Competition Law and Machine-Mediated Supply Chain Governance

1. Introduction

Machine-mediated supply chain governance refers to the use of artificial intelligence (AI), machine learning, automated contracting systems, algorithms, Internet of Things (IoT) devices, predictive analytics, blockchain, and digital platforms to manage relationships across a supply chain.

A modern supply chain may be governed by machines at several levels:

Supplier → Manufacturer → Distributor → Platform → Retailer → Consumer

At each stage, algorithms may determine:

supplier selection;

prices;

inventory levels;

delivery schedules;

allocation of orders;

quality standards;

access to platforms;

credit limits;

discounts;

logistics;

customer allocation.

These technologies can produce significant efficiencies. However, they can also create competition-law concerns where a powerful platform or automated system is used to exclude competitors, coordinate suppliers, discriminate among distributors, restrict market access, facilitate collusion, or reinforce market power.

Machine-mediated supply-chain governance is not itself a separate competition-law offence. Existing principles concerning cartels, abuse of dominance, vertical restraints, tying, exclusive dealing, refusal to deal, discrimination, information exchange, and merger control remain applicable.

2. Meaning of Machine-Mediated Supply Chain Governance

Traditional supply-chain governance relies heavily on human decisions.

For example:

A manufacturer negotiates prices with suppliers and manually allocates production.

A machine-mediated system might instead operate as:

Data → Algorithm → Supplier evaluation → Automated allocation → Contract execution → Monitoring

The machine may continuously assess:

price;

quality;

delivery performance;

inventory;

demand;

creditworthiness;

geographic coverage.

The system can then automatically modify commercial relationships.

3. Main Technologies

Machine-mediated supply chains can involve:

1. Artificial intelligence

Used for:

demand forecasting;

supplier selection;

risk assessment;

logistics optimization.

2. Machine learning

Used to identify:

supplier performance;

demand patterns;

price trends;

delivery risks.

3. IoT

Connected devices can provide real-time information about:

inventory;

production;

transportation;

equipment.

4. Blockchain

Can facilitate:

transaction records;

automated verification;

smart contracts;

supply-chain traceability.

5. Smart contracts

Contracts can automatically execute when predefined conditions are satisfied.

6. Digital platforms

Platforms can connect:

suppliers;

manufacturers;

distributors;

retailers;

consumers.

4. Why Competition Law Is Relevant

Machine-mediated governance can change the competitive structure of a supply chain.

A powerful undertaking may control:

the platform;

data;

logistics;

suppliers;

distribution;

payment systems;

algorithms.

This can create vertical leverage.

For example:

A dominant marketplace controls both the digital platform and logistics network and requires sellers to use its logistics service.

This could raise questions concerning:

tying;

bundling;

foreclosure;

self-preferencing;

refusal to deal;

discriminatory access.

However, vertical integration or automated management is not automatically unlawful.

The competitive effects must be examined.

5. Vertical Supply Chains

Competition law distinguishes between:

Horizontal relationships

Competitors at the same level.

Example:

Manufacturer A ↔ Manufacturer B

Vertical relationships

Businesses operating at different levels.

Example:

Manufacturer → Distributor → Retailer

Machine-mediated supply chains can affect both.

6. Vertical Foreclosure

A major risk occurs when a dominant company uses control over one level of the supply chain to disadvantage competitors at another level.

Example:

A dominant cloud provider acquires an important AI infrastructure supplier and gives preferential access to its own AI products.

Potential concern:

Input foreclosure

The company may make an important input less accessible to competing downstream firms.

Another possibility is:

Customer foreclosure

A powerful distributor may restrict suppliers from reaching competing distribution channels.

7. Algorithmic Supplier Selection

Algorithms may rank suppliers according to:

price;

quality;

reliability;

delivery speed;

geographic location;

historical performance.

This can improve efficiency.

However, competition concerns can arise if a dominant platform's algorithm systematically disadvantages competing suppliers without legitimate justification.

Possible issues include:

discriminatory access;

exclusion;

self-preferencing;

retaliation;

preferential treatment for affiliated suppliers.

8. Self-Preferencing

Self-preferencing occurs where a platform gives preferential treatment to its own products or services compared with competing products.

A machine-mediated supply chain may automatically rank:

Platform's own product → Position 1

while competing supplier products appear lower.

The competition-law question is whether such conduct produces unlawful exclusionary effects.

9. Algorithmic Discrimination

A supply-chain algorithm may offer:

lower fees to affiliated suppliers;

faster delivery;

better search rankings;

better credit terms;

greater inventory access.

If competitors are treated less favourably, authorities may investigate whether the differentiation is:

objectively justified;

efficiency-based;

discriminatory;

exclusionary.

The existence of an algorithm does not itself establish discrimination.

10. Machine-Mediated Information Exchange

Supply-chain platforms can collect huge amounts of information.

For example:

supplier prices;

production volumes;

inventory;

delivery schedules;

customer demand;

future supply plans.

This creates competition risks where sensitive information concerning competing firms is exchanged or used to coordinate behaviour.

11. Algorithmic Coordination Among Suppliers

Suppose several competing manufacturers use the same supply-chain platform.

The platform collects their:

prices;

inventory;

future production;

discounts.

If the platform uses that information to recommend coordinated prices, competition concerns may arise.

This resembles the broader problem of algorithmic collusion.

12. Hub-and-Spoke Risk

A platform can act as the central “hub” connecting competing businesses.

Example:

Supplier A → Digital Platform ← Supplier B

If the platform facilitates an agreement among A and B, the platform's role may become competition-law relevant.

However, simply using the same platform does not prove an unlawful arrangement.

Evidence concerning:

knowledge;

communications;

platform design;

instructions;

data flows;

commercial conduct

would be important.

13. Exclusive Supply Arrangements

Machine-mediated systems can automatically impose:

exclusivity;

minimum purchase requirements;

preferred supplier arrangements;

minimum inventory obligations.

Exclusive arrangements may have legitimate commercial purposes.

But they can become competition concerns where a dominant company uses them to foreclose rivals.

14. Loyalty Mechanisms

Algorithms may automatically reward suppliers that:

use only one platform;

meet certain sales targets;

maintain minimum inventory;

avoid competing platforms.

This can create loyalty or exclusivity effects.

Competition authorities may examine:

market power;

duration;

coverage;

foreclosure;

availability of alternatives;

efficiencies.

15. Tying and Bundling

Machine-mediated supply chains may combine several services.

For example:

“To access our marketplace, you must use our logistics service.”

Or:

“To receive favourable algorithmic ranking, you must purchase our payment service.”

Such arrangements may raise tying or bundling concerns where the undertaking has significant market power and the arrangement forecloses competitors.

16. Refusal of Access

A dominant digital supply-chain platform may control an important:

marketplace;

logistics network;

data system;

API;

warehouse network;

payment infrastructure.

If access is denied to competitors, the legal analysis may involve refusal-to-deal or essential-facilities principles, depending on the jurisdiction.

But a company does not automatically have a competition-law obligation to provide access to every competitor.

17. Data as a Supply-Chain Asset

Modern supply chains generate extensive data.

Examples:

supplier performance;

product demand;

consumer behaviour;

delivery information;

pricing;

inventory.

Data can become a competitive input.

A dominant undertaking's control over unique supply-chain data may increase barriers to entry.

However:

Control over data does not automatically establish dominance or an obligation to share it.

Its competitive significance must be assessed in context.

18. Smart Contracts and Competition Law

Smart contracts automatically execute transactions when predetermined conditions are met.

For example:

If inventory falls below 1,000 units → automatically place an order.

This can increase efficiency.

But smart contracts can also be used to implement:

resale-price restrictions;

exclusivity;

discriminatory access;

coordinated pricing;

automatic penalties.

The technology does not remove competition-law scrutiny.

19. Resale Price Maintenance

A manufacturer may use software to monitor retailers' prices.

If the system automatically penalizes retailers that sell below a specified price, this can raise resale-price-maintenance (RPM) concerns in jurisdictions where RPM is restricted.

The algorithm may simply automate what could otherwise be a traditional contractual or monitoring mechanism.

20. Geographic Restrictions

Machine systems can automatically prevent suppliers or distributors from serving certain territories.

For example:

Distributor A's software prevents sales to customers in Distributor B's territory.

Territorial restrictions may sometimes be legitimate but can raise competition concerns depending on the applicable legal regime and circumstances.

21. Input Foreclosure

Suppose a dominant manufacturer controls a critical input and uses an algorithm to allocate that input.

The algorithm could:

reduce supply to rivals;

prioritize affiliated businesses;

delay deliveries to competitors;

increase prices for rivals.

The competition question is whether the system produces unlawful foreclosure.

22. Customer Foreclosure

The reverse can occur.

A powerful distributor may control access to customers.

Its algorithm could:

favour selected suppliers;

exclude rivals;

restrict product visibility;

allocate shelf or platform space.

This may disadvantage suppliers that depend on the distributor.

23. Dynamic Competition

Machine-mediated supply chains can make competition highly dynamic.

A company may compete not only through today's price but through:

faster delivery;

superior forecasting;

better inventory;

technological innovation;

logistics networks.

Competition authorities therefore need to consider dynamic efficiency and innovation, not merely current prices.

24. Important Case Laws

Because “machine-mediated supply chain governance” is a modern conceptual category, the cases below are foundational or analogous authorities rather than cases that directly concern AI-controlled supply chains.

Case 1: United States v. Microsoft Corp.

Citation: 253 F.3d 34 (D.C. Cir. 2001)

Principle

The case concerned Microsoft's use of its operating-system position and contractual/technical strategies affecting browser competition.

Relevance

It demonstrates how a powerful firm can use control at one level of a technological ecosystem to influence competition at another level.

This is analogous to a dominant supply-chain platform using its infrastructure or algorithms to disadvantage competing suppliers.

25. Case 2: United States v. Terminal Railroad Association

Citation: 224 U.S. 383 (1912)

Principle

The Supreme Court addressed control over an important transportation facility and the competitive implications of denying meaningful access.

Relevance

The case is a foundational authority for access-related competition concerns.

Modern machine-mediated supply chains may similarly depend upon:

logistics networks;

digital platforms;

warehouses;

APIs;

data infrastructure.

The analogy is that control over an important bottleneck can affect downstream competition.

26. Case 3: Aspen Skiing Co. v. Aspen Highlands Skiing Corp.

Citation: 472 U.S. 585 (1985)

Principle

The Supreme Court considered exclusionary conduct involving a dominant ski operator's termination of cooperation with a smaller rival.

Relevance

The case illustrates the narrow circumstances in which termination of an established commercial relationship can raise monopolization concerns.

In a machine-mediated supply chain, automated termination of a supplier relationship should not automatically be treated as unlawful. The broader factual and competitive context remains important.

27. Case 4: Verizon Communications Inc. v. Trinko

Citation: 540 U.S. 398 (2004)

Principle

The Supreme Court emphasized that competition law generally does not impose a broad obligation on firms to share their assets with competitors.

Relevance

This is important for digital supply-chain infrastructure.

A company controlling:

data;

software;

logistics;

APIs;

infrastructure

does not automatically have a competition-law duty to provide access.

The strict limits on compulsory access must be considered.

28. Case 5: Bronner v. Mediaprint

Citation: Case C-7/97, Court of Justice of the European Union, 1998

Principle

The Court established stringent conditions for treating refusal to provide access to a facility as an abuse of dominance.

Relevance

This is directly useful by analogy where a dominant machine-mediated supply-chain platform controls an important distribution or delivery network.

The fact that access would be useful to competitors is not enough; the legal criteria for compulsory access must be satisfied.

29. Case 6: IMS Health GmbH & Co. KG v. NDC Health

Citation: Joined Cases C-418/01, Court of Justice of the European Union, 2004

Principle

The Court considered refusal to license intellectual property and the exceptional circumstances in which compulsory access may be required.

Relevance

Modern machine-mediated supply chains may depend on:

proprietary software;

databases;

APIs;

technical standards;

digital infrastructure.

IMS Health provides an important framework for analysing demands for access to such proprietary systems.

30. Case 7: Google Shopping

Case: Google and Alphabet v European Commission

Citation: Case T-612/17, General Court, 2021

Principle

The case concerned Google's treatment of competing comparison-shopping services and preferential positioning within its search ecosystem.

Relevance

The case is important by analogy to algorithmic self-preferencing.

A supply-chain platform may use an algorithm to rank:

its own suppliers;

affiliated logistics services;

private-label products.

The competitive effects of such preferential treatment may require investigation.

31. Case 8: Slovak Telekom v Commission

Citation: Joined Cases C-152/19 P and C-165/19 P, Court of Justice of the European Union, 2021

Principle

The case involved exclusionary conduct and access to telecommunications infrastructure.

Relevance

It illustrates how control over an important upstream network can affect downstream competitors.

The same analytical logic may become relevant to machine-mediated infrastructure such as:

digital logistics;

cloud systems;

supply-chain platforms;

telecommunications networks.

32. Case 9: Intel Corp. v Commission

Citation: Case C-413/14 P, Court of Justice of the European Union, 2017

Principle

The Court required proper consideration of the effects of certain exclusivity-related practices where the undertaking is dominant.

Relevance

Automated supply-chain systems may create:

exclusive-supplier arrangements;

loyalty incentives;

conditional rebates.

Intel demonstrates the importance of analysing their actual or potential foreclosure effects rather than treating every commercial incentive as automatically unlawful.

33. Case 10: Hoffmann-La Roche v Commission

Citation: Case 85/76, Court of Justice of the European Communities, 1979

Principle

The Court established important principles concerning dominance and exclusionary loyalty arrangements.

Relevance

Machine-mediated supplier systems could automatically reward suppliers for exclusive or loyalty-based relationships.

Where a dominant undertaking is involved, such mechanisms may require careful competition analysis.

34. Machine-Mediated Supply Chain and Competition Effects

A useful framework is:

Stage 1 — Input

What information enters the machine?

↓

Stage 2 — Processing

How does the algorithm process the information?

↓

Stage 3 — Decision

What supplier/distributor decision does it make?

↓

Stage 4 — Implementation

Is the decision automatically enforced?

↓

Stage 5 — Market Effect

Does it:

reduce competition?

increase efficiency?

exclude competitors?

facilitate coordination?

This framework prevents the technology itself from becoming the focus of the legal analysis.

35. Legitimate Efficiency vs Anti-Competitive Governance

36. Competition Risks for Small Businesses

Machine-mediated systems can create particular challenges for smaller suppliers.

Potential barriers include:

algorithmic exclusion;

high platform fees;

lack of access to data;

automated delisting;

opaque ranking systems;

dependence on a single platform;

inability to challenge algorithmic decisions.

But small-firm disadvantage alone does not necessarily establish an antitrust violation.

The key question is whether the conduct harms the competitive process.

37. Algorithmic Transparency

A dominant platform may need to consider whether suppliers can understand:

ranking criteria;

eligibility conditions;

pricing rules;

penalties;

termination criteria.

Competition law does not universally require complete disclosure of algorithms.

However, opaque systems can make it harder to determine whether discrimination or exclusion is occurring.

38. Interoperability

Machine-mediated supply chains may rely on interconnected systems.

For example:

Supplier software ↔ Platform API ↔ Warehouse system ↔ Logistics system

If a dominant platform intentionally prevents competing systems from interoperating, competition concerns may arise depending on the circumstances.

Interoperability can therefore become an important competitive parameter.

39. Data Portability

Suppliers may depend on their historical:

sales data;

customer information;

inventory records;

performance history.

If they cannot transfer this information when changing platforms, switching costs may increase.

High switching costs can reinforce platform dependence.

Again, however, data portability is not automatically a competition-law obligation in every situation.

40. Remedies

Where competition concerns are established, possible remedies include:

Structural remedies

divestiture;

separation of businesses;

removal of ownership links.

Behavioural remedies

non-discrimination;

access commitments;

licensing;

interoperability;

restrictions on exclusivity.

Algorithmic remedies

independent audits;

ranking safeguards;

algorithmic monitoring;

data segregation;

human review of automated exclusion decisions.

41. Compliance Framework for Businesses

Companies using machine-mediated supply chains should consider:

1. Competition-risk assessment

Review algorithms before deployment.

2. Data segregation

Prevent inappropriate sharing of competitor information.

3. Algorithm testing

Test for discriminatory or exclusionary outcomes.

4. Human oversight

Allow review of important supplier decisions.

5. Audit trails

Keep records of algorithmic decisions.

6. Contract review

Check automated contracts for:

exclusivity;

RPM;

territorial restrictions;

discriminatory terms.

7. Periodic review

Market conditions and algorithms can change over time.

42. Key Case-Law Lessons

43. Short Revision Points

Machine-mediated supply chains use AI and automated systems to govern commercial relationships.

Automation itself is not an antitrust violation.

Algorithms can create vertical foreclosure risks.

Digital platforms may become supply-chain bottlenecks.

Data can become an important competitive input.

Self-preferencing can occur through algorithmic rankings.

Automated exclusivity can create foreclosure concerns.

Smart contracts can implement restrictive arrangements automatically.

Common platforms can create information-exchange risks.

Dominance remains distinct from mere technological superiority.

Refusal to provide access is subject to jurisdiction-specific legal standards.

Efficiency benefits must be distinguished from exclusionary effects.

Algorithmic decisions should be auditable in high-risk settings.

Human oversight can assist competition-law compliance.

Post-deployment monitoring is important because algorithms evolve.

44. Conclusion

Machine-mediated supply chain governance can substantially improve efficiency, reliability, forecasting, logistics, and resource allocation. At the same time, it can increase competition-law risks where powerful undertakings use automated systems to control access to suppliers, customers, data, infrastructure, or distribution channels.

The principal areas of concern are:

vertical foreclosure;

self-preferencing;

exclusive dealing;

tying and bundling;

algorithmic discrimination;

information exchange;

refusal of access;

data concentration;

algorithmic coordination;

control over essential digital infrastructure.

The cases of Terminal Railroad, Microsoft, Aspen Skiing, Trinko, Bronner, IMS Health, Google Shopping, Slovak Telekom, Intel, and Hoffmann-La Roche provide foundational principles for analysing these modern problems.

The central principle is:

Competition law should regulate the competitive effects of machine-mediated supply-chain governance, not the mere fact that machines are being used. Efficient automation should remain possible, while automated systems that facilitate exclusion, coordination, discrimination, or foreclosure remain subject to ordinary competition-law scrutiny.