the algorithms may learn that maintaining higher prices produces greater returns.

The competition-law challenge becomes:

When does autonomous algorithmic adaptation become unlawful coordination?

III. Algorithmic Hub-and-Spoke Coordination

A platform may operate as a hub, while competing sellers operate as spokes.

The platform can:

• collect pricing information;
• recommend prices;
• enforce pricing rules;
• monitor deviations;
• penalize discounts; and
• automatically adjust marketplace conditions.

The resulting arrangement can potentially reproduce the economic effects of a cartel even when direct communication between competitors is absent.

IV. Abuse of Dominance Through Computational Systems

A dominant digital undertaking can potentially use algorithms to exclude rivals.

Examples include:

A. Algorithmic self-preferencing

A platform may rank its own products above competing products.

B. Algorithmic discrimination

The system may give different:

• prices;
• rankings;
• commissions;
• access conditions; or
• visibility

to similarly situated businesses.

C. Automated exclusion

An algorithm may automatically downgrade or exclude competitors.

D. Data foreclosure

A dominant platform may prevent competitors from obtaining data necessary to compete.

E. Interoperability restrictions

A platform may technically prevent competing services from interacting with its infrastructure.

V. Essential Facilities and Computational Infrastructure

Some computational systems may become economically indispensable.

Examples include:

• digital identity infrastructure;
• payment rails;
• cloud infrastructure;
• app stores;
• operating systems;
• digital advertising exchanges;
• interoperability protocols;
• large-scale datasets; and
• API infrastructure.

Where a dominant undertaking controls an indispensable computational infrastructure, competition authorities may examine whether refusal of access constitutes exclusionary conduct.

The traditional essential-facilities doctrine therefore acquires a technological dimension.

VI. Data as a Competitive Resource

Data can function as an important competitive input.

Computational legal systems can create competition concerns where one undertaking controls:

• transaction data;
• consumer behavioural data;
• industrial datasets;
• search data;
• mobility data;
• financial data; or
• platform interaction data.

Competition authorities may therefore examine:

1. who controls the data;
2. whether rivals can obtain comparable data;
3. whether data portability is available;
4. whether data is technically interoperable;
5. whether access is discriminatory; and
6. whether data accumulation creates durable market power.

VII. AI and Competition Law

AI introduces several new dimensions.

An AI system can potentially:

• determine prices;
• rank competitors;
• select suppliers;
• determine advertising allocation;
• identify customers;
• recommend contractual terms;
• allocate scarce resources; and
• predict competitors' behaviour.

The competition-law question is consequently shifting from:

Who made the decision?

to:

Who designed, trained, deployed, controlled, and benefited from the computational decision system?

Responsibility may potentially extend beyond the immediate user of an algorithm to:

• platform operators;
• software developers;
• data suppliers;
• algorithm vendors;
• system integrators; and
• controlling firms,

depending on applicable law and the evidence establishing their involvement.

VIII. Smart Contracts and Competition Law

Smart contracts can automatically execute commercial arrangements.

They can:

• determine prices;
• distribute revenues;
• impose contractual penalties;
• restrict transactions;
• automatically terminate relationships; and
• enforce exclusivity.

Their automated nature does not necessarily remove them from competition-law scrutiny.

If the underlying arrangement is anti-competitive, automation may simply make implementation faster and more persistent.

IX. Computational Governance and Merger Control

Computational systems also affect merger analysis.

A merger between two technology companies may combine:

• datasets;
• algorithms;
• cloud infrastructure;
• user networks;
• AI models;
• APIs;
• intellectual property; and
• distribution channels.

Traditional turnover thresholds may fail to capture acquisitions of innovative or data-rich firms with relatively low revenues.

Consequently, modern competition governance increasingly considers:

• innovation competition;
• potential competition;
• data concentration;
• ecosystem effects;
• network effects;
• interoperability;
• algorithmic capabilities; and
• future competitive constraints.

X. Relevant Case Laws

1. United States v. Apple Inc. — U.S. Department of Justice

The Apple litigation concerns alleged exclusionary conduct involving the iPhone ecosystem.

The broader competition-law significance lies in examining how control over a technologically integrated ecosystem can affect:

• application distribution;
• payments;
• interoperability;
• switching;
• access to platform functionality; and
• competition from complementary services.

Principle

Computational architecture can itself become an instrument through which market power is exercised.

2. Epic Games, Inc. v. Apple Inc.

The litigation involving Epic Games and Apple examined Apple's rules governing the iOS ecosystem and App Store.

Issues included:

• platform restrictions;
• payment systems;
• distribution;
• developer access;
• commissions; and
• restrictions on alternative payment mechanisms.

Computational-governance significance

The case demonstrates that technical architecture and contractual rules can operate together as competitive restraints.

A platform does not necessarily compete only through prices. Its control over the technical environment can determine which competitors are able to reach consumers.

3. United States v. Google LLC — Search and Search Advertising

The Google search litigation examined Google's conduct concerning search distribution and related competitive arrangements.

The case is important for computational competition law because search markets depend heavily upon:

• algorithms;
• default settings;
• data;
• scale;
• search quality;
• distribution agreements; and
• feedback effects.

Principle

Algorithmically mediated markets can create competitive advantages that reinforce themselves through scale and data.

4. Google Shopping — European Commission

The European Commission's Google Shopping decision concerned Google's treatment of its comparison-shopping service in search results.

The Commission found that Google had given prominent placement to its own comparison-shopping service while applying demotion mechanisms to competing services.

Computational significance

This is a central example of algorithmic ranking becoming a competition-law issue.

The ranking algorithm was not merely a neutral technical instrument; its operation affected the visibility of competing businesses.

Principle

Where a dominant digital platform controls the ranking mechanism through which competitors reach consumers, discriminatory ranking can have exclusionary consequences.

5. Google Android — European Commission

The Android decision concerned several practices associated with Google's Android ecosystem, including arrangements concerning:

• search;
• browser distribution;
• app stores;
• licensing; and
• device manufacturers.

Computational significance

The case illustrates how an integrated digital ecosystem can combine:

• operating-system control;
• application distribution;
• default arrangements;
• data advantages; and
• network effects.

Principle

Competition analysis may need to consider the entire computational ecosystem, rather than examining each technical component in isolation.

6. Amazon Marketplace — European Commission

The European Commission investigated Amazon's use of marketplace data and its relationship with third-party sellers.

The concern included the possibility that Amazon could use competitively sensitive information generated by independent sellers to strengthen its own retail activities.

Computational significance

The case demonstrates the competitive importance of data generated by a platform's users and business partners.

Principle

A platform that simultaneously operates infrastructure and competes with businesses using that infrastructure may create a structural conflict between:

• platform governance; and
• competitive neutrality.

7. FTC v. Amazon

The U.S. Federal Trade Commission's litigation against Amazon addresses alleged practices involving Amazon's marketplace and its competitive environment.

Among the issues examined are:

• seller practices;
• pricing;
• marketplace rules;
• advertising;
• fulfillment;
• competition among sellers; and
• platform control.

Computational significance

The case illustrates how platform rules, ranking, marketplace architecture, and automated systems can collectively influence competitive conditions.

8. Eturas v. Lietuvos Respublikos konkurencijos taryba — CJEU

Eturas is particularly relevant to algorithmic coordination.

An online travel-booking platform sent a system-wide message concerning a limitation on discounts available to participating travel agencies.

The Court of Justice considered whether participating undertakings could be treated as engaging in concerted practice where they received and potentially acted upon the platform's message.

Principle

Digital platforms can facilitate coordination among competitors, and competition law can examine the conduct of participants even where coordination occurs through an electronic system.

This case is especially important for understanding the transition from human communication to digitally mediated coordination.

9. Uber Spain — Asociación Profesional Elite Taxi v Uber Systems Spain

The CJEU considered the nature of Uber's service and its relationship with the transportation market.

Although not primarily an antitrust case, it has broader significance for computational-market governance because Uber's business model relied heavily upon:

• digital intermediation;
• algorithmic matching;
• platform coordination;
• pricing systems; and
• network effects.

Principle

Competition regulation of digital markets may require identifying the economic substance of the platform's activity, rather than focusing exclusively on its technological form.

X. Competition Governance Architecture

Competition governance in computational legal systems should operate at several levels.

1. Ex Ante Regulation

Authorities can establish rules before anti-competitive effects become entrenched.

Possible mechanisms include:

• interoperability obligations;
• data portability;
• transparency requirements;
• access obligations;
• merger notification rules;
• algorithmic accountability;
• non-discrimination requirements; and
• restrictions on self-preferencing.

2. Ex Post Enforcement

Traditional competition law remains important.

Authorities can investigate:

• cartels;
• abuse of dominance;
• exclusionary conduct;
• discriminatory access;
• tying;
• refusal to deal;
• predatory strategies;
• anti-competitive agreements; and
• unlawful mergers.

XI. Algorithmic Auditing

Competition authorities may increasingly need technical auditing capabilities.

An algorithmic competition audit may examine:

1. input data;
2. model architecture;
3. optimization objectives;
4. pricing variables;
5. ranking criteria;
6. feedback mechanisms;
7. competitor monitoring;
8. human intervention;
9. automated enforcement; and
10. historical decision logs.

This creates an important concept:

Competition law enforcement may require both legal evidence and computational evidence.

XII. Explainability and Evidentiary Problems

Computational systems create unusual evidentiary difficulties.

A competition authority may need to establish:

• what the algorithm did;
• why it produced a particular result;
• who designed the objective;
• whether employees intervened;
• what data was used;
• whether competitors' information was considered;
• whether the system learned independently; and
• whether the resulting conduct was foreseeable.

This makes algorithmic explainability and audit trails increasingly important.

XIII. Human Responsibility for Autonomous Systems

A major governance question is whether an undertaking can argue:

"The algorithm made the decision."

Competition law generally cannot function effectively if automation becomes a complete shield against liability.

The relevant investigation may therefore examine:

• who commissioned the system;
• who selected its objectives;
• who supplied the data;
• who controlled its deployment;
• who monitored its results;
• who benefited economically; and
• whether safeguards were implemented.

Automation may change the mechanism of conduct without necessarily eliminating the legal responsibility of the undertaking operating the system.

XIV. Computational Neutrality

A new regulatory principle can be described as computational neutrality.

A platform performing a quasi-market-governance function should not necessarily be permitted to use its control over:

• rankings;
• APIs;
• data;
• identity;
• payment infrastructure;
• technical standards; or
• access protocols

to systematically disadvantage competing businesses.

This is particularly significant where the platform simultaneously acts as:

1. infrastructure provider;
2. rule-maker;
3. market intermediary; and
4. competitor.

XV. Interoperability as a Competition Remedy

Interoperability can become an important remedy.

Authorities may require dominant systems to permit:

• API access;
• data portability;
• technical interoperability;
• messaging interoperability;
• payment interoperability; or
• compatibility with competing services.

The purpose is not necessarily to eliminate large firms, but to reduce artificial barriers that prevent competitors from exercising meaningful competitive constraints.

XVI. Data Portability and Switching

Consumers and businesses may become locked into computational ecosystems because of:

• accumulated data;
• learning histories;
• reputation scores;
• transaction records;
• proprietary formats; and
• network relationships.

Data portability can reduce switching costs and potentially facilitate entry.

XVII. Competition Between Algorithms

A computational market may contain multiple layers of competition:

Layer 1 — Human firms

Traditional undertakings compete.

Layer 2 — Algorithms

Algorithms determine prices, rankings, or allocations.

Layer 3 — Platforms

Platforms control the infrastructure.

Layer 4 — Data

Data determines algorithmic performance.

Layer 5 — Ecosystems

Entire digital ecosystems compete with each other.

Competition law therefore increasingly needs an ecosystem perspective.

XVIII. Key Challenges for Competition Authorities

A. Attribution

Who is responsible for an autonomous decision?

B. Detection

How can hidden algorithmic coordination be discovered?

C. Proof

What evidence establishes algorithmic intent or effect?

D. Technical expertise

Can regulators understand complex AI systems?

E. Speed

Can enforcement respond quickly enough to rapidly changing digital markets?

F. Cross-border enforcement

Algorithms may operate simultaneously across numerous jurisdictions.

G. Regulatory fragmentation

Competition, data-protection, AI, consumer-protection, financial, and cybersecurity regulators may investigate overlapping conduct.

XIX. Future Competition-Governance Model

A mature computational competition regime could combine:

Competition Law + AI Governance + Data Governance + Platform Regulation + Consumer Protection + Cybersecurity + Technical Auditing

A possible institutional model is:

Market Monitoring → Algorithmic Risk Assessment → Data/Architecture Audit → Competition Investigation → Interim Measures → Remedy → Continuous Monitoring

This represents a movement from traditional event-based enforcement toward continuous computational market governance.

XX. Major Legal Principles Emerging

The case law and regulatory developments discussed above support several broad principles:

1. Algorithms are not outside competition law merely because they operate automatically.
2. Digital platforms can facilitate coordination among competitors.
3. Algorithmic ranking can have competitive significance.
4. Data concentration can contribute to market power.
5. Technical architecture can constitute a mechanism of exclusion.
6. Platform governance rules can have competition-law consequences.
7. Interoperability may become an important competition remedy.
8. Smart contracts cannot automatically immunize anti-competitive arrangements.
9. Competition authorities increasingly require technical evidence.
10. Computational systems should be assessed according to both their legal structure and their economic effects.

XXI. Conclusion

Competition law in computational legal systems represents the convergence of antitrust law, digital regulation, AI governance, data governance, and technological infrastructure regulation.

The central transformation is from markets governed predominantly by human decisions to markets increasingly governed by code, algorithms, platforms, data, and automated rules.

The cases involving Google Shopping, Google Android, Amazon, Eturas, Apple, Epic Games, Uber, and Google Search demonstrate different dimensions of this transformation: algorithmic ranking, ecosystem control, platform data, digitally mediated coordination, technical restrictions, and automated market intermediation.

The future challenge for competition governance is therefore not simply to determine whether computational systems should be regulated. It is to develop legal mechanisms capable of ensuring that automation does not conceal coordination, technical architecture does not become an instrument of exclusion, and control over data or digital infrastructure does not eliminate effective competition.