Indirect Network Externalities in Digital Markets

1. Introduction

Indirect network externalities arise when the value of a digital platform to one group of users increases because the number, activity, or quality of participants on another side of the platform increases.

They are particularly important in multi-sided digital markets, where a platform simultaneously serves different but interdependent groups—for example:

• users and advertisers;
• consumers and app developers;
• buyers and sellers;
• drivers and passengers;
• merchants and payment users;
• content creators and viewers;
• advertisers and publishers.

The essential feature is cross-group feedback.

For example, a mobile operating system becomes more attractive to consumers when more developers create useful applications. At the same time, developers have greater incentives to develop for that operating system when it has a large consumer base.

Thus:

More users → greater attractiveness to complementary participants → more complements/services → greater attractiveness to users.

This feedback can generate substantial efficiencies, but it can also create market concentration, entry barriers, tipping, self-preferencing incentives, and durable platform power.

2. Meaning of Indirect Network Externalities

A traditional network externality occurs when the value of a product increases directly with the number of other users of that same product.

Direct network externality

For example:

More WhatsApp users → WhatsApp becomes more useful to another WhatsApp user.

This is a direct network effect.

Indirect network externality

An indirect network effect operates through a different participant group.

For example:

More iPhone users → larger potential customer base → more developers create iOS applications → more applications → greater value of iPhone to consumers.

The effect therefore operates indirectly:

Group A grows → Group B becomes more attractive → Group B grows → Group A becomes more attractive.

3. Why Indirect Network Effects Matter in Competition Law

Indirect network effects can have several competition-law consequences.

A. Barriers to entry

A new platform may have difficulty attracting users because it lacks complementary participants.

A new app store, for example, may have difficulty attracting consumers without applications, while developers may refuse to participate because there are insufficient consumers.

This produces a chicken-and-egg problem.

B. Market tipping

A market may eventually move toward one or a few dominant platforms.

Once a platform reaches sufficient scale, its network advantages can make entry increasingly difficult.

C. Feedback loops

Digital markets can generate reinforcing loops:

Users → data → better service → more users → more data → better service.

Where another participant group is involved:

Users → advertisers → advertising revenue → investment → better platform → more users.

D. Switching costs

Indirect network effects can make switching difficult because leaving one platform may mean losing access to:

• applications;
• merchants;
• social connections;
• accumulated reputation;
• content;
• payment relationships;
• interoperability;
• advertising reach.

E. Self-preferencing

A dominant platform may have an incentive to favor its own complementary services because strengthening the complementary side can strengthen the platform's overall ecosystem.

F. Exclusionary conduct

A dominant platform may attempt to restrict rivals' access to one side of the ecosystem.

Examples include:

• exclusive distribution;
• restrictions on alternative payment systems;
• contractual restrictions on developers;
• discriminatory ranking;
• tying;
• refusal to interoperate;
• restricting multi-homing.

4. Economic Structure of Indirect Network Effects

Consider a platform PP connecting two groups:

• Group A = consumers
• Group B = sellers

The utility of a consumer can be represented conceptually as:

UA=f(NB)U_A = f(N_B)

where NBN_B represents the number or quality of sellers.

Conversely:

UB=g(NA)U_B = g(N_A)

where NAN_A represents the number of consumers.

The platform therefore experiences cross-side network effects.

The competitive significance becomes particularly strong when:

∂UA∂NB>0\frac{\partial U_A}{\partial N_B}>0

and

∂UB∂NA>0\frac{\partial U_B}{\partial N_A}>0

That is, additional participants on one side increase the value received by the other side.

5. Indirect Network Effects and Market Definition

Competition authorities cannot always analyze digital platforms as ordinary one-sided markets.

For example, a search engine may appear to provide free services to users, but simultaneously operates an advertising market.

Likewise:

• app stores connect developers and users;
• payment systems connect merchants and consumers;
• marketplaces connect buyers and sellers;
• advertising platforms connect advertisers and publishers/users.

A competition authority therefore needs to consider:

1. the different sides of the platform;
2. cross-side demand;
3. network effects;
4. multi-homing;
5. switching costs;
6. platform intermediation;
7. data advantages;
8. the possibility of market tipping.

6. Important Case Laws

Case 1: Ohio v. American Express Co. (2018)

Facts

The case concerned American Express's contractual restrictions preventing merchants from steering customers toward alternative payment methods.

American Express operated a two-sided transaction platform, connecting cardholders and merchants.

The Supreme Court recognized the importance of considering both sides of the platform in assessing competitive effects.

Legal significance

The case is highly relevant to indirect network externalities because increased participation on one side of a transaction platform can increase the value of participation on the other.

More cardholders can make acceptance attractive to merchants.

More merchants accepting the card can make the card more attractive to consumers.

Thus:

Cardholders ↔ Merchants

creates a cross-side feedback mechanism.

Principle

The competitive analysis of a two-sided platform may need to account for both sides simultaneously, particularly where the platform facilitates a single transaction between them.

Relevance to digital markets

The reasoning is highly relevant to:

• payment platforms;
• marketplaces;
• app stores;
• advertising exchanges;
• online platforms connecting buyers and sellers.

7. United States v. Microsoft Corp. (2001)

Facts

Microsoft was accused of using its dominance in PC operating systems to restrict competition from Netscape and Java.

The case involved important network and platform dynamics surrounding Windows.

Indirect network effect

Windows benefited from the availability of a large number of applications.

Developers, in turn, had strong incentives to write applications for Windows because of its enormous installed user base.

This created a reinforcing relationship:

More Windows users → more software development → more value for Windows users → more Windows adoption.

Competition concern

Microsoft's conduct could therefore be understood against the background of a platform whose installed base generated significant ecosystem advantages.

Principle

Network effects can strengthen incumbent power and make apparently small exclusionary actions strategically significant.

Relevance

The case remains important for understanding:

• software ecosystems;
• platform dominance;
• application ecosystems;
• compatibility;
• entry barriers;
• ecosystem foreclosure.

8. United States v. Apple Inc. — e-books (2013)

Facts

Apple was found liable for conspiring with publishers to raise e-book prices.

The case involved Apple's entry into the digital distribution ecosystem through the iPad and iBookstore.

Network-effect dimension

Digital content platforms can benefit from complementary content.

More attractive content can increase consumer demand for the platform, while a larger consumer base makes the platform more attractive to publishers.

Thus:

Consumers → publishers/content → consumers

can create an indirect network effect.

Competition significance

The case demonstrates how control over a digital distribution ecosystem can influence competition among complementary suppliers.

It also illustrates why contractual arrangements within digital ecosystems may affect competition beyond the immediate contracting parties.

9. European Commission — Google Shopping (2017)

Facts

The European Commission found that Google had abused its dominant position by systematically favoring its own comparison-shopping service in general search results.

Indirect network-effect dimension

Search engines and online advertising ecosystems involve interconnected groups.

More users generate:

• search queries;
• data;
• advertising opportunities;
• traffic;
• commercial information.

Those resources can increase the platform's attractiveness to advertisers and merchants.

At the same time, merchants' participation increases the commercial usefulness of the search ecosystem.

Competition concern

Preferential treatment of Google's own comparison-shopping service could therefore strengthen the platform's position in an adjacent market while benefiting from advantages generated by the dominant search ecosystem.

Principle

A dominant platform may not necessarily compete only within a single product market. Its control over one ecosystem layer can create advantages in neighboring markets.

10. Google Android — European Commission (2018)

Facts

The European Commission found that Google imposed restrictions concerning Android devices, including requirements relating to the pre-installation of Google Search and Chrome and restrictions concerning Android forks.

Indirect network effects

Android represents a classic multi-sided ecosystem involving:

• users;
• device manufacturers;
• application developers;
• advertisers;
• search services.

More Android users can attract more developers.

More applications can make Android devices more attractive to users.

More users can increase advertising and search revenues.

The resulting cycle can be represented as:

Users → developers → applications → users

and:

Users → search activity → advertising revenue → ecosystem investment → users.

Competition concern

The Commission's reasoning demonstrates how contractual restrictions can protect ecosystem advantages and potentially reinforce dominance.

Principle

In platform ecosystems, conduct affecting one participant group may have consequences across the entire ecosystem.

11. Epic Games, Inc. v. Apple Inc. (2021)

Facts

Epic Games challenged Apple's App Store policies, particularly restrictions concerning alternative payment mechanisms and app distribution.

The case involved the relationship between:

• iOS users;
• application developers;
• Apple's App Store;
• payment systems.

Indirect network effects

The App Store exhibits a classic cross-side relationship:

More users → more potential customers for developers.

At the same time:

More developers/apps → more value for users.

This creates:

Users↔DevelopersUsers \leftrightarrow Developers

The payment system is embedded within this ecosystem.

Competition significance

The case illustrates why restrictions imposed on developers can have consequences for consumers even when the immediate contractual restriction concerns only developers.

Restrictions on alternative payment mechanisms may affect:

• developer costs;
• consumer prices;
• innovation;
• platform entry;
• alternative distribution channels.

Principle

Competition analysis of digital ecosystems must consider the interaction between platform governance and cross-side network effects.

12. European Commission — Google AdSense (2019)

Facts

The European Commission found that Google imposed restrictive contractual provisions concerning online search advertising intermediation.

Indirect network-effect dimension

Advertising intermediaries connect:

• advertisers;
• publishers;
• users.

Greater participation by advertisers can increase the attractiveness of the advertising platform to publishers, while greater publisher reach makes the platform more attractive to advertisers.

Thus:

Advertisers ↔ Publishers

operate through an interconnected platform.

Competition concern

Contractual restrictions that reduce the ability of publishers to use competing advertising intermediaries may protect an incumbent's ecosystem advantages.

Principle

Network effects can make exclusionary contractual restrictions more significant because restricting rivals' access to one side can undermine their ability to achieve the scale needed to compete on the other.

13. Facebook/Meta — Social Networking and Platform Ecosystem Cases

Competition authorities have repeatedly examined Facebook/Meta's ecosystem through the interaction between users, advertisers, developers and complementary services.

The fundamental economic structure is:

Users → attention/data → advertisers

and:

Users → developers/content creators → more content → users.

Competition significance

A platform with a large user base may become increasingly attractive to advertisers and developers.

That can generate additional revenue and content, which can then be reinvested into the platform.

The resulting feedback can reinforce incumbent advantages.

This is particularly relevant to theories of:

• data-driven network effects;
• attention markets;
• ecosystem lock-in;
• platform concentration;
• entry barriers.

14. Summary of the Case Law

15. Indirect Network Effects and Tipping

One of the most important consequences is market tipping.

Suppose Platform A has:

• 60 million users;
• 1 million developers.

Platform B has:

• 5 million users;
• 100,000 developers.

Even if Platform B has better technology, developers may prefer Platform A because it offers access to a larger customer base.

Consumers then prefer Platform A because it offers more applications.

The resulting cycle is:

Large user base

↓

More developers

↓

More applications/services

↓

Higher consumer value

↓

More users

↓

Even more developers

This can make a digital market highly concentrated.

16. Multi-Homing as a Constraint

Indirect network effects do not automatically establish durable monopoly power.

A critical question is whether users can multi-home.

Multi-homing

A user participates in multiple platforms simultaneously.

Examples:

• sellers using several marketplaces;
• advertisers using several advertising networks;
• developers publishing applications across multiple ecosystems.

If multi-homing is easy, network effects may be weaker as an entry barrier.

If multi-homing is expensive or technically restricted, the incumbent's network advantage can become substantially stronger.

17. Single-Homing and Platform Power

Single-homing occurs when participants primarily use one platform.

Single-homing can strengthen indirect network effects.

For example:

Consumers single-home → sellers need that platform → sellers attract more consumers → rival platforms struggle to obtain sellers.

This creates an entry barrier that is not necessarily based on superior technology.

It is based on ecosystem coordination.

18. Data as an Indirect Network Externality

Modern digital platforms may combine traditional network effects with data feedback effects.

For example:

More users

↓

More behavioral data

↓

Better recommendation algorithm

↓

Better user experience

↓

More users

This can be combined with an advertising ecosystem:

More users → more advertising inventory → more advertisers → more revenue → greater investment → better service → more users.

Therefore, digital platforms can possess multiple reinforcing advantages simultaneously:

1. network effects;
2. data advantages;
3. economies of scale;
4. switching costs;
5. ecosystem integration.

19. Competition Risks

Indirect network effects may produce several forms of competitive harm.

1. Entry barriers

New competitors cannot easily attract both sides simultaneously.

2. Tipping

One platform may become the dominant ecosystem.

3. Foreclosure

The incumbent can prevent rivals from obtaining sufficient scale.

4. Self-preferencing

The platform can favor its own complementary products.

5. Exclusivity

Participants may be prevented from joining rival platforms.

6. Interoperability restrictions

The dominant platform may make interoperability difficult.

7. Data accumulation

A large network can generate disproportionately large quantities of data.

8. Innovation suppression

Potential competitors may abandon entry because they cannot achieve sufficient scale.

20. Competition-Law Assessment Framework

A competition authority examining indirect network effects should ask:

Step 1 — Identify the platform sides

Who are the relevant participant groups?

Step 2 — Identify cross-side effects

Does growth on one side increase the value of participation on another?

Step 3 — Measure network strength

Consider:

• number of users;
• transaction volume;
• developer participation;
• merchant participation;
• advertising reach;
• content availability.

Step 4 — Examine multi-homing

Can participants easily use competing platforms?

Step 5 — Examine switching costs

Are users locked into:

• data;
• applications;
• reputation;
• contracts;
• social connections;
• payment systems?

Step 6 — Examine interoperability

Can rivals connect to the incumbent ecosystem?

Step 7 — Examine exclusionary conduct

Has the dominant platform:

• restricted access;
• imposed exclusivity;
• tied products;
• self-preferenced;
• restricted interoperability;
• discriminated against rivals?

Step 8 — Examine efficiencies

Network effects can generate legitimate benefits:

• lower transaction costs;
• better matching;
• improved quality;
• innovation;
• greater variety;
• lower search costs.

The existence of network effects therefore does not itself constitute anticompetitive conduct.

21. Key Distinction: Network Effect vs Network Externality

The terms are sometimes used interchangeably, but the distinction can be useful.

Network effect

A change in the number of participants changes the value of the platform.

Network externality

That change in value is experienced by other participants without necessarily being fully reflected in the private decision of the participant creating the effect.

For competition law, the important issue is whether these effects create self-reinforcing advantages that competitors cannot readily replicate.

22. Indirect Network Effects and Article 102 / Abuse of Dominance

Under abuse-of-dominance frameworks, indirect network effects can be relevant to:

• defining the relevant market;
• assessing dominance;
• determining barriers to entry;
• evaluating foreclosure;
• assessing exclusionary conduct;
• evaluating indispensability;
• assessing leveraging into adjacent markets.

The central question is not:

"Does the platform have network effects?"

Rather:

"Does the platform's network advantage materially strengthen market power or make exclusion of equally efficient competitors more likely?"

23. Indirect Network Effects and Article 101 / Agreements

Agreements among platform participants can also exploit network effects.

Potentially problematic arrangements include:

• exclusivity;
• most-favored-nation clauses;
• anti-steering provisions;
• restrictions on alternative payment systems;
• restrictions on multi-homing;
• discriminatory access terms.

Where an incumbent controls a sufficiently important ecosystem, contractual restrictions can prevent rival platforms from reaching the minimum scale necessary to compete.

24. Remedies

Possible remedies include:

Structural remedies

• divestiture;
• separation of business units.

Behavioral remedies

• prohibition of exclusivity;
• non-discrimination requirements;
• access obligations.

Interoperability remedies

• API access;
• technical interoperability;
• data portability.

Data remedies

• restrictions on combining datasets;
• data access;
• portability.

Choice remedies

• default-choice screens;
• alternative payment options;
• alternative app stores.

Monitoring remedies

• independent compliance monitoring;
• algorithmic auditing;
• reporting obligations.

25. Conclusion

Indirect network externalities are among the most important sources of market power in digital markets.

They arise because participation by one group increases the value of the platform to another group. This creates powerful feedback loops:

Users → complementary participants → greater platform value → more users.

Cases such as Ohio v. American Express, Microsoft, Apple, Google Shopping, Google Android, Epic Games v. Apple, and Google AdSense demonstrate different ways in which cross-side platform relationships can affect competition.

The central competition-law challenge is therefore to distinguish between:

pro-competitive network efficiencies

and

network effects that are being exploited to entrench dominance or foreclose rivals.

In digital markets, the strongest competitive advantage may not be a traditional physical asset. It may be the self-reinforcing ecosystem created by millions of interconnected participants.