Interdependence-Based Dominance In Duopolistic Platform Ecosystems
Interdependence-Based Dominance in Duopolistic Platform Ecosystems
1. Introduction
Interdependence-based dominance in duopolistic platform ecosystems describes a situation in which two large digital platforms jointly structure the competitive environment, while each platform's market power depends partly upon the existence, behaviour, infrastructure, data, users, developers, merchants, or ecosystem of the other.
This is different from a conventional monopoly.
A monopoly involves:
one dominant undertaking → market control
A conventional duopoly involves:
two major competitors → substantial market shares
An ecosystem duopoly can be more complicated:
Platform A ↔ Platform B
↓ ↓
Users Developers
Merchants Advertisers
Data Infrastructure
↓ ↓
mutual dependence + competitive rivalry
The central legal question is therefore:
Can an undertaking possess legally relevant dominance even when its market power is partly generated or constrained by the existence of another powerful platform?
The answer under modern competition law can be yes, but interdependence itself does not automatically establish dominance or unlawful conduct.
2. Meaning of Interdependence-Based Dominance
Interdependence-based dominance arises where the competitive position of a platform is determined not simply by its independent market share but by its position within a network of relationships.
These relationships may involve:
users;
rival platforms;
app developers;
advertisers;
merchants;
payment providers;
cloud providers;
data brokers;
operating systems;
content providers; and
complementary applications.
For example:
Platform A
may control 55% of users, while
Platform B
controls 45%.
Neither has a traditional monopoly.
Yet developers may need access to both platforms, advertisers may depend on both, and consumers may use both simultaneously.
This produces ecosystem interdependence.
3. Duopoly Does Not Mean Absence of Dominance
A common misconception is:
"If there are two major competitors, neither can be dominant."
That is incorrect.
Dominance is not necessarily equivalent to monopoly.
An undertaking may possess a dominant position despite the existence of significant competitors if it has sufficient economic strength to behave to an appreciable extent independently of competitive constraints.
Thus, a 60/40 or even 55/45 market may require careful analysis of:
entry barriers;
network effects;
switching costs;
data;
interoperability;
user lock-in;
multi-homing;
ecosystem strength;
innovation;
countervailing buyer power.
4. Interdependence Is Different from Collective Dominance
This distinction is extremely important.
Individual dominance
Platform A independently possesses substantial market power.
Collective dominance
Two or more undertakings may jointly occupy a dominant position under applicable legal doctrine.
Interdependence
Each undertaking's competitive position is materially affected by the conduct and ecosystem of the other.
Interdependence does not automatically establish collective dominance.
A regulator must identify the appropriate legal theory and demonstrate the necessary elements.
5. Why Digital Duopolies Are Different
Traditional duopolies usually involve two firms selling substitutable products.
Digital ecosystems can involve much more complicated relationships.
Platform A may compete with Platform B for:
users,
while simultaneously depending upon Platform B for:
interoperability;
content;
advertisers;
payments;
cross-platform users;
data;
technical standards.
The firms are therefore:
competitors + complements + infrastructure partners + potential gatekeepers
at the same time.
6. Network Effects
Network effects are central.
Suppose Platform A has more users.
More users attract:
→ more developers
which attract:
→ more applications
which attract:
→ more users.
Platform B experiences a similar feedback loop.
The result may be a highly concentrated market in which both ecosystems become difficult to displace.
The critical competition question becomes whether the duopoly is:
contestable, or
structurally entrenched.
7. Direct and Indirect Network Effects
Direct network effects
The value of a service increases as more users join.
Examples include:
messaging;
social networks;
communication platforms.
Indirect network effects
The value to one side increases because another side becomes larger.
For example:
More users
→ attract advertisers.
More advertisers
→ generate revenue.
Revenue
→ funds platform improvements.
Improved platform
→ attracts more users.
This feedback structure can reinforce dominance.
8. Switching Costs
Duopolistic ecosystems can create significant switching costs.
Users may have accumulated:
contacts;
purchase histories;
digital identities;
subscriptions;
photographs;
playlists;
loyalty benefits;
application libraries;
cloud data.
Businesses may have invested in:
APIs;
software;
advertising systems;
developer tools;
platform-specific applications.
The result is ecosystem-specific investment.
A competitor may technically exist but still be unable to exert meaningful competitive pressure.
9. Multi-Homing
Multi-homing is an important counterweight.
If users and businesses can easily use both platforms, the competitive constraint may be stronger.
For example:
User → Platform A + Platform B
may prevent either platform from exercising substantial unilateral power.
But if the platforms impose:
exclusivity;
loyalty incentives;
technical restrictions;
incompatible standards;
high switching costs,
multi-homing may decline.
This can make dominance more durable.
10. Data Interdependence
Platform ecosystems frequently depend upon data flows.
Platform A may possess:
search data;
transaction data;
behavioural data.
Platform B may possess:
social data;
app usage data;
location data.
Third-party businesses may need access to both ecosystems.
Data therefore becomes an interdependent competitive resource.
The competition concern arises where one platform restricts data portability or interoperability in ways that weaken the competitive position of rivals.
11. Interoperability
Interoperability is particularly important in a duopoly.
Suppose:
Platform A ↔ Platform B
allow users to communicate or transfer data freely.
Competition may remain relatively fluid.
But if:
Platform A —X— Platform B
where X represents technical incompatibility, users may become locked into one ecosystem.
A dominant platform may therefore use interoperability restrictions as a form of exclusion.
12. Standards and Ecosystem Control
Two platforms may compete over technical standards.
The winning standard can determine:
developer access;
hardware compatibility;
data portability;
payment architecture;
advertising infrastructure.
Control over a technical standard can therefore become a source of structural market power.
13. Algorithmic Interdependence
Modern platform duopolies may also exhibit algorithmic interdependence.
Each platform continuously observes:
competitor pricing;
ranking;
advertising prices;
user migration;
product launches;
engagement levels.
Algorithms can respond rapidly.
This can potentially reduce uncertainty between competitors.
However:
algorithmic parallelism alone does not automatically constitute an unlawful agreement.
Competition authorities must establish the legal basis for intervention.
14. Coordinated Effects Versus Individual Dominance
A duopolistic ecosystem can generate two different concerns.
Individual dominance
Platform A uses its market power to:
exclude rivals;
self-preference;
impose discriminatory conditions;
exploit users.
Coordinated effects
Platforms A and B may become mutually accommodating or coordinate their conduct.
These theories must not be conflated.
A market can be highly concentrated without evidence of unlawful coordination.
15. Relevant Case Law
1. Airtours v Commission, Case T-342/99
Airtours is one of the most important authorities concerning collective dominance and coordinated effects.
The General Court developed criteria concerning whether firms can sustain coordinated behaviour in concentrated markets.
The analysis includes factors such as:
transparency;
ability to monitor deviations;
credible retaliation;
sustainability of coordination.
Relevance to platform ecosystems
Digital duopolies can be extraordinarily transparent.
Platforms can observe:
prices;
ranking;
advertising conditions;
user behaviour;
product launches.
Algorithmic monitoring may make deviations from coordinated conduct easier to detect.
However, Airtours concerns collective dominance, not merely interdependence.
16. Case 2: Gencor v Commission, Case T-102/96
Gencor established important principles concerning collective dominance in concentrated markets.
The Court accepted that dominance may arise through an economic structure that allows firms to behave collectively without necessarily having an explicit agreement.
Relevance
In a two-platform ecosystem, structural conditions could potentially facilitate coordinated behaviour where:
market transparency is high;
competitors are highly symmetric;
entry barriers are substantial;
retaliation is credible.
Again, structural interdependence alone does not prove collective dominance.
17. Case 3: Impala v Commission, Case T-464/04
The Impala litigation concerned the assessment of collective dominance in the music-recording industry.
The case demonstrated the evidentiary difficulty of proving coordinated dominance.
Relevance
Digital-platform authorities should not infer collective dominance merely because:
two firms have large market shares;
prices are similar;
strategies appear parallel.
A sufficiently robust evidentiary basis is required.
This is particularly important where algorithms produce similar outcomes independently.
18. Case 4: Microsoft v Commission, Case T-201/04
The Microsoft litigation is one of the most important authorities for digital ecosystem dominance.
Microsoft's control over the Windows operating-system environment allowed it to influence neighbouring markets.
Relevance
The case demonstrates that platform power may arise from control over:
technical interfaces;
interoperability;
software distribution;
ecosystem access.
In a duopoly, one platform may possess substantial power because it controls a critical ecosystem layer even though another large platform exists.
19. Case 5: Google Shopping, Case T-612/17
Google Shopping demonstrates how dominance can be exercised through ranking and visibility.
Google's search ecosystem gave it the ability to influence the competitive position of comparison-shopping services.
Relevance
In a platform duopoly, the key competitive weapon may not be price.
It may be:
visibility.
A platform could theoretically:
demote rival services;
favour its own products;
manipulate ranking;
alter recommendation systems.
Such conduct may weaken the competitive constraint imposed by the second ecosystem.
20. Case 6: Android, Case T-604/18
The Android case demonstrates how contractual and technical restrictions can reinforce ecosystem power.
The case concerned Google's arrangements involving Android and related services.
Relevance
In a duopolistic platform environment, contractual conditions concerning:
default settings;
app distribution;
search access;
application compatibility;
licensing;
may affect whether rival ecosystems can effectively compete.
The important lesson is that dominance may be maintained through ecosystem architecture rather than explicit exclusion.
21. Case 7: United Brands, Case 27/76
United Brands provides the classic definition of a dominant position.
Dominance involves a position of economic strength enabling an undertaking to behave to an appreciable extent independently of competitors, customers and consumers.
Relevance
The existence of a second major platform does not eliminate the possibility of dominance.
The question is whether Platform A possesses sufficient economic strength to operate with a substantial degree of independence despite Platform B.
22. Case 8: Hoffmann-La Roche, Case 85/76
Hoffmann-La Roche established the principle that a dominant undertaking has a special responsibility not to allow its conduct to impair genuine undistorted competition.
Relevance
If Platform A occupies a dominant position within one layer of a duopolistic ecosystem, it may have heightened obligations concerning:
exclusivity;
loyalty incentives;
interoperability;
data access;
discriminatory treatment.
Its responsibility does not arise simply because it is large, but because of its legally established dominant position.
23. Case 9: Bronner, Case C-7/97
Bronner is particularly relevant to interdependence and access.
The case concerned refusal to provide access to an infrastructure that a competitor considered important.
The Court imposed stringent conditions for mandatory access.
Relevance
In platform ecosystems, Platform B may argue:
"We depend on Platform A's infrastructure."
But dependency alone is insufficient.
The infrastructure must generally satisfy the demanding conditions associated with compulsory access.
This prevents competition law from converting every form of commercial interdependence into a mandatory-sharing obligation.
24. Case 10: IMS Health, Joined Cases C-418/01 P and C-7/01 P
IMS Health further developed the exceptional circumstances doctrine concerning refusal to license intellectual property.
Relevance
A platform may possess:
proprietary APIs;
data structures;
software interfaces;
interoperability technologies.
Competitors cannot automatically demand access merely because these assets are commercially important.
The case therefore provides an important limit on intervention in ecosystem disputes.
25. The Difference Between Dependency and Dominance
This distinction is crucial.
Dependency
A business may be economically dependent upon Platform A.
Dominance
Platform A possesses substantial market power within a relevant market.
Abuse
Platform A uses that position in a manner prohibited by competition law.
Thus:
Dependency → dominance → abuse
are three distinct analytical steps.
One cannot simply infer the third from the first.
26. The “Mutual Dependency” Problem
The most difficult scenario occurs when both platforms depend upon each other.
For example:
Platform A
controls hardware distribution.
Platform B
controls essential applications.
Users need both.
Platform A therefore depends upon B's applications, while B depends upon A's installed user base.
This creates mutual dependency.
Competition law must determine whether:
neither platform is dominant;
one platform is dominant in a particular market;
both possess separate forms of dominance;
collective dominance exists; or
the relationship is simply competitive interdependence.
27. Ecosystem Bottlenecks
A particularly important concept is the ecosystem bottleneck.
A platform may control a layer through which competitors must pass.
Examples include:
app stores;
operating systems;
search engines;
payment rails;
cloud infrastructure;
identity systems;
digital advertising exchanges.
Even if another platform competes downstream, the bottleneck can provide the platform controlling the infrastructure with significant power.
28. Competitive Constraint from the Rival Platform
The second platform can nonetheless impose competitive constraints.
For example, if Platform A raises prices or reduces quality, users may move to Platform B.
The strength of that constraint depends upon:
switching costs;
portability;
multi-homing;
compatibility;
user preferences;
network effects.
Therefore, a duopoly can constrain dominance more effectively when users can switch easily.
29. Lock-In Can Make a Duopoly Less Competitive
Suppose two platforms have:
50% + 50%
market shares.
At first glance, the market appears balanced.
But suppose each user has invested heavily in one ecosystem.
Then:
Platform A users → difficult to migrate
and
Platform B users → difficult to migrate.
The result can be a market with two enormous but relatively insulated ecosystems.
This may reduce the competitive pressure ordinarily associated with a duopoly.
30. Ecosystem Switching Costs
Switching costs can include:
loss of data;
loss of applications;
loss of subscriptions;
retraining employees;
rewriting software;
rebuilding advertising campaigns;
changing payment systems;
rebuilding customer relationships.
Business switching costs can therefore be considerably larger than consumer switching costs.
31. Multi-Homing as a Regulatory Safeguard
Competition authorities may therefore favour conditions that facilitate multi-homing.
Examples include:
data portability;
interoperability;
open APIs;
non-discriminatory access;
standardised interfaces.
The objective is not to eliminate successful ecosystems.
It is to ensure that users and businesses can realistically choose between competing ecosystems.
32. Self-Preferencing in a Duopoly
Suppose Platform A operates:
a marketplace; and
its own competing retail service.
It may rank its own retail products above those of sellers.
Platform B may do the same.
The result could be:
two ecosystems competing externally while restricting internal competition.
This distinction is important.
The existence of rivalry between ecosystems does not necessarily ensure competition within each ecosystem.
33. Vertical Integration
Digital duopolies frequently involve vertical integration.
For example:
Hardware → Operating system → App store → Payment → Advertising
A platform controlling multiple levels can use power from one level to strengthen another.
This is the classic leveraging problem.
The relevant competition-law question becomes:
Does the undertaking use power in one market to foreclose competition in an adjacent market?
34. Data Advantage and Feedback Effects
Suppose Platform A has more users than Platform B.
A's data advantage may improve:
search;
recommendations;
advertising;
fraud detection;
AI models.
This produces a feedback loop:
market share → data → better service → market share.
Platform B may be unable to reproduce this advantage.
Thus, interdependence can coexist with asymmetric power.
35. When Interdependence Becomes Legally Significant
Interdependence becomes competition-law significant when it produces or reinforces:
substantial market power;
exclusionary effects;
barriers to entry;
dependency;
discriminatory access;
foreclosure;
coordinated conduct.
The mere existence of interdependence is not unlawful.
Modern markets are inherently interdependent.
The legal question concerns how that interdependence is structured and exploited.
36. German Competition-Law Perspective
Under the GWB, this issue can be examined through:
Section 19
Abuse of dominant position.
Section 20
Relative or substantial market power and dependency.
Section 19a
Undertakings of paramount significance across markets.
This makes German law particularly interesting for platform duopolies.
A platform may not necessarily satisfy conventional dominance criteria in every market but may nevertheless create significant dependency for business users.
37. Section 19a and Duopolistic Ecosystems
Section 19a is particularly suited to ecosystem analysis.
The Bundeskartellamt can examine whether an undertaking possesses:
cross-market significance;
financial strength;
access to data;
vertical integration;
network effects;
strategic importance;
ability to leverage its position.
This is particularly relevant where two ecosystems dominate different layers of the digital economy.
38. Collective Dominance Under EU Law
EU competition law has recognised the possibility of collective dominance.
However, proving it requires more than showing that two undertakings are powerful.
The Airtours/Gencor line of jurisprudence demonstrates that authorities must establish the necessary structural and economic conditions for coordinated behaviour.
Therefore:
Duopoly ≠ collective dominance.
And:
Interdependence ≠ collective dominance.
39. Enforcement Difficulties
Several problems arise in enforcement.
First: market definition
Should regulators define:
a platform market;
an ecosystem market;
a service market;
a data market?
Second: dynamic competition
Technology changes rapidly.
Third: causation
Authorities must distinguish legitimate network effects from exclusionary conduct.
Fourth: innovation
Intervention can unintentionally reduce incentives to develop superior platforms.
Fifth: remedies
Breaking up or regulating ecosystems can create technical and economic complications.
40. Appropriate Competition-Law Remedies
Possible remedies include:
Interoperability obligations
Require technical compatibility where legally justified.
Data portability
Facilitate user movement between ecosystems.
Non-discrimination
Prevent discriminatory access conditions.
Restrictions on self-preferencing
Prevent a platform from unfairly favouring its own downstream services.
Contractual restrictions
Limit unjustified exclusivity.
Transparency
Require greater explanation of ranking or access rules in appropriate circumstances.
Structural remedies
In extreme cases, separation of infrastructure and downstream competitive activities may be considered.
41. Economic Test for Interdependence-Based Dominance
A useful analytical framework is:
Step 1
Identify the relevant market.
Step 2
Identify the two principal ecosystems.
Step 3
Measure market shares and concentration.
Step 4
Assess network effects.
Step 5
Assess switching costs.
Step 6
Measure multi-homing.
Step 7
Examine data advantages.
Step 8
Identify ecosystem bottlenecks.
Step 9
Assess vertical integration.
Step 10
Determine whether either undertaking can behave independently.
Step 11
If relevant, assess collective dominance.
Step 12
Analyse the specific conduct and competitive effects.
42. Conceptual Model
The phenomenon can be represented as:
Platform A
Data
↓
Users
↓
Developers
↓
Applications
↓
Network effects
↓
Market power
↕ Interdependence
Platform B
Data
↓
Users
↓
Developers
↓
Applications
↓
Network effects
↓
Market power
The legal analysis must determine whether the arrows between the platforms create:
competitive constraint;
mutual dependence;
coordinated power; or
asymmetric dominance.
43. Key Distinction: Competition Between Ecosystems vs Competition Within Ecosystems
This is perhaps the most important conceptual point.
A market can exhibit vigorous competition:
between Platform A and Platform B
while simultaneously suffering from weak competition:
inside Platform A and inside Platform B.
For example, two operating systems may vigorously compete for users while each restricts developers within its own ecosystem.
Competition authorities therefore increasingly need to consider both inter-ecosystem and intra-ecosystem competition.
44. Conclusion
Interdependence-based dominance in duopolistic platform ecosystems represents a significant evolution of traditional dominance analysis.
The central insight is that digital power cannot always be measured simply by asking:
"Which company has the largest market share?"
Instead, regulators must examine:
ecosystem structure;
network effects;
switching costs;
multi-homing;
data accumulation;
interoperability;
technical bottlenecks;
vertical integration;
dependency;
algorithmic interaction; and
cross-market leverage.
The leading cases provide complementary principles:
United Brands and Hoffmann-La Roche establish the foundations of individual dominance.
Airtours and Gencor provide the framework for analysing collective dominance and coordinated market structures.
Impala demonstrates the evidentiary difficulties involved in establishing collective dominance.
Microsoft demonstrates ecosystem and interoperability power.
Google Shopping illustrates ranking and self-preferencing concerns.
Android demonstrates how contractual and technical ecosystem arrangements can reinforce platform power.
Bronner and IMS Health establish limits on compulsory access to privately controlled infrastructure and intellectual property.
The ultimate principle is therefore:
A duopoly does not eliminate dominance, and interdependence does not itself establish dominance. Competition law must determine whether the relationship between the two ecosystems creates genuine competitive constraints or instead produces entrenched, asymmetric, or collectively exercisable market power.
In modern platform markets, the most important unit of analysis may consequently shift from the individual firm to the interdependent ecosystem—while preserving the requirement that any finding of dominance or abuse be grounded in the applicable legal criteria and evidence.

comments