Competition Law and Strategic Data Asset Concentration

1. Introduction

Strategic data asset concentration refers to a situation in which one undertaking, platform, digital ecosystem, or group of undertakings accumulates control over large quantities of commercially valuable data and uses that concentration to obtain or reinforce market power.

In modern digital markets, data can function as a strategically important competitive asset because it may be used to:

• improve products and algorithms;
• reduce search and transaction costs;
• personalise services and advertising;
• train artificial-intelligence and machine-learning systems;
• identify customers and competitors;
• predict demand;
• optimise prices;
• develop credit or risk models;
• control access to digital ecosystems;
• create switching costs; and
• establish barriers to entry.

Competition law generally does not prohibit the mere possession of large datasets. The legal concern arises where concentration of data contributes to a substantial lessening of competition, facilitates exclusionary conduct, strengthens dominance, enables discriminatory access, or results from an anticompetitive merger or acquisition.

2. Meaning of Strategic Data Asset Concentration

Strategic data concentration can arise through several mechanisms.

A. Organic accumulation

A platform may accumulate data because millions of users continuously interact with it.

Examples include:

• search queries;
• purchase histories;
• location information;
• browsing behaviour;
• financial information;
• advertising interactions;
• technical telemetry; and
• transaction records.

B. Acquisition of data-rich companies

A dominant undertaking may acquire another company primarily or partly because of:

• its customer database;
• proprietary datasets;
• behavioural information;
• transaction histories;
• data-processing technology; or
• access to a particularly valuable user population.

C. Data aggregation across services

A company operating several complementary services may combine datasets from:

• search;
• social networking;
• payments;
• e-commerce;
• advertising;
• cloud computing;
• mapping; and
• mobile operating systems.

The competitive concern becomes stronger where rivals cannot obtain comparable data.

D. Exclusive data arrangements

A dominant undertaking may obtain exclusive access to a commercially significant dataset through contracts with:

• retailers;
• banks;
• advertisers;
• hospitals;
• manufacturers;
• payment providers;
• public authorities; or
• application developers.

E. Data generated through an essential digital interface

An undertaking controlling a major platform may be able to collect data from businesses that depend upon its infrastructure.

This creates a potential data chokepoint.

3. Why Data Can Become a Strategic Competitive Asset

Data has several characteristics that distinguish it from ordinary commercial inputs.

3.1 Data can improve algorithms

Large datasets may enable firms to develop:

• better recommendation systems;
• fraud detection;
• advertising targeting;
• search rankings;
• AI models;
• credit scoring; and
• demand forecasting.

3.2 Data can generate economies of scale

More users can generate more data, which can improve the service, attracting more users and producing still more data.

This can create a feedback loop:

Users → Data → Better service → More users → More data

This is sometimes described as a data-driven network effect.

3.3 Data may create entry barriers

A new entrant may possess good technology but lack:

• historical consumer data;
• behavioural observations;
• transaction records;
• training datasets; or
• sufficiently broad user interactions.

The resulting disadvantage can make entry more difficult.

3.4 Data can facilitate market intelligence

Access to detailed information concerning competitors, suppliers or customers can potentially reduce uncertainty and facilitate:

• coordinated conduct;
• discriminatory pricing;
• targeted exclusion;
• personalised offers; or
• algorithmic coordination.

4. Competition-Law Framework

Strategic data concentration can be examined principally through four areas.

A. Abuse of dominance

Where a firm has substantial market power, competition authorities may examine whether it uses control over data to:

• deny access to an important dataset;
• discriminate against competitors;
• impose unfair contractual conditions;
• tie data access to another service;
• engage in self-preferencing;
• impose exclusivity;
• exploit data-dependent customers; or
• prevent interoperability.

B. Merger control

Data can be relevant to merger analysis even where the acquired company's turnover is relatively small.

A transaction may raise concerns where it combines:

• two large datasets;
• complementary datasets;
• competing AI-training resources;
• advertising data;
• financial and consumer data; or
• datasets that competitors cannot readily replicate.

C. Cartel and coordination rules

Data concentration can facilitate coordination when competitors exchange strategically sensitive information concerning:

• prices;
• customers;
• capacity;
• inventory;
• costs; or
• future commercial strategies.

D. Essential-facility and access theories

In exceptional circumstances, control over an indispensable dataset or data infrastructure may raise questions concerning access obligations.

However, competition law generally does not automatically create a right of access to privately held data.

5. Important Case Laws

1. Google Search / Google Shopping — European Commission, 2017

Facts

The European Commission found that Google had abused its dominant position in general search by systematically giving prominent placement to its own comparison-shopping service while placing rival comparison-shopping services in less favourable positions.

Relevance to strategic data concentration

The case is important because Google's search infrastructure generated enormous quantities of user interactions and information.

The competition concern was not simply that Google possessed data. Rather, its control over an important digital gateway could be combined with its data, algorithms and ranking mechanisms to disadvantage competitors.

Principle

Control over a data-rich platform becomes particularly significant when the platform also controls the mechanism through which competing services reach consumers.

Significance

The case demonstrates the interaction between:

Data + algorithmic control + platform access + dominance.

6. Google Android — European Commission, 2018

Facts

The European Commission found that Google imposed contractual restrictions concerning Android devices, including requirements relating to the pre-installation and distribution of Google's applications.

Data significance

Mobile operating systems generate extensive information concerning:

• application use;
• searches;
• device activity;
• location;
• consumer behaviour; and
• interactions across Google's ecosystem.

Consequently, control of the operating-system layer can reinforce control over data-producing services.

Competition principle

Competition analysis in digital ecosystems may need to consider not only the immediate contractual restriction but also its effect on the accumulation of data and reinforcement of ecosystem advantages.

7. Facebook/WhatsApp — European Commission, 2014

Facts

The European Commission examined Facebook's acquisition of WhatsApp.

The transaction raised questions concerning the combination of Facebook's extensive user data with WhatsApp's user base and information.

The Commission ultimately cleared the transaction subject to the EU merger-control framework applicable at the time.

Importance

This is one of the most important examples of data-driven merger analysis.

The case illustrates that authorities may ask whether a transaction combines:

• large user bases;
• valuable datasets;
• advertising capabilities;
• consumer profiles; and
• complementary sources of information.

Principle

Data may constitute an important competitive parameter even where the transaction does not involve a conventional physical asset.

8. Facebook/WhatsApp — European Commission, 2017 enforcement concerning merger information

A later development concerning the Facebook/WhatsApp transaction demonstrated another important competition-law issue: the accuracy and completeness of information supplied during merger review.

The Commission fined Facebook for providing incorrect or misleading information concerning its ability to match Facebook and WhatsApp user accounts.

Competition relevance

This illustrates that data capabilities can be relevant not only to substantive merger assessment but also to the procedural integrity of merger control.

Principle

Where data integration is commercially significant, representations concerning the technical ability to combine datasets can themselves become relevant to competition enforcement.

9. Microsoft/LinkedIn — European Commission, 2016

Facts

The European Commission reviewed Microsoft's acquisition of LinkedIn.

LinkedIn possessed a substantial professional-network database, while Microsoft operated important productivity and enterprise software businesses.

The Commission examined potential competition concerns involving, among other matters, the use of LinkedIn data and access to professional-networking services.

Relevance

The transaction demonstrates the concept of complementary data assets.

Microsoft and LinkedIn possessed different but potentially valuable categories of information:

Microsoft → enterprise/productivity data

LinkedIn → professional-network and employment-related information

Combining complementary datasets can potentially create competitive advantages that neither company possesses independently.

Principle

Merger analysis may examine whether combining datasets creates a significant competitive advantage, particularly where competitors cannot readily reproduce the combined information.

10. Meta Platforms / Bundeskartellamt — Germany, 2019

Facts

Germany's Federal Cartel Office examined Facebook's practice of combining data collected from Facebook with data obtained from other Facebook services and third-party websites.

The Bundeskartellamt concluded that Facebook's terms concerning the combination of data constituted an abuse of its dominant position.

The matter subsequently generated important litigation before the German courts and the Court of Justice of the European Union.

Importance

This is one of the clearest examples of data concentration being examined directly through abuse-of-dominance law.

The case demonstrated that competition authorities may consider:

• the accumulation of data;
• cross-service data combination;
• consumer dependency;
• the importance of data to platform competition; and
• the relationship between privacy-related conditions and competition law.

Principle

A dominant platform's ability to impose extensive data-combination conditions may be examined as a potential exploitative or exclusionary abuse.

11. Google/DoubleClick — European Commission, 2008

Facts

The European Commission examined Google's acquisition of DoubleClick.

DoubleClick was an important provider of online advertising technology and possessed significant expertise and information concerning digital advertising.

Data significance

The transaction illustrated the importance of combining:

• search data;
• advertising information;
• publisher relationships;
• advertiser relationships; and
• behavioural information.

Competition principle

A merger can raise competition questions where combining data resources strengthens an undertaking's position in an adjacent or vertically related market.

The case is particularly relevant to the modern ad-tech data concentration debate.

12. Apple/Shazam — European Commission, 2018

Facts

Apple proposed acquiring Shazam, a music-recognition service with a large user base and significant music-related behavioural information.

The European Commission investigated whether the transaction could adversely affect competition in music streaming.

The transaction was ultimately cleared.

Relevance

The case demonstrates that authorities may examine whether an acquisition of a data-rich application could:

• provide commercially valuable information;
• disadvantage competing services;
• facilitate customer targeting; or
• reinforce an ecosystem.

Principle

Even relatively specialised datasets can have strategic importance when they relate to a commercially significant user population.

13. Amazon/eBay and Data-Driven Platform Power

The competition-law issues surrounding major online marketplaces illustrate another dimension of strategic data concentration.

Marketplace operators can potentially obtain information about:

• seller performance;
• consumer demand;
• prices;
• product turnover;
• inventory;
• conversion rates; and
• customer preferences.

Where the platform simultaneously competes with sellers using the platform, concerns may arise regarding dual-role data advantages.

The theoretical competition problem is:

Platform operator → collects seller data → observes commercially sensitive information → competes against sellers.

This is particularly relevant to self-preferencing and leveraging theories.

14. Strategic Data Concentration and Merger Control

Data-driven merger analysis should consider several questions.

Question 1: Is the data genuinely scarce?

If identical or equivalent information is freely available, concentration may have limited competitive significance.

Question 2: Is the dataset replicable?

A dataset may be valuable because competitors cannot easily reproduce its:

• scale;
• historical depth;
• accuracy;
• variety; or
• frequency of updating.

Question 3: Is the dataset unique?

Unique transaction or behavioural information can create a stronger competitive advantage.

Question 4: Can the datasets be combined?

Two datasets may become more valuable when integrated.

For example:

Consumer identity data + purchasing data + financial data + location data

may provide significantly greater commercial insight than any dataset individually.

Question 5: Are competitors dependent upon the dataset?

Dependence strengthens the potential competitive significance of the asset.

15. Data Concentration and Barriers to Entry

A dominant data holder can potentially create a barrier to entry through a data scale advantage.

A simplified model is:

Incumbent

Large user base
↓
Large dataset
↓
Better algorithms
↓
Better product
↓
More users
↓
More data

A new entrant may face the reverse problem:

New entrant

Small user base
↓
Limited data
↓
Less accurate algorithms
↓
Less attractive service
↓
Difficulty attracting users
↓
Limited data accumulation

This can produce a data feedback loop.

However, the existence of such a loop does not by itself establish an infringement. Authorities must generally demonstrate its actual or likely competitive significance.

16. Data Portability as a Competition Remedy

One possible remedy is data portability.

A consumer may be allowed to transfer relevant information from one provider to another.

This can reduce:

• switching costs;
• lock-in;
• customer acquisition barriers; and
• informational advantages of incumbents.

Competition law can therefore interact with data-protection and digital-regulation frameworks.

17. Data Access Remedies

In exceptional cases, competition authorities may consider:

• mandatory data sharing;
• interoperability;
• API access;
• non-discriminatory access;
• data portability;
• licensing;
• separation of datasets; or
• restrictions on data combination.

However, mandatory access can also create:

• privacy risks;
• cybersecurity risks;
• free-riding;
• incentives to reduce investment in data collection; and
• difficulties concerning data quality and governance.

Therefore, access remedies require careful calibration.

18. Data Concentration and Self-Preferencing

A particularly important modern problem arises where a platform acts simultaneously as:

1. infrastructure provider;
2. data collector;
3. marketplace operator; and
4. competitor.

For example:

Marketplace → collects seller data → analyses sales → identifies successful products → launches competing products

The competition question is whether the platform is using its privileged informational position to obtain an unfair competitive advantage.

This makes data governance and structural platform governance closely connected.

19. Data Concentration and Artificial Intelligence

AI significantly increases the strategic importance of data.

Modern AI competition can depend upon access to:

• training datasets;
• labelled datasets;
• proprietary industrial data;
• consumer interaction data;
• scientific datasets;
• copyrighted databases;
• real-time behavioural information; and
• feedback generated by users.

Consequently, acquisition of a company possessing a unique dataset can potentially affect future competition even if the acquired company's current revenues are modest.

The central concern becomes:

Could control over the dataset prevent future competitors from developing viable competing AI systems?

20. Data Concentration and Algorithmic Competition

Large datasets may allow a dominant undertaking to develop sophisticated algorithms for:

• pricing;
• recommendations;
• advertising;
• credit assessment;
• logistics;
• fraud prevention;
• product ranking; and
• consumer segmentation.

This can create an important distinction:

Traditional market power

Market share → market power

Digital market power

Data → algorithms → users → network effects → data

Therefore, traditional market-share analysis may need to be supplemented by analysis of data access and ecosystem effects.

21. Data Concentration and Consumer Exploitation

Data concentration may also produce exploitative effects.

Consumers may effectively "pay" for digital services with:

• personal information;
• behavioural information;
• attention;
• location data; or
• permission to combine information across services.

Competition authorities may therefore examine whether data-related contractual conditions are particularly burdensome when imposed by a dominant undertaking.

The Meta/Bundeskartellamt litigation is particularly relevant to this issue.

22. Data Concentration and Privacy

Competition law and privacy law pursue different objectives but can overlap.

Privacy law asks:

Is personal data processed lawfully and fairly?

Competition law asks:

Does the undertaking's conduct distort or restrict competition?

The same conduct may therefore potentially have consequences under both regimes.

For example:

Cross-service data combination

may simultaneously raise:

• privacy concerns;
• consumer-protection concerns;
• dominance concerns; and
• merger concerns.

23. Strategic Data Asset Concentration: Major Competition Risks

24. Legal Tests for Competition Authorities

When analysing strategic data concentration, an authority may consider:

Step 1 — Define the relevant market

Determine whether the relevant market concerns:

• search;
• online advertising;
• social networking;
• e-commerce;
• cloud computing;
• AI services;
• financial technology;
• data analytics; or another product/service.

Step 2 — Establish market power

Relevant indicators can include:

• market share;
• network effects;
• switching costs;
• barriers to entry;
• control over infrastructure;
• ecosystem position;
• access to data; and
• dependence of business users.

Step 3 — Identify the data asset

The authority should determine:

• what data is controlled;
• who generates it;
• whether it is proprietary;
• whether it is unique;
• whether it is replicable;
• whether it is commercially valuable.

Step 4 — Examine exclusionary conduct

Possible conduct includes:

• refusal to provide access;
• discriminatory access;
• exclusivity;
• tying;
• bundling;
• self-preferencing;
• interoperability restrictions;
• data combination;
• exploitative terms.

Step 5 — Examine competitive effects

The authority should consider whether conduct:

• excludes competitors;
• increases entry barriers;
• reduces innovation;
• increases switching costs;
• restricts consumer choice;
• facilitates coordination; or
• entrenches an existing dominant position.

Step 6 — Consider efficiencies

Data concentration can produce legitimate efficiencies such as:

• improved fraud detection;
• better product recommendations;
• reduced transaction costs;
• improved cybersecurity;
• improved logistics; and
• better AI performance.

Competition analysis should distinguish these benefits from conduct that unnecessarily excludes competitors.

25. Six Core Cases at a Glance

26. Emerging Concept: Data as a Strategic Chokepoint

The most important development is the transformation of data from an ordinary business resource into a strategic competitive chokepoint.

A strategic chokepoint exists where control over a particular resource gives an undertaking the ability to influence the competitive conditions under which other firms operate.

The structure can be represented as:

Data Asset
↓
Information Advantage
↓
Algorithmic Advantage
↓
Better Products / Targeting
↓
More Users
↓
More Data
↓
Greater Market Power

Competition law becomes concerned when this cycle becomes sufficiently self-reinforcing that competitors cannot effectively constrain the incumbent.

27. Important Limitations

Strategic data concentration should not automatically be treated as unlawful.

Large datasets can result from legitimate competition and investment.

Furthermore:

• data may depreciate rapidly;
• consumers may use multiple platforms;
• datasets may be replicable;
• synthetic data may substitute for some information;
• publicly available data may reduce scarcity;
• privacy restrictions may limit data use;
• competitors may obtain equivalent data through alternative sources.

Therefore, data volume alone is not a reliable measure of market power.

The critical question is whether control over the data creates a material and durable competitive advantage and whether the undertaking engages in conduct that unlawfully exploits or reinforces that advantage.

28. Conclusion

Strategic Data Asset Concentration represents an increasingly important intersection of competition law, digital markets, privacy, artificial intelligence and merger control.

The principal competition-law concern is not simply that a company possesses enormous amounts of data. The deeper issue is whether control over unique or difficult-to-replicate data allows an undertaking to acquire, maintain or reinforce market power, particularly through exclusionary conduct or anticompetitive acquisitions.

The major legal themes emerging from cases such as Google Shopping, Google Android, Facebook/WhatsApp, Microsoft/LinkedIn, Meta/Bundeskartellamt, Google/DoubleClick and Apple/Shazam include:

1. data can constitute an important competitive asset;
2. data advantages can reinforce network effects;
3. data-rich mergers can require careful scrutiny;
4. cross-service data combination can have dominance implications;
5. platform control and data control can mutually reinforce each other;
6. unique datasets can contribute to entry barriers;
7. data access and interoperability may become potential remedies; and
8. AI is increasing the strategic importance of proprietary and high-quality datasets.

Ultimately, the modern competition-law problem can be expressed as:

Who controls the data, who can access it, how difficult it is to reproduce, and whether that control can be used to restrict the competitive process?