Competition Law and Industrial Data Marketplaces and Antitrust

1. Introduction

Industrial data marketplaces are digital platforms through which industrial data is collected, aggregated, exchanged, licensed, accessed, or monetised among manufacturers, suppliers, technology providers, logistics companies, energy companies, research institutions, and other businesses.

Examples include marketplaces for:

manufacturing and machine-performance data;

industrial IoT data;

supply-chain and logistics information;

energy-consumption data;

predictive-maintenance datasets;

agricultural and industrial sensor data;

engineering and design information;

emissions and environmental data;

industrial benchmarks and quality data;

AI-training datasets;

digital-twin information; and

data generated by connected machinery and industrial platforms.

Industrial data marketplaces can improve competition by reducing information asymmetries and allowing smaller firms to access valuable datasets. At the same time, they can create new forms of market power because data can be difficult to replicate, switching costs can be high, and the platform controlling the marketplace may simultaneously participate as a buyer, seller, data processor, or downstream competitor.

Competition law therefore has to address both traditional antitrust problems and data-specific mechanisms of market power.

2. Why Industrial Data Marketplaces Raise Competition Issues

The central competition question is not simply whether a company possesses a large amount of industrial data.

Data ownership or possession does not automatically establish dominance.

The relevant questions are:

What is the relevant product and geographic market?

How important is the particular dataset to competition?

Can competitors obtain equivalent data elsewhere?

Can the data be replicated?

Does the platform control access to an important data ecosystem?

Does the platform discriminate between competing users?

Does it use marketplace data to compete against marketplace participants?

Are exclusivity arrangements preventing rival data marketplaces from developing?

Are mergers eliminating potential competitors or combining strategically important datasets?

Does collaboration through the marketplace facilitate coordination between competitors?

3. Relevant Market Definition

Several markets may exist simultaneously.

A. Industrial data supply market

A platform may supply:

machine data;

production data;

sensor information;

engineering datasets;

logistics information;

industrial benchmarks.

B. Industrial data marketplace/intermediation market

The platform may provide the infrastructure through which data buyers and sellers transact.

C. Data analytics market

Data may be transformed into:

predictive models;

industrial analytics;

maintenance forecasts;

optimisation services;

benchmarking tools.

D. Downstream industrial markets

Data may ultimately affect competition in:

manufacturing;

automotive;

energy;

chemicals;

logistics;

pharmaceuticals;

agriculture;

construction.

Thus, an antitrust investigation may need to analyse several vertically connected markets.

4. Data as a Source of Market Power

Industrial data can contribute to market power through several mechanisms.

4.1 Data scale

A platform serving thousands of industrial machines may receive vastly more information than a new entrant.

The resulting dataset can improve:

algorithms;

predictive accuracy;

maintenance recommendations;

benchmarking;

AI models.

This may create a feedback loop:

more users → more data → better services → more users → still more data.

4.2 Data variety

The value of industrial data often depends on combining different categories.

For example:

machine data + maintenance data + energy data + supply-chain data

may be significantly more valuable than any one dataset individually.

A dominant platform that controls several complementary datasets may therefore possess an ecosystem advantage.

4.3 Data velocity

Real-time or continuously generated data may be especially valuable.

A competitor cannot necessarily recreate five years of historical sensor information simply by entering the market today.

4.4 Data quality

Quantity alone does not determine competitive significance.

A smaller dataset may be more valuable if it contains:

highly accurate measurements;

proprietary machine information;

rare failure events;

detailed maintenance histories;

high-frequency sensor observations.

5. Network Effects

Industrial data marketplaces can exhibit direct and indirect network effects.

A larger number of data suppliers attracts more buyers.

More buyers, in turn, attract additional data suppliers.

This creates:

Data suppliers → marketplace → data buyers → more suppliers

A platform may consequently reach a tipping point where competitors find it difficult to attract enough participants to compete effectively.

6. Multi-Homing and Single-Homing

An important competition-law issue is whether industrial customers can participate in several marketplaces.

Multi-homing

A manufacturer can supply data to:

Marketplace A;

Marketplace B; and

Marketplace C.

This can constrain market power.

Single-homing

If technical, contractual, financial, or interoperability barriers effectively require customers to use one marketplace, market power may become stronger.

Competition authorities therefore examine:

API compatibility;

data-portability arrangements;

contractual restrictions;

technical integration;

switching costs;

migration costs.

7. Exclusive Data Agreements

A dominant industrial-data platform might enter agreements under which manufacturers agree to provide data exclusively to that platform.

Such agreements can create competition concerns when they substantially foreclose rival marketplaces.

Potential concerns include:

exclusion of new entrants;

raising rivals' data-acquisition costs;

preventing multi-homing;

locking customers into one ecosystem;

depriving competing AI providers of necessary training data.

The legality depends upon factors such as:

duration;

coverage;

market power;

foreclosure;

availability of alternative data sources;

efficiencies.

8. Refusal to Provide Data

One of the most difficult issues is whether a dominant firm can be required to provide access to industrial data.

Competition law generally does not create a universal obligation to share every commercially valuable dataset.

However, refusal may become problematic in exceptional circumstances where:

the firm is dominant;

the data/resource is indispensable;

duplication is practically or economically impossible;

refusal eliminates effective competition;

access is objectively necessary for downstream competition;

there is no legitimate justification.

This connects industrial data marketplaces with the essential-facilities and refusal-to-deal doctrines.

9. Self-Preferencing

Suppose a dominant industrial data marketplace operates a platform on which independent analytics providers compete.

The platform also offers its own predictive-maintenance service.

It could potentially use its control over marketplace data to:

obtain competitors' commercially sensitive information;

identify successful products;

copy their strategies;

rank its own service more prominently;

restrict competitors' access to data;

impose inferior API access on rivals.

This creates a possible self-preferencing problem.

The key question is whether the platform is using an infrastructural position to advantage its own downstream business.

10. Use of Competitors' Data

This is particularly important in industrial ecosystems.

Imagine that:

ten manufacturers use a data marketplace;

each provides operational information;

the marketplace operator observes their production patterns;

the operator also sells competing industrial equipment.

The platform might gain information about:

production volumes;

capacity utilisation;

procurement needs;

inventory;

equipment failures;

future investment;

pricing strategies.

Using such competitively sensitive information against marketplace participants may create serious antitrust concerns.

11. Information Exchange and Cartel Risks

Industrial data marketplaces can unintentionally become mechanisms for competitor coordination.

Suppose competing manufacturers upload:

current prices;

future prices;

production capacity;

inventories;

customer allocations;

output forecasts.

If the platform makes this information available to competitors, the marketplace may reduce uncertainty between competitors.

This can facilitate:

price coordination;

output coordination;

market allocation;

bid coordination;

capacity coordination.

Therefore, data governance is also a cartel-compliance issue.

12. Hub-and-Spoke Concerns

A marketplace may act as a hub, while competing businesses constitute the spokes.

The risk becomes:

Competitor A → platform → Competitor B → platform → Competitor C

If the platform facilitates the transmission of strategically sensitive information between competitors, traditional cartel principles can become relevant.

Important factors include:

whether the information is competitively sensitive;

whether disclosure is necessary for the service;

whether competitors know information is being shared;

whether the platform intentionally facilitates coordination;

whether safeguards prevent unnecessary disclosure.

13. Algorithmic Coordination

Industrial marketplaces increasingly use algorithms to:

recommend prices;

match buyers and sellers;

allocate capacity;

forecast demand;

optimise procurement;

recommend suppliers.

Algorithms can potentially make coordination easier by:

increasing price transparency;

rapidly responding to rivals;

reducing uncertainty;

implementing coordinated strategies automatically.

The fact that coordination is algorithmically implemented does not itself eliminate competition-law responsibility.

14. Tying and Bundling

A dominant industrial platform may offer:

Data marketplace + cloud storage + analytics + AI + equipment monitoring

and require customers to purchase several services together.

Competition concerns may arise if a dominant platform uses one market to extend power into another.

For example:

industrial data access may be conditioned on using the platform's cloud;

marketplace participation may require its analytics product;

access to APIs may require purchasing its hardware;

data portability may be restricted unless the customer uses its software.

The relevant legal analysis generally examines dominance, foreclosure, coercion, and effects.

15. Interoperability and API Restrictions

Interoperability is particularly important for industrial data.

Industrial companies may use:

sensors from one manufacturer;

cloud services from another;

analytics software from a third party;

machinery from several suppliers.

If a dominant marketplace restricts API access, it can make it technically difficult for competitors to participate.

Potential concerns include:

discriminatory API access;

technical degradation;

proprietary protocols;

refusal to provide interoperability;

unreasonable certification requirements.

16. Data Portability and Switching Costs

A customer may have accumulated years of:

maintenance records;

machine histories;

production information;

sensor observations;

AI models.

Moving to another marketplace may require expensive data conversion and system integration.

High switching costs can therefore reinforce market power.

A competition analysis may examine:

financial switching costs + technical switching costs + contractual switching costs + loss of accumulated data.

17. Predatory Pricing and Free Data Services

A large platform might offer industrial-data marketplace services below cost to acquire users.

For example:

Free data storage + free analytics + free marketplace access

may initially appear beneficial.

However, if the platform has substantial resources and later uses the installed customer base to exclude competitors, competition authorities may examine whether the conduct constitutes exclusionary pricing or another form of anticompetitive strategy.

Free services are not automatically unlawful.

18. Merger Control and Data Concentration

Mergers involving industrial data may create competition problems even when traditional turnover measures underestimate their significance.

Consider:

Industrial IoT platform + industrial analytics company

The transaction could combine:

machine data;

customer relationships;

cloud infrastructure;

AI models;

analytics capabilities.

Potential theories include:

Horizontal effects

Elimination of a competing data marketplace.

Vertical effects

Control over an upstream data resource could disadvantage downstream analytics competitors.

Conglomerate effects

Combining several complementary industrial datasets could reinforce ecosystem dominance.

Potential competition

A smaller data company might represent a future competitive constraint even if its current market share is low.

19. Case Law

1. IMS Health GmbH & Co. OHG v NDC Health GmbH & Co. KG

The Court of Justice of the European Union's decision in IMS Health is highly relevant to data-intensive markets.

The case concerned access to a commercially valuable information structure and the circumstances in which refusal to license an intellectual-property-related resource could constitute abuse.

The Court identified exceptional conditions relevant to compulsory access, including circumstances involving indispensability and elimination of competition.

Relevance to industrial data marketplaces

It provides an important framework for asking:

Is the dataset genuinely indispensable?

Can competitors realistically recreate it?

Would refusal eliminate effective competition?

Is access objectively necessary?

It supports a cautious approach to treating valuable data as an essential facility.

2. Bronner v Mediaprint

Oscar Bronner GmbH & Co. KG v Mediaprint Zeitungs und Zeitschriftenverlag GmbH & Co. KG is another major refusal-to-deal case.

The Court applied a demanding standard before requiring a dominant undertaking to provide access to infrastructure.

Relevance

Industrial-data infrastructure may similarly involve substantial investment.

The case helps establish that:

Mere usefulness of another firm's infrastructure does not automatically create a competition-law duty to provide access.

Indispensability and elimination of competition are particularly important considerations.

3. Commercial Solvents v Commission

In Commercial Solvents Corp. v Commission, the EU courts addressed refusal to supply by a dominant undertaking where the upstream input was important to downstream competition.

Relevance

An industrial-data marketplace could occupy a similar vertical position.

For example:

upstream industrial data → analytics → downstream industrial services.

If a dominant data provider cuts off an important input in order to disadvantage a downstream competitor, Commercial Solvents provides a foundational framework for analysing the conduct.

4. Microsoft Corp. v Commission

The Microsoft litigation is particularly important for technology markets involving interoperability.

The European Commission found competition concerns concerning Microsoft's refusal to provide interoperability information to competing work-group server products.

Relevance

Industrial data ecosystems often depend upon interoperability between:

machinery;

sensors;

cloud systems;

APIs;

analytics platforms.

The case therefore provides an important analogy for analysing situations in which technical restrictions prevent competing services from interoperating with a dominant platform.

5. Google Shopping

The Google Shopping litigation concerns the treatment of Google's own comparison-shopping service within its general search results.

The case is important to the broader debate concerning platform self-preferencing.

Relevance

An industrial-data marketplace might:

operate the marketplace;

host third-party analytics providers; and

offer its own analytics service.

If it systematically favours its own downstream service through control over marketplace access, ranking, visibility, or data, the principles arising from platform-discrimination cases become relevant.

The precise legal assessment depends on the market structure and conduct involved.

6. Google Android

The Google Android case concerned Google's practices surrounding the Android mobile ecosystem, including tying and restrictions affecting competing search and browser services.

Relevance

Industrial platforms increasingly operate ecosystems containing:

operating systems;

cloud infrastructure;

data services;

applications;

analytics.

The case illustrates how contractual or technical restrictions within an ecosystem can potentially extend market power from one layer into another.

7. Magill

RTE and ITP v Commission (Magill) is a foundational case concerning access to information protected by intellectual-property rights.

The Court recognised exceptional circumstances in which refusal to license information could constitute abuse.

Relevance

Industrial datasets may be protected through combinations of:

copyright;

database rights;

trade secrets;

contractual restrictions.

Magill demonstrates that intellectual-property protection does not automatically immunise conduct from competition law, while also emphasising the exceptional nature of compulsory access.

8. United Brands v Commission

In United Brands v Commission, the Court examined abuse of dominance and exclusionary conduct involving a powerful market position.

Relevance

Industrial-data marketplaces may similarly involve a dominant undertaking capable of imposing commercial conditions on customers and trading partners.

The case is useful for understanding broader principles concerning:

dominance;

abusive conduct;

discriminatory conditions;

exclusionary effects.

9. Hoffmann-La Roche v Commission

Hoffmann-La Roche v Commission is a foundational EU case concerning abusive exclusionary practices by dominant firms, particularly loyalty-inducing arrangements.

Relevance

If an industrial data marketplace requires customers to provide data exclusively to it, or provides incentives designed to prevent customers from using competing marketplaces, the principles concerning exclusionary loyalty arrangements may become relevant.

10. Airtours v Commission

Airtours v Commission is principally important for merger analysis and the assessment of coordinated effects.

Relevance

Industrial data marketplaces can increase transparency among competitors.

A merger may therefore be relevant not only because of unilateral market power but also because increased information transparency could affect the conditions for coordination.

20. Indian Competition Law Perspective

In India, industrial-data marketplace issues can potentially engage the Competition Act, 2002.

Section 3 — Anti-competitive agreements

Section 3 can become relevant where agreements between businesses:

restrict data access;

impose exclusivity;

facilitate information exchange;

allocate customers;

coordinate prices;

restrict technological interoperability.

Horizontal agreements deserve particular scrutiny where competitors exchange competitively sensitive data.

Section 4 — Abuse of dominant position

Section 4 becomes relevant where a dominant industrial-data platform engages in conduct such as:

unfair or discriminatory conditions;

unfair pricing;

denial of market access;

leveraging dominance;

tying or bundling;

exclusionary arrangements.

The important point is that possession of industrial data alone does not establish dominance.

Market power must be assessed in the relevant market.

Sections 5 and 6 — Combinations

Industrial-data mergers can require competition scrutiny where the transaction crosses the applicable statutory thresholds and falls within the combination framework.

The analysis can include:

data concentration;

loss of potential competitors;

vertical foreclosure;

ecosystem effects;

access to essential datasets;

network effects.

21. Important Competition Indicators

Competition authorities examining an industrial-data marketplace may consider:

22. Competition Risks Across the Industrial Data Value Chain

The problem can be visualised as:

Data generation
↓
Industrial equipment / sensors
↓
Data aggregation
↓
Industrial data marketplace
↓
Data analytics / AI
↓
Predictive services
↓
Industrial products and services

Competition concerns can arise at every layer.

A firm dominant at one layer may attempt to leverage its position into another.

23. Pro-Competitive Benefits

Industrial-data marketplaces can produce substantial efficiencies.

They may:

reduce information asymmetry;

improve supply-chain efficiency;

reduce equipment downtime;

facilitate predictive maintenance;

improve industrial safety;

allow SMEs to access data;

promote AI innovation;

reduce transaction costs;

improve resource allocation;

enable new entrants to develop data-driven services.

Competition law should therefore avoid treating data concentration itself as automatically unlawful.

The proper question is whether particular conduct substantially restricts competition or creates an abusive exercise of market power.

24. Possible Remedies

Where competition concerns are established, possible remedies may include:

Structural remedies

divestiture in exceptional merger cases;

separation of competing business units.

Behavioural remedies

non-discriminatory access;

interoperability obligations;

API access;

data portability;

restrictions on exclusivity;

information-firewall requirements;

transparency obligations.

Data-related remedies

controlled data access;

standardised data formats;

portability mechanisms;

restrictions on use of competitor data.

Merger remedies

divestiture of overlapping assets;

access commitments;

licensing;

interoperability commitments.

25. Key Legal Distinction

A crucial distinction is:

Data concentration ≠ market dominance ≠ abuse.

These are three separate questions.

First:

Does a firm possess a large or strategically valuable dataset?

Second:

Does that dataset contribute to substantial market power?

Third:

Has the firm used that market power in an anticompetitive manner?

Only after these questions are separated can a sound antitrust analysis be conducted.

26. Analytical Framework

For an industrial-data marketplace, a competition-law investigation can follow this sequence:

Step 1 — Define the relevant market

Identify the specific data, marketplace, analytics, or downstream industrial service.

Step 2 — Assess market power

Examine market share, data scale, network effects, switching costs and entry barriers.

Step 3 — Assess data significance

Determine whether the data is unique, replicable, substitutable and commercially indispensable.

Step 4 — Examine platform governance

Study:

access rules;

APIs;

ranking;

pricing;

data portability;

technical standards.

Step 5 — Examine contractual restrictions

Look for:

exclusivity;

loyalty arrangements;

tying;

bundling;

MFN-type clauses;

restrictions on multi-homing.

Step 6 — Examine information flows

Determine whether competitor information is being exchanged or exploited.

Step 7 — Examine vertical effects

Consider whether the platform is favouring its own downstream operations.

Step 8 — Examine mergers

Assess whether a transaction combines datasets or eliminates an emerging competitor.

Step 9 — Consider efficiencies

Evaluate genuine benefits such as:

better security;

interoperability;

reduced costs;

innovation;

improved data quality.

Step 10 — Select proportionate remedies

Remedies should address the identified competitive harm without unnecessarily eliminating legitimate commercial incentives to collect and invest in data.

27. Conclusion

Industrial data marketplaces represent a significant intersection between competition law, digital platforms, industrial organisation, intellectual property, interoperability and data governance.

The most important competition concerns include:

data-driven market power;

network effects and tipping;

exclusive data arrangements;

refusal to provide strategically important data;

self-preferencing;

use of competitors' data;

API and interoperability restrictions;

information-exchange and cartel risks;

algorithmic coordination;

tying and bundling;

high switching costs;

data-driven mergers and ecosystem concentration.

The leading cases—IMS Health, Bronner, Commercial Solvents, Microsoft, Google Shopping, Google Android, Magill, United Brands, Hoffmann-La Roche, and Airtours—provide different doctrinal building blocks for analysing these problems.

The central principle is that competition law should distinguish between legitimate data-driven scale and exclusionary exploitation of data control. A large industrial dataset can be an important competitive advantage without itself being unlawful; the antitrust question becomes considerably more significant when control over that data is combined with exclusionary contracts, discriminatory access, interoperability restrictions, competitor-data exploitation, coordination mechanisms, or strategic acquisitions.