Competition Law and Competition Implications of Reflexive Digital Markets

1. Introduction

Reflexive digital markets are digital markets in which the behaviour of market participants is continuously observed, measured, processed and fed back into the market itself.

In a traditional market, demand and supply influence prices and output. In a reflexive digital market, the process is more dynamic:

User behaviour → Data collection → Algorithmic analysis → Market decision → Changed user behaviour → New data → Further algorithmic adjustment

The market therefore becomes self-referential or reflexive.

Examples include:

• online marketplaces;
• search engines;
• social-media platforms;
• digital advertising;
• app stores;
• ride-hailing platforms;
• online travel platforms;
• financial technology platforms;
• AI-based marketplaces; and
• algorithmic pricing systems.

The competition-law significance is that a dominant digital undertaking may not merely respond to market conditions—it may possess the ability to observe, predict, influence and modify those conditions.

Importantly, “reflexive digital market” is an analytical concept, not a separate statutory competition-law offence. Existing doctrines such as abuse of dominance, monopolization, exclusionary conduct, tying, refusal to deal, discriminatory access and coordinated conduct are used to analyse the relevant behaviour.

2. Meaning of Reflexive Digital Markets

A reflexive digital market is a market where:

1. market participants generate digital data;
2. platforms continuously collect that data;
3. algorithms analyse the data;
4. the resulting information affects market decisions;
5. those decisions change participant behaviour; and
6. the changed behaviour produces new data.

Simple example

Suppose an online marketplace observes that consumers increasingly search for a particular product.

The platform:

collects searches → predicts demand → changes rankings → promotes selected products → consumers change purchasing behaviour → platform collects new data.

The platform is therefore not merely observing the market. Its decisions can feed back into the market.

3. Traditional Market vs Reflexive Digital Market

4. Core Features

A. Continuous Data Collection

Digital platforms can continuously collect:

• searches;
• clicks;
• purchases;
• location information;
• browsing behaviour;
• reviews;
• transaction information;
• advertising responses.

This gives platforms an unusually detailed view of market behaviour.

B. Algorithmic Feedback

Algorithms transform data into decisions.

For example:

Data → Prediction → Ranking → Consumer response → New data

The cycle may operate thousands or millions of times.

C. Behavioural Influence

Platforms can influence behaviour through:

• recommendations;
• rankings;
• personalised advertising;
• notifications;
• default settings;
• search results;
• discounts.

Consequently, a platform can sometimes move from being a market observer to a market shaper.

5. Why Reflexivity Matters to Competition Law

Competition law traditionally asks:

How does a firm's conduct affect competition?

In a reflexive market, another question becomes important:

How does the firm's conduct change the market conditions that the firm subsequently observes and uses?

This can create a feedback loop:

Market power → better data → better prediction → stronger service → more users → more market power

This may create substantial barriers for new entrants.

6. Data Feedback Loops

One of the most important competition implications is the data feedback loop.

Example

A search platform has:

1. millions of users;
2. enormous search data;
3. sophisticated algorithms;
4. better predictions;
5. more accurate recommendations.

Better recommendations attract more users.

More users generate more data.

More data improves the algorithms.

Thus:

Scale → Data → Algorithmic improvement → More scale

This may create a self-reinforcing competitive advantage.

7. Network Effects

Reflexive markets frequently involve network effects.

For example:

More users → More sellers → More products → More consumers → More sellers

A platform with strong network effects may become difficult to challenge.

When combined with data feedback, the effect can become even stronger:

Network effects + data advantage + algorithmic learning = potentially powerful entry barriers

8. Algorithmic Pricing

Algorithms may continuously analyse:

• competitor prices;
• demand;
• inventory;
• customer behaviour;
• time;
• location.

They can then automatically adjust prices.

Competition concerns

Algorithmic pricing may potentially:

• facilitate coordination;
• make price changes extremely rapid;
• increase transparency among competitors;
• discriminate between consumers;
• reduce independent price-setting.

However, algorithmic pricing is not automatically unlawful. The competition-law question depends on the circumstances and competitive effects.

9. Algorithmic Coordination

Reflexive markets can make coordination easier because algorithms can observe and react to competitors quickly.

The theoretical cycle is:

Competitor changes price → Algorithm observes → Algorithm responds → Competitor's algorithm observes → Further response

This can produce rapid price convergence even without traditional human communication.

Competition authorities therefore increasingly consider the relationship between:

• algorithms;
• information;
• pricing;
• market transparency;
• coordination.

10. Self-Preferencing

A dominant platform may use data generated by third-party businesses to compete against them.

For example:

1. marketplace hosts independent sellers;
2. platform observes seller sales;
3. platform obtains information about consumer demand;
4. platform launches competing products;
5. platform uses its control over ranking or visibility.

The competition concern is that the platform may combine intermediary power with competitive activity.

11. Ranking and Search Algorithms

Search and ranking systems are particularly important.

A platform determines:

• which products appear first;
• which sellers receive visibility;
• which advertisements are displayed;
• which recommendations are made.

In a reflexive market, rankings influence consumer choices.

Those choices generate new data.

That data may then influence future rankings.

Therefore:

Ranking can become part of the market-feedback mechanism.

12. Consumer Choice Architecture

Digital platforms can influence choices through:

• default settings;
• recommendation systems;
• personalised interfaces;
• notifications;
• search design;
• subscription prompts.

Competition concerns may arise if a dominant platform systematically directs users toward its own products or away from competitors.

This overlaps with issues involving:

• self-preferencing;
• tying;
• exclusion;
• consumer choice restrictions.

13. Case Law

Because reflexive digital markets are an emerging concept, courts generally have not created a doctrine formally called “reflexive digital markets.” The following cases provide relevant principles by analogy.

Case 1: United States v. Microsoft Corp., 253 F.3d 34 (D.C. Cir. 2001)

Facts

Microsoft possessed substantial power in PC operating systems. It engaged in various contractual and technical practices concerning web browsers, computer manufacturers and software developers.

Competition issue

The court examined whether Microsoft's conduct protected its operating-system monopoly by restricting competitive threats.

Principle

The case illustrates how a dominant technology company can use its control over an ecosystem to restrict competitive opportunities.

Relevance to reflexive markets

Microsoft's control over the Windows environment demonstrates an important digital-market principle:

Control over an important technological platform can influence the competitive conditions faced by other businesses.

In modern reflexive markets, this principle can extend to data, algorithms, interfaces and digital infrastructure.

14. Case 2: Google Shopping – Google and Alphabet v European Commission, Case C-48/22 P

Facts

Google operated a general search engine and its own comparison-shopping service.

The European Commission found that Google systematically gave its own comparison-shopping service preferential positioning and display in general search results while competing comparison-shopping services received less favourable treatment.

Competition issue

The case concerned Google's use of its dominant general-search position and treatment of competing services.

Principle

The case demonstrates that the way a dominant platform structures its search and ranking mechanisms can have competition-law significance.

Relevance to reflexive markets

Google's search system does not simply report market information.

Search rankings can influence:

• consumer attention;
• traffic;
• commercial visibility;
• business success.

Those outcomes subsequently generate new information for the search system.

Thus, the case is highly relevant to the idea of algorithmic market feedback.

15. Case 3: Ohio v. American Express Co., 585 U.S. 529 (2018)

Facts

American Express operated a two-sided payment network connecting cardholders and merchants.

Its merchant agreements contained anti-steering provisions.

Competition issue

The U.S. Supreme Court considered the competitive effects of the conduct in the context of a two-sided platform.

Principle

The Court emphasised the importance of analysing both sides of a transaction platform where indirect network effects connect the two groups.

Relevance

Reflexive digital markets often contain multiple interconnected groups:

• consumers;
• sellers;
• advertisers;
• developers;
• service providers.

Changes on one side can influence the other side.

Lesson

Competition analysis of digital platforms may require consideration of the interconnected feedback relationships among different groups.

16. Case 4: Eturas, Case C-74/14

Facts

Eturas operated a common online travel-booking system used by travel agencies.

The platform sent an electronic message implementing a technical restriction on discounts that agencies could offer.

Competition issue

The case concerned whether the platform's technical mechanism could facilitate coordinated conduct.

Principle

A digital platform's technical architecture and communications can be relevant to competition law where they facilitate coordination.

Relevance to reflexive markets

This is particularly important because the platform itself can influence the behaviour that it subsequently observes.

The cycle can be:

Platform rule → business behaviour → market data → platform adjustment

Lesson

Technical design can affect competitive behaviour, not merely provide neutral infrastructure.

17. Case 5: United States v. Apple Inc. – E-books, 791 F.3d 290 (2d Cir. 2015)

Background

The litigation concerned Apple's role in the e-books market and alleged coordination with publishers.

Competition issue

The Second Circuit considered whether Apple participated in a scheme that increased e-book prices.

Relevance

The case illustrates the importance of examining how a digital intermediary's contractual arrangements can influence market outcomes.

In a reflexive digital environment, contractual structures may alter:

• pricing;
• distribution;
• market transparency;
• competitive responses.

Lesson

Digital intermediaries can influence market conditions through contractual and platform arrangements.

18. Case 6: Aspen Skiing Co. v. Aspen Highlands Skiing Corp., 472 U.S. 585 (1985)

Facts

Aspen Skiing and Aspen Highlands operated competing ski facilities.

Aspen Skiing discontinued cooperation involving a multi-area ticketing arrangement.

Competition issue

The Supreme Court considered whether the refusal to continue cooperation constituted unlawful exclusionary conduct.

Relevance

The case is not a digital-market case, but its refusal-to-deal principles can be relevant by analogy where a dominant digital platform:

• withdraws interoperability;
• removes previously available access;
• terminates an established technical integration.

Limitation

The case does not mean that every digital platform must provide access to competitors.

Lesson

Refusal to cooperate becomes a competition-law issue only under the applicable legal conditions.

19. Case 7: Bronner v. Mediaprint, Case C-7/97

Facts

Bronner sought access to Mediaprint's newspaper-delivery system.

Competition issue

The European Court considered whether refusal to provide access to infrastructure could amount to abuse of dominance.

Principle

The Court established strict conditions for treating refusal to provide access as abusive.

Relevance

The case is relevant to digital infrastructure such as:

• APIs;
• operating systems;
• cloud infrastructure;
• interoperability systems.

Lesson

Control over infrastructure does not automatically create an obligation to share it with competitors.

20. Case 8: CCI – Google Android, Case No. 39 of 2018

Background

The Competition Commission of India examined Google's Android ecosystem and related practices involving:

• mobile operating systems;
• app stores;
• search services;
• licensing arrangements;
• restrictions affecting competition.

Competition relevance

The proceedings illustrate how digital ecosystems can connect several markets.

Relevance to reflexive markets

An ecosystem can generate data and behavioural feedback across several services.

For example:

Android users → app usage → search data → advertising information → improved services → more users

Lesson

Digital ecosystems can create interconnected competitive advantages across multiple related markets.

21. Case 9: Matrimony.com v. Google, CCI Cases 07 and 30 of 2012

Background

The Competition Commission of India examined Google's conduct relating to online search and search advertising.

Competition relevance

The proceedings involved issues concerning:

• search;
• online advertising;
• search bias;
• preferential treatment.

Relevance to reflexive markets

Search rankings can affect traffic and consumer behaviour.

That behaviour subsequently produces additional data.

Consequently:

Search algorithm → consumer behaviour → data → algorithmic adjustment

can create a reflexive market environment.

22. Case 10: Epic Games v. Google

Background

The litigation concerned Google's Android ecosystem, including Google Play's distribution and payment arrangements.

Competition relevance

The case illustrates how control over app distribution can influence:

• developer access;
• payment options;
• distribution channels;
• consumer access.

Relevance to reflexive digital markets

An app store does not merely facilitate transactions. Its rules can affect:

• which applications succeed;
• what payment systems are used;
• what data is generated;
• how developers respond.

This creates a continuous feedback relationship between platform governance and market behaviour.

23. Major Competition Implications

A. Reinforcement of Market Power

Reflexivity can reinforce an incumbent's position.

Cycle

Market power

↓

More users

↓

More data

↓

Better algorithms

↓

Better service

↓

More users

This can make entry increasingly difficult.

24. B. Raising Barriers to Entry

New entrants may face disadvantages because they lack:

• historical data;
• user information;
• behavioural insights;
• sophisticated algorithms;
• network effects;
• established reputation.

Consequently, a market may technically be open but practically difficult to enter.

25. C. Data Advantage

Data can become a strategic competitive asset.

A dominant platform may know:

• what consumers search for;
• what consumers buy;
• how much consumers pay;
• which advertisements work;
• which sellers perform best.

A new entrant may lack equivalent information.

26. D. Feedback-Driven Self-Preferencing

A platform can potentially use information generated through its intermediary role to improve competing products.

For example:

Marketplace operation → seller data → consumer data → platform's competing product → improved ranking

This raises questions about:

• data access;
• conflicts of interest;
• self-preferencing;
• discriminatory ranking.

27. E. Algorithmic Discrimination

Algorithms can differentiate between:

• sellers;
• advertisers;
• consumers;
• products.

Competition concerns can arise if a dominant firm applies discriminatory rules to disadvantage competitors without legitimate justification.

28. F. Reduced Market Transparency

Reflexive systems may make it difficult for market participants to understand:

• why prices changed;
• why rankings changed;
• why visibility declined;
• how recommendations are generated.

Opacity may increase dependence on the platform.

29. G. Switching Costs

Reflexive platforms may learn continuously from user behaviour.

If users accumulate:

• profiles;
• recommendations;
• transaction history;
• ratings;
• social connections;
• personalised settings,

switching to another platform may become more difficult.

This can reduce competitive pressure.

30. H. Multi-Homing

Multi-homing means using several competing platforms.

Reflexive markets may reduce multi-homing through:

• loyalty programs;
• technical restrictions;
• incompatible systems;
• contractual conditions;
• data portability barriers.

Lower multi-homing can strengthen platform power.

31. I. Consumer Manipulation and Competition

Competition law generally focuses on competitive process rather than every undesirable consumer-interface practice.

However, where interface design is used by a dominant firm to:

• exclude competitors;
• prevent switching;
• steer consumers;
• favour its own products,

it may become relevant to competition analysis.

32. J. Innovation Effects

Reflexive markets can have two opposing effects.

Positive

Data-driven feedback can produce:

• improved products;
• faster innovation;
• personalised services;
• better matching;
• lower transaction costs.

Negative

Strong feedback loops may:

• protect incumbents;
• discourage entry;
• reduce experimentation;
• make alternative technologies difficult to scale.

Therefore, competition analysis should examine actual market effects rather than assuming that algorithmic feedback is inherently harmful.

33. Reflexive Markets and Tipping

A digital market may experience tipping.

The process can be:

Initial advantage

↓

More users

↓

More data

↓

Better service

↓

More users

↓

Competitors lose scale

↓

Further incumbent advantage

If network effects and data advantages are sufficiently strong, a market can become increasingly concentrated.

However, concentration alone does not establish unlawful conduct.

34. Reflexive Markets and Self-Reinforcing Dominance

A useful analytical model is:

Stage 1 — Initial advantage

A firm has better technology or more users.

Stage 2 — Data accumulation

It obtains more information.

Stage 3 — Algorithmic improvement

The additional information improves predictions.

Stage 4 — Behavioural influence

Better rankings or recommendations attract more users.

Stage 5 — Market reinforcement

The resulting scale produces additional data.

This can create a self-reinforcing competitive advantage.

35. Competition Law Analysis

A competition authority can consider the following questions:

Question 1: What is the relevant market?

Is it:

• search;
• online advertising;
• app distribution;
• payment services;
• online retail;
• cloud computing?

Question 2: Does the undertaking possess substantial market power?

Relevant factors include:

• market share;
• network effects;
• switching costs;
• data;
• entry barriers;
• ecosystem dependence.

Question 3: What is the feedback mechanism?

Identify:

• data collection;
• ranking;
• recommendation;
• pricing;
• contractual restrictions.

Question 4: Does the conduct exclude competitors?

Examine:

• foreclosure;
• discriminatory access;
• self-preferencing;
• tying;
• refusal to deal;
• exclusive arrangements.

Question 5: Are there legitimate efficiencies?

Possible explanations include:

• security;
• privacy;
• fraud prevention;
• quality;
• innovation;
• efficiency.

36. Indian Competition Law Perspective

The Competition Act, 2002 provides several relevant provisions.

Section 3

Section 3 addresses agreements that cause or are likely to cause an appreciable adverse effect on competition.

Potential digital issues include:

• coordination;
• restrictive agreements;
• platform-based restrictions;
• information exchange.

Section 4

Section 4 addresses abuse of dominant position.

Reflexive digital markets may raise questions concerning:

• unfair conditions;
• discriminatory conditions;
• denial of market access;
• tying;
• leveraging;
• exclusionary conduct.

Sections 19 and 26

These provisions are relevant to investigation and assessment of competition concerns.

In digital markets, the CCI may need to consider:

• network effects;
• data;
• switching costs;
• multi-homing;
• platform dependence;
• ecosystem relationships.

37. Regulatory Challenges

1. Rapid technological change

Algorithms may change faster than traditional enforcement processes.

2. Difficult causation

It can be difficult to establish whether market outcomes resulted from:

• legitimate innovation; or
• exclusionary conduct.

3. Data complexity

Competition authorities may need to understand enormous datasets.

4. Multi-sided markets

The platform may serve several groups simultaneously.

5. Dynamic competition

Today's competitive advantage may disappear through innovation tomorrow.

6. Algorithmic opacity

Authorities may not easily understand how algorithms generate outcomes.

38. Possible Competition Remedies

Depending on the infringement and jurisdiction, remedies may include:

Behavioural remedies

• prohibit discriminatory ranking;
• prohibit exclusionary contracts;
• require transparent access conditions.

Interoperability

Require appropriate technical compatibility.

Data portability

Allow users to transfer relevant data where legally appropriate.

Access remedies

Provide access to certain infrastructure under legally justified conditions.

Structural remedies

In exceptional circumstances, authorities may consider structural separation.

Monitoring

Require continuing compliance monitoring.

39. Pro-Competitive and Anti-Competitive Effects

40. Reflexive Digital Markets and Consumer Welfare

Consumer welfare should be considered in both short-term and long-term terms.

Short-term benefits

• lower prices;
• convenience;
• better recommendations;
• improved services;
• faster transactions.

Long-term competition concerns

• reduced choice;
• fewer competitors;
• higher switching costs;
• reduced innovation;
• increased platform dependence.

Thus, an apparently beneficial algorithm may still deserve competition-law examination if it contributes to long-term exclusion.

41. Important Distinction: Reflexivity Is Not Automatically Anti-Competitive

A market does not become unlawful merely because:

• it uses algorithms;
• it collects data;
• it has network effects;
• consumers receive personalised recommendations;
• firms use dynamic pricing.

The legal question is whether market power is combined with conduct that unlawfully restricts competition, taking account of legitimate efficiencies and applicable legal standards.

42. Key Case-Law Principles

43. Exam-Oriented Framework

For an examination answer, remember:

R-D-A-B-E

R – Reflexivity
Market behaviour continuously feeds back into the digital system.

D – Data
Data creates information and potentially competitive advantages.

A – Algorithms
Algorithms convert data into rankings, prices and recommendations.

B – Barriers
Network effects, data advantages and switching costs may create barriers.

E – Exclusion
Competition law focuses on whether market power is used to exclude competitors.

44. Six Most Important Points

1. Reflexive digital markets continuously learn from market behaviour.
2. Data and algorithms create feedback loops that may reinforce competitive advantages.
3. Network effects can amplify those feedback loops.
4. Dominant platforms may influence markets through ranking, recommendation, pricing and access rules.
5. Self-preferencing, exclusion, discriminatory access, tying and coordination can create competition concerns.
6. Competition law should distinguish legitimate data-driven innovation from conduct that unlawfully protects or extends market power.

45. Conclusion

Reflexive digital markets represent a major development in competition economics because digital platforms increasingly operate within continuous feedback loops. They observe market behaviour, process data through algorithms, make decisions that influence consumers and businesses, and then use the resulting behaviour as new market information.

This creates a potentially powerful cycle:

Data → Prediction → Decision → Behavioural change → New data → Better prediction

Where a firm possesses substantial market power, this cycle may reinforce its position through network effects, data advantages, switching costs and algorithmic optimisation.

Cases such as United States v. Microsoft, Google Shopping, Ohio v. American Express, Eturas, Bronner, Aspen Skiing, Google Android, Matrimony.com v. Google, Apple E-books and Epic Games v. Google provide principles that can be applied to different aspects of this emerging problem.

The central competition-law principle is therefore:

Reflexivity itself is not anti-competitive. The competition concern arises when a powerful digital undertaking uses data, algorithms, platform control or feedback mechanisms in a manner that unlawfully excludes competitors, restricts market access, facilitates coordination, or reinforces market power without sufficient legitimate justification.