Competition Law and Industrial Data Monopolization Concerns

1. Introduction

Industrial data monopolization refers to a situation in which one or a small number of undertakings obtain substantial control over data that is important for production, innovation, distribution, maintenance, or competition in an industrial market.

Industrial data can include:

machine-generated data;

sensor and IoT data;

manufacturing-process data;

industrial equipment performance data;

predictive-maintenance information;

supply-chain data;

logistics information;

energy-consumption data;

production data;

engineering datasets;

industrial customer data;

technical specifications;

operational algorithms; and

data generated by connected products.

Industrial data can become a source of market power when competitors cannot obtain equivalent information or when users are locked into a particular data ecosystem.

The competition-law concern is not simply that a company possesses large amounts of data. The important questions are whether the data provide substantial and durable competitive advantages, whether access is reasonably available elsewhere, and whether the undertaking uses its position in an exclusionary or exploitative manner.

2. Why Industrial Data Matters to Competition

Modern industrial production increasingly depends upon data.

For example:

Industrial machine

↓

Sensors

↓

Operational data

↓

AI analytics

↓

Predictive maintenance

↓

Lower production costs

↓

Competitive advantage

A company possessing a superior industrial dataset may therefore improve:

productivity;

equipment reliability;

product quality;

energy efficiency;

logistics;

research and development.

This can create a data-driven competitive feedback loop.

3. What Is Industrial Data?

Industrial data can broadly be divided into several categories.

A. Machine-generated data

Examples:

temperature;

vibration;

pressure;

operating cycles;

machine utilisation.

B. Production data

Examples:

output volumes;

production times;

defect rates;

manufacturing costs.

C. Maintenance data

Examples:

failure history;

replacement schedules;

component performance.

D. Supply-chain data

Examples:

inventories;

transportation;

supplier performance;

demand forecasts.

E. Customer and usage data

Information generated by industrial products while they are being used by customers.

F. Research and engineering data

Examples:

testing results;

simulation datasets;

product-development information.

4. When Does Data Become a Competition Concern?

Data concentration ≠ automatically monopoly.

A firm may lawfully possess a large dataset because it:

invested in data collection;

developed superior technology;

served many customers;

created an innovative product.

Competition concerns become stronger where several conditions combine:

the undertaking has substantial market power;

the data are competitively significant;

rivals cannot reasonably reproduce or obtain equivalent data;

the data are important for competing effectively;

the undertaking restricts access or interoperability; and

the conduct produces or is capable of producing exclusionary effects.

5. Relevant Market Definition

Authorities may define the relevant market around:

industrial data itself;

data-access services;

industrial analytics;

predictive-maintenance services;

cloud-based industrial software;

connected machinery;

industrial IoT platforms;

aftermarket services.

A particularly important question is whether the data constitute a separate economic product or simply an input into another product.

6. Data as an Entry Barrier

A dominant undertaking may possess years of accumulated industrial data.

A new entrant may face:

insufficient training data;

inferior predictive models;

limited machine-performance information;

inability to benchmark equipment;

lack of historical failure information.

This can produce an economy-of-scale in data.

The incumbent obtains more data → develops better products → attracts more customers → obtains still more data.

7. Network Effects and Data Feedback

The competitive mechanism may be represented as:

More customers

↓

More industrial data

↓

Better analytics

↓

Better industrial services

↓

More customers

This is a data network effect.

It can make markets particularly difficult for new entrants because the incumbent's competitive advantage increases over time.

8. Case Law 1 — Google Shopping

Google Search (Shopping), Case AT.39740

Google Shopping provides an important precedent concerning the use of a dominant digital gateway to favour a related service.

Relevance to industrial data

Imagine a dominant industrial-data platform that:

collects equipment data;

operates an industrial analytics marketplace; and

also offers its own predictive-maintenance service.

The platform could theoretically use its control over data and rankings to favour its own analytics service.

The Google Shopping principles are therefore relevant to self-preferencing and leveraging through a data gateway.

9. Case Law 2 — Google Android

Google Android, Case AT.40099

The Android proceedings examined Google's contractual arrangements within the Android ecosystem.

Relevance

Industrial-data ecosystems may similarly combine:

hardware;

operating systems;

applications;

cloud services;

data;

analytics.

A dominant industrial-platform provider might impose contractual conditions that make it difficult for competing data or analytics services to operate.

The case demonstrates how ecosystem restrictions can reinforce platform power.

10. Case Law 3 — Microsoft

Microsoft Corp. v Commission, Case T-201/04

Microsoft is an important case concerning interoperability and access to technical information.

Relevance to industrial data

Industrial systems increasingly require interoperability between:

machinery;

software;

sensors;

cloud systems;

analytics platforms.

If a dominant equipment provider restricts access to necessary interfaces or technical information, independent analytics providers may be unable to compete effectively.

The Microsoft principles therefore provide an important framework for considering interoperability-based data foreclosure.

11. Case Law 4 — Magill

RTE and ITP v Commission, Joined Cases C-241/91 P and C-242/91 P

Magill is a foundational case concerning intellectual property, information and refusal to license.

Relevance

Industrial data may be protected through:

copyright;

database rights;

contractual confidentiality;

trade secrets.

The fact that information is legally protected does not automatically remove it from competition-law analysis.

Magill demonstrates the exceptional circumstances in which control over protected information can interact with abuse-of-dominance principles.

12. Case Law 5 — IMS Health

IMS Health GmbH & Co. OHG v NDC Health GmbH & Co. KG, Case C-418/01

IMS Health concerned access to a proprietary information structure.

Relevance

Industrial data providers may develop proprietary:

data architectures;

classifications;

technical datasets;

analytical systems.

Where competitors claim that access is indispensable, IMS Health provides an important framework for analysing the interaction between intellectual property and competition.

The case also confirms that competition law does not establish a general right to obtain competitors' proprietary information.

13. Case Law 6 — Bronner

Oscar Bronner GmbH & Co. KG v Mediaprint, Case C-7/97

Bronner concerned access to infrastructure controlled by a dominant undertaking.

Relevance

An industrial-data platform could control infrastructure such as:

machine-data interfaces;

industrial cloud platforms;

sensor networks;

proprietary data exchanges.

A competitor might argue that access is indispensable.

Bronner establishes a demanding framework for such refusal-to-deal claims.

14. Case Law 7 — Amazon Marketplace

European competition proceedings concerning Amazon provide an important example of competition concerns arising where a platform simultaneously operates:

an intermediary marketplace; and

competing commercial services.

Relevance to industrial data

A dominant industrial-data marketplace could similarly:

host independent analytics providers;

collect data about their activities; and

operate its own competing analytics service.

Access to commercially sensitive data could potentially strengthen the platform's competitive position.

15. Case Law 8 — Qualcomm

Qualcomm Inc. v European Commission

Qualcomm litigation concerning conditional payments provides useful guidance on exclusionary arrangements in technologically concentrated markets.

Relevance

Industrial-data ecosystems can involve powerful vertically integrated companies controlling:

chips;

sensors;

machinery;

software;

cloud services;

data analytics.

Commercial arrangements that restrict customers from using competing technologies may therefore raise competition questions.

16. Refusal to Provide Industrial Data

One of the most difficult issues is whether a dominant undertaking should be required to provide data to competitors.

Possible examples:

aircraft engine performance data;

industrial-machine data;

vehicle telematics;

energy-consumption data;

agricultural-machine data.

A competitor might argue:

"Without access to this data, we cannot provide an effective competing service."

But competition law generally does not impose a universal duty to share every valuable dataset.

Authorities must examine the relevant legal requirements, including:

indispensability;

duplication possibilities;

access feasibility;

elimination of competition;

objective justification.

17. Essential-Facility Theory

Industrial data may sometimes be argued to constitute an essential facility.

For example, suppose a manufacturer controls:

90% of a particular industrial machine market;

the machine-data interface;

the historical operational dataset.

Independent maintenance companies may argue that they cannot compete without access to the data.

The relevant question would be whether the legal requirements for intervention are satisfied.

The Bronner and IMS Health lines of authority are particularly relevant.

18. Data Portability

Industrial customers may want to move data between providers.

For example:

Machine manufacturer A

→ Customer's equipment data

→ Independent analytics provider B.

If the manufacturer prevents the customer from transferring the relevant data, switching costs can increase.

Data portability can therefore become a competition tool by reducing:

lock-in;

switching costs;

dependency on proprietary analytics.

19. Interoperability

Interoperability may be even more important than raw data ownership.

A company may technically allow customers to access their data but provide it in a format that competing software cannot easily process.

Potential competition issues can therefore arise from:

incompatible formats;

proprietary APIs;

restricted access credentials;

technical restrictions;

delayed access.

The Microsoft case provides an important conceptual basis for analysing such problems.

20. Industrial Aftermarkets

Industrial data monopolisation often becomes particularly important in aftermarkets.

Consider:

Original equipment

→ Maintenance

→ Spare parts

→ Diagnostics

→ Software updates

→ Data analytics

The manufacturer may be dominant in the aftermarket even if competition existed when the original equipment was purchased.

If only the manufacturer can access machine data, independent repair or maintenance providers may face competitive disadvantages.

21. Tying Industrial Data to Equipment

A dominant equipment manufacturer could potentially require customers to purchase:

Machine + proprietary analytics

or:

Machine + manufacturer-controlled maintenance

or:

Machine + proprietary cloud service

Such arrangements may raise tying or leveraging concerns where the applicable legal requirements are satisfied.

The Microsoft jurisprudence provides useful guidance on this category of conduct.

22. Self-Preferencing in Industrial Analytics

Suppose an industrial platform hosts:

100 third-party analytics firms; and

its own analytics service.

It controls the data and the ranking algorithm.

Potential competitive concerns could arise if it:

gives its own service privileged data access;

ranks its own analytics first;

gives competitors delayed data;

restricts competing APIs;

uses third-party performance data to improve its own product.

This is a classic platform-plus-competitor conflict.

23. Data Combination

A large industrial company may combine datasets from multiple markets.

For example:

Machine data

Customer data

Maintenance data

Supply-chain data

Energy data

=

Integrated industrial intelligence

This can create competitive advantages that smaller competitors cannot easily reproduce.

Data combination itself is not automatically unlawful, but its competitive consequences may become important where it strengthens already substantial market power.

24. Algorithmic Advantages

Industrial AI systems may rely heavily upon historical datasets.

For predictive maintenance:

More historical failures

→ Better prediction

→ More accurate maintenance

→ Better product

→ More customers

→ More data.

This can create a data-driven entry barrier.

Competition authorities may therefore need to examine whether data access determines the ability to compete in AI-enabled industrial markets.

25. Data Exclusivity Agreements

A dominant industrial-data provider might enter agreements requiring customers to provide:

exclusive machine data;

exclusive maintenance information;

exclusive analytics access.

Such arrangements may prevent rival analytics providers from obtaining sufficient data.

The competitive analysis would consider:

duration;

market coverage;

alternatives;

foreclosure;

efficiencies.

26. Data Licensing

A dominant firm may license industrial data to competitors.

Problems can arise if licensing conditions are:

discriminatory;

excessive;

restrictive;

exclusive;

technically burdensome.

For example:

Competitor A: receives complete real-time data.

Competitor B: receives delayed or incomplete data.

If there is no objective justification, discriminatory access could potentially become competition-relevant.

27. Industrial Data and Innovation

Data access can determine whether innovators can develop:

predictive-maintenance systems;

energy-optimisation software;

manufacturing AI;

robotics applications;

logistics solutions.

Therefore, data foreclosure may reduce innovation competition even where consumers do not immediately experience higher prices.

28. Industrial Data and SMEs

Small industrial-technology firms may be particularly dependent upon data controlled by large equipment manufacturers.

An SME may possess:

better analytics;

better AI;

better maintenance technology,

but lack access to the data necessary to demonstrate its product's effectiveness.

This can create a competitive paradox:

The company with the best technology may lose because the incumbent controls the data needed to use that technology.

Competition policy may therefore consider whether data access is functioning as an entry barrier.

29. Intellectual Property and Trade Secrets

Industrial data can be protected through:

patents;

copyright;

database rights;

trade secrets;

confidentiality agreements.

Competition law must balance:

Protection of investment and innovation

against

prevention of exclusionary conduct.

This is why cases such as Magill and IMS Health remain important.

30. Indian Competition-Law Perspective

Under the Competition Act, 2002, industrial data monopolisation may implicate several provisions.

Section 3

Potentially problematic arrangements could include:

exclusive data-sharing agreements;

restrictive licensing;

anti-competitive vertical agreements.

Section 4

Potential abuse of dominance could potentially involve:

denial of market access;

discriminatory access to data;

tying;

leveraging;

unfair conditions;

limiting technical development.

Sections 5 and 6

Acquisitions involving major industrial-data assets may require examination under India's merger-control framework where the statutory requirements are met.

31. Competition Concerns in Different Industries

32. Industrial Data and Cloud Concentration

Industrial data is increasingly stored and processed in cloud environments.

If a small number of cloud providers control:

storage;

processing;

AI infrastructure;

analytics;

industrial APIs,

companies may become dependent upon a limited number of technological ecosystems.

Potential competition concerns include:

cloud switching costs;

data portability;

interoperability;

bundling;

technical restrictions;

preferential treatment.

33. Industrial Data Marketplaces

An emerging model is the industrial data marketplace, where companies buy and sell datasets.

Such a marketplace could potentially become dominant.

Potential concerns include:

exclusion of rival data exchanges;

discriminatory access;

excessive commissions;

self-preferencing;

exclusive data arrangements;

acquisition of competing data providers.

34. Merger Control

Industrial-data concentration can also arise through mergers and acquisitions.

A large company might acquire:

an industrial IoT startup;

a predictive-maintenance company;

a specialised dataset;

a cloud analytics company;

an industrial software provider.

The acquisition may combine:

data + technology + distribution

and create competitive advantages that are difficult for rivals to replicate.

Authorities may therefore examine effects on:

current competition;

potential competition;

innovation;

data access.

35. Killer Acquisitions in Industrial Data

A small startup may have:

little revenue;

an innovative algorithm;

a unique industrial dataset.

A large incumbent may acquire it before it becomes a serious competitor.

Traditional revenue-based thresholds may not always reflect the strategic value of the target.

This makes potential competition an important consideration in data-intensive markets.

36. Possible Remedies

Where unlawful conduct is established, possible remedies may include:

A. Data access

Controlled access to relevant datasets.

B. Data portability

Customers can transfer their industrial data.

C. API access

Independent providers receive appropriate technical access.

D. Non-discrimination

Comparable competitors receive comparable access.

E. Interoperability

Different industrial systems can communicate.

F. Restrictions on exclusivity

Preventing unnecessarily broad exclusive-data arrangements.

G. Structural remedies

Exceptional separation of vertically integrated businesses.

37. Balancing Competition and Data Security

Not all data-access restrictions are anticompetitive.

Industrial data may contain:

trade secrets;

cybersecurity information;

personal information;

safety-critical information;

national-security-sensitive information.

A company may therefore have legitimate reasons to restrict access.

Competition analysis must distinguish:

legitimate security/confidentiality

from

unjustified exclusion.

38. Key Case-Law Summary

39. Major Competition-Law Questions for the Future

Industrial data monopolisation will increasingly require authorities to consider:

Who owns machine-generated data?

Can customers transfer their data?

Can independent repairers access diagnostic information?

Can competing AI systems train on relevant industrial data?

Can industrial platforms restrict third-party APIs?

Can manufacturers exclusively control aftermarket data?

Can a dominant platform use customer data to compete against its own customers?

When does data become an indispensable input?

Can data advantages constitute an entry barrier?

How should data-intensive mergers be assessed?

40. Conclusion

Industrial data is becoming a strategic input comparable in importance to capital, technology and physical infrastructure. Concentration of industrial data can generate legitimate efficiencies because large datasets may improve prediction, maintenance, safety and production efficiency. However, the same concentration can potentially create durable entry barriers where competitors cannot obtain comparable data.

The principal competition-law theories include refusal to deal, essential-facilities concerns, interoperability restrictions, self-preferencing, tying, exclusive data arrangements, discriminatory access, leveraging, vertical foreclosure and data-driven barriers to entry.

The cases of Google Shopping, Google Android, Microsoft, Magill, IMS Health, Bronner, Amazon Marketplace and Qualcomm provide established legal principles that can be applied to these emerging industrial-data problems.

The key distinction is therefore between legitimate data advantages resulting from investment and innovation and conduct through which a dominant undertaking uses control over industrial data to exclude competitors, restrict innovation, foreclose downstream markets or entrench its market position.