Global Routing Governance (BGP) and Competitive Implications

Introduction

Border Gateway Protocol (BGP) is the principal inter-domain routing protocol used to exchange reachability information between autonomous systems (ASes) on the Internet. Autonomous systems include telecommunications operators, Internet service providers (ISPs), cloud providers, content-delivery networks, universities, enterprises and other network operators.

BGP is not merely a technical mechanism. Its governance can have significant competition-law implications because control over routing can affect:

• which networks can reach customers;
• the quality, latency and reliability of connectivity;
• interconnection and peering conditions;
• access to transit markets;
• traffic engineering;
• resilience and redundancy;
• cloud and CDN competition;
• data-centre connectivity;
• costs imposed on rivals;
• and ultimately the ability of competing digital services to reach users.

A competition-law analysis therefore asks whether routing governance is being used as a neutral coordination mechanism or as a means of foreclosure, discrimination, exclusion or strategic control over an essential digital infrastructure layer.

1. What Is BGP?

BGP enables autonomous systems to advertise IP prefixes and determine routes through which traffic can reach those prefixes.

A simplified structure is:

User → Access ISP → Transit/Peering Network → Other AS → Destination

BGP does not necessarily select the geographically shortest path. Operators establish policies concerning:

• preferred routes;
• transit relationships;
• peering;
• route preference;
• path length;
• traffic engineering;
• redundancy;
• filtering;
• security;
• and commercial agreements.

Consequently, routing decisions are partly technical and partly economic.

Example

Suppose ISP A connects to:

• Transit Provider X;
• Transit Provider Y; and
• CDN Z.

If ISP A systematically gives Z's competitor inferior routing while providing Z with better paths, the issue may move beyond ordinary network management into a potential competitive-discrimination problem.

2. Global Governance of BGP

BGP governance is highly decentralized.

Important participants include:

1. IETF – develops Internet routing standards.
2. Regional Internet Registries (RIRs) – administer Internet number resources.
3. Internet exchange points (IXPs) – facilitate interconnection and peering.
4. Network operators – establish routing policies.
5. Transit providers – sell connectivity between networks.
6. Cloud and CDN operators – increasingly operate substantial routing infrastructure.
7. National regulators – regulate telecommunications and digital infrastructure.
8. Competition authorities – intervene where routing practices affect competition.
9. Courts and tribunals – assess exclusionary conduct, essential facilities, discrimination and market power.

The important competition-law question is therefore:

Can a decentralized technical routing system nevertheless become a locus of private economic power?

The answer can be yes.

3. Relevant Competition Markets

BGP-related conduct can affect several overlapping markets.

A. IP Transit

A provider supplies connectivity allowing one network to reach many other networks.

B. Internet Access

ISPs connect end users to the Internet.

C. Peering

Networks exchange traffic directly rather than relying entirely on paid transit.

D. Interconnection

Network operators establish physical or logical connections for traffic exchange.

E. CDN Services

Content providers distribute content through geographically distributed infrastructure.

F. Cloud Connectivity

Cloud providers compete partly through connectivity, latency and network reach.

G. Data-Centre Interconnection

Access to strategically located data centres and IXPs can determine competitive network performance.

H. Managed Connectivity and Enterprise Networks

Large enterprises increasingly purchase connectivity, security and routing services from integrated providers.

4. Why BGP Creates Competition Concerns

A. Routing as a Bottleneck

A dominant network can potentially influence the ability of another network to reach customers.

Where alternative routes are technically or economically inadequate, control over routing may become a bottleneck facility.

The competitive concern increases when the network is:

• difficult to replicate;
• geographically indispensable;
• connected to large numbers of users;
• connected to important cloud/CDN infrastructure; or
• supported by strong network effects.

5. Peering and Transit Discrimination

One of the most important issues is discrimination between networks.

A powerful network could potentially:

• refuse peering;
• impose excessive transit prices;
• degrade interconnection;
• restrict route advertisements;
• selectively withdraw routes;
• impose discriminatory technical conditions;
• or favour affiliated services.

Not every refusal or commercial disagreement violates competition law.

The critical question is whether the conduct substantially harms competition rather than merely harming an individual competitor.

6. BGP and Essential-Facilities Doctrine

The essential-facilities concept can become relevant where access to a network facility is indispensable for competition.

A claimant generally has to establish concepts such as:

1. control of an important facility;
2. indispensability;
3. inability to reasonably duplicate it;
4. refusal or discriminatory access;
5. potential elimination of effective competition; and
6. absence of objective justification.

The doctrine must be applied cautiously because forcing infrastructure sharing can reduce incentives to invest.

BGP therefore creates a difficult balance:

Interoperability and access versus network investment and operational autonomy.

7. Route Leaks and Route Hijacking as Competition Issues

BGP was historically designed around trust between network operators.

A malicious or erroneous announcement can cause traffic to be:

• diverted;
• intercepted;
• black-holed;
• delayed;
• or redirected through another network.

Most route leaks are not automatically competition violations.

However, deliberate manipulation could theoretically become relevant where a firm uses routing control strategically to:

• disrupt a competitor;
• impair access to customers;
• create artificial congestion;
• undermine a rival's service quality;
• or favour an affiliated service.

This potentially transforms network-security conduct into competition-law conduct.

8. BGP Communities and Selective Routing

BGP communities allow operators to communicate routing preferences.

They are technically useful but potentially commercially significant.

For example, an operator might use routing policies to:

• prioritize its own CDN;
• deprioritize a rival;
• restrict announcements;
• influence geographic traffic flows;
• or alter the effective quality of competing services.

The competition-law inquiry should distinguish:

legitimate traffic engineering

from

strategic foreclosure.

9. Vertical Integration

Modern Internet infrastructure is increasingly vertically integrated.

A large technology company might operate:

Cloud + CDN + Backbone + Data Centres + DNS + Security + Connectivity

A telecommunications operator might operate:

Access ISP + Transit + Data Centre + Cloud Connectivity + Content Services

Vertical integration can produce efficiencies, but it also creates opportunities for self-preferencing.

For example:

A vertically integrated network could theoretically provide superior routing to its own cloud/CDN service while imposing inferior conditions on competing services.

That could raise concerns analogous to other digital-platform self-preferencing cases.

10. BGP and Network Effects

Internet routing exhibits powerful network effects.

A network connected to many other networks becomes more valuable because it can reach more destinations.

This can create a reinforcing cycle:

More Customers → More Traffic → More Peering → Better Reachability → More Customers

Large network operators may therefore develop structural advantages that are difficult for smaller entrants to replicate.

Competition authorities may need to examine not merely market share but:

• AS connectivity;
• route diversity;
• peering relationships;
• customer concentration;
• geographic reach;
• latency;
• switching costs;
• network capacity;
• and access to critical IXPs.

11. Six Important Case Laws

BGP-specific reported competition cases are relatively uncommon because routing disputes frequently arise through telecommunications regulation, private contracts, technical incidents or cybersecurity proceedings. Therefore, the most useful jurisprudence consists of telecommunications, interoperability, essential-facility, network-access and digital-infrastructure cases whose principles can be applied to BGP governance.

Case 1 — Bronner v Mediaprint

Case: Oscar Bronner GmbH & Co. KG v Mediaprint Zeitungs und Zeitschriftenverlag GmbH & Co. KG, C-7/97, Court of Justice of the European Union.

Principle

The CJEU established a demanding test for compulsory access to infrastructure controlled by a dominant undertaking.

The facility must generally be indispensable and there must be no realistic alternative.

BGP relevance

A network operator's backbone, transit network or interconnection facility should not automatically become an essential facility merely because access would be commercially convenient.

For BGP-related claims, a competitor should demonstrate:

• genuine indispensability;
• lack of economically viable alternatives;
• significant foreclosure;
• and the absence of objective justification.

Importance: Bronner prevents competition law from becoming a general infrastructure-sharing obligation.

Case 2 — IMS Health

Case: IMS Health GmbH & Co. OHG v NDC Health GmbH & Co. KG, Joined Cases C-241/00 P and C-242/00 P.

Principle

The case developed the exceptional circumstances under which refusal to provide access to an indispensable infrastructure or resource may constitute abuse of dominance.

BGP relevance

The principle is potentially relevant to:

• critical peering facilities;
• routing information;
• interconnection arrangements;
• proprietary network interfaces;
• and access to strategically indispensable infrastructure.

A competition authority would need to distinguish an ordinary commercial refusal from conduct capable of eliminating competition.

Importance: IMS Health demonstrates why indispensability must be assessed rigorously.

Case 3 — Deutsche Telekom

Case: Deutsche Telekom AG v Commission, C-280/08 P.

Principle

The CJEU upheld the principle that a dominant vertically integrated telecommunications operator can abuse its position through pricing practices that squeeze competitors operating at downstream levels.

BGP relevance

This is particularly important for vertically integrated network operators.

Imagine a company controlling:

• wholesale network infrastructure;
• transit;
• Internet access;
• and downstream digital services.

If wholesale routing/interconnection conditions make effective downstream competition impossible, competition authorities could examine the conduct through an access-price or margin-squeeze framework.

Importance: Deutsche Telekom shows that telecommunications infrastructure can be assessed as a platform between upstream and downstream markets.

Case 4 — TeliaSonera

Case: Konkurrensverket v TeliaSonera Sverige AB, C-52/09.

Principle

The CJEU examined margin-squeeze conduct involving telecommunications infrastructure and emphasized that such conduct can constitute an independent form of abuse in appropriate circumstances.

BGP relevance

BGP infrastructure can similarly form an upstream input for downstream competitors.

For example:

Wholesale connectivity → ISP/Cloud/CDN service → End users

A dominant provider could potentially distort downstream competition by manipulating the economic conditions under which competitors obtain connectivity.

Importance: TeliaSonera is especially useful for analysing vertically integrated routing and connectivity markets.

Case 5 — Slovak Telekom

Case: Slovak Telekom a.s. v Commission, Joined Cases C-165/19 P and C-166/19 P.

Principle

The CJEU considered exclusionary conduct involving access to telecommunications infrastructure and the relationship between sector-specific regulation and Article 102 TFEU.

BGP relevance

This is highly relevant to modern routing governance because Internet connectivity frequently sits at the intersection of:

• telecommunications regulation;
• competition law;
• access obligations;
• infrastructure investment;
• and technical standards.

A regulated telecommunications operator cannot necessarily escape competition-law scrutiny simply because the conduct concerns regulated infrastructure.

Importance: Slovak Telekom demonstrates the importance of examining regulatory obligations together with competition-law principles.

Case 6 — Microsoft v Commission

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

Principle

The General Court considered Microsoft's refusal to provide interoperability information to competitors.

The case is a foundational authority concerning interoperability, technological ecosystems and exclusionary conduct.

BGP relevance

BGP itself is an interoperability protocol. Competition concerns can arise where a powerful ecosystem operator controls technical interfaces or information necessary for effective interoperability.

Relevant questions include:

• Who controls routing information?
• Are competitors given equivalent technical access?
• Are proprietary routing extensions used discriminatorily?
• Does a platform restrict interoperability with competing networks?
• Does technical incompatibility create artificial switching costs?

Importance: Microsoft demonstrates that technical interoperability can have direct competition significance.

Case 7 — Google Shopping

Case: Google and Alphabet v Commission, Case T-612/17.

Principle

The General Court upheld the Commission's finding concerning Google's preferential treatment of its own comparison-shopping service in search results.

BGP relevance

Although Google Shopping is not a routing case, its broader principle is important for self-preferencing through a platform-controlled infrastructure layer.

The analogy is:

Search infrastructure → preferential visibility

versus

Network infrastructure → preferential routing

If a dominant infrastructure operator systematically improves the effective technical conditions of its own downstream service while disadvantaging rivals, competition authorities may investigate whether the infrastructure is being used to distort downstream competition.

Importance: The case provides a modern framework for analysing infrastructure-mediated self-preferencing.

12. Regulatory Governance Versus Competition Governance

BGP governance creates an institutional problem.

Technical institutions may prioritize:

• security;
• stability;
• routing efficiency;
• resilience;
• engineering simplicity.

Competition authorities prioritize:

• competitive access;
• non-discrimination;
• consumer welfare;
• market contestability;
• prevention of exclusion.

These objectives can conflict.

Example

An ISP may claim:

"We prioritize routes according to technical reliability."

A competitor may respond:

"The alleged technical preference systematically disadvantages our service."

The authority must determine whether the routing preference is:

technically justified

or

commercially discriminatory.

13. Global Competition-Law Dimensions

European Union

Potential provisions include:

• Article 101 TFEU;
• Article 102 TFEU;
• EU telecommunications regulation;
• Digital Markets Act where its conditions are satisfied;
• merger control.

Article 102 is particularly relevant to:

• refusal to deal;
• discriminatory access;
• tying;
• self-preferencing;
• margin squeeze;
• exclusionary routing;
• and infrastructure foreclosure.

United States

Potential legal frameworks include:

• Sherman Act §1;
• Sherman Act §2;
• Clayton Act §7;
• telecommunications regulation;
• FCC authority;
• state competition laws.

Section 2 can become relevant where a dominant network deliberately uses routing or interconnection control to exclude competitors.

Section 1 becomes particularly relevant where independent networks coordinate commercially sensitive routing or access conditions.

United Kingdom

The principal framework includes:

• Competition Act 1998;
• Chapter I prohibition;
• Chapter II prohibition;
• Telecommunications regulation;
• Digital Markets, Competition and Consumers Act 2024.

The UK framework is particularly relevant to large digital infrastructure providers that combine:

cloud + backbone + CDN + platform + data-centre infrastructure.

India

Relevant institutions and frameworks include:

• Competition Act 2002;
• Competition Commission of India;
• Telecommunications Act 2023;
• TRAI regulatory framework;
• Indian Internet Exchange and interconnection ecosystem.

Potential competition concerns include:

• discriminatory interconnection;
• exclusionary network practices;
• refusal of access;
• vertical foreclosure;
• abuse of dominance;
• and coordinated conduct among network operators.

14. BGP and Algorithmic Competition

Modern BGP management increasingly involves automated systems.

Algorithms can determine:

• route selection;
• congestion avoidance;
• traffic engineering;
• capacity allocation;
• peering preferences;
• anomaly responses.

This creates a new competition-law question:

Who is legally responsible when an automated routing system repeatedly produces exclusionary outcomes?

A firm cannot necessarily avoid competition scrutiny merely by saying:

"The algorithm made the decision."

The relevant inquiry may instead examine:

1. who designed the algorithm;
2. what objectives were programmed;
3. what data it uses;
4. whether commercial preferences were encoded;
5. whether discriminatory outcomes were foreseeable;
6. whether monitoring systems detected them; and
7. whether corrective mechanisms existed.

15. Algorithmic Coordination Between Networks

BGP operators are competitors in some markets while cooperating technically in others.

That creates an unusual competition problem.

Operators may exchange information concerning:

• capacity;
• routing;
• outages;
• traffic;
• peering;
• network performance.

Legitimate operational coordination can become problematic if it evolves into coordination concerning:

• prices;
• customers;
• market allocation;
• capacity restrictions;
• refusal to supply;
• or commercially sensitive strategies.

Thus:

Technical coordination ≠ automatically cartel conduct

but

technical coordination can create a platform for commercially anticompetitive coordination.

16. Cybersecurity and Competition Law

Routing-security systems such as route validation can have competitive effects.

Technologies designed to prevent route hijacking may include:

• RPKI;
• route-origin validation;
• filtering;
• routing security policies.

These generally enhance competition by increasing reliability.

However, governance questions can arise if:

• one operator controls a critical validation mechanism;
• participation becomes discriminatory;
• security standards are manipulated to exclude legitimate competitors;
• or compliance requirements create disproportionate entry barriers.

The appropriate regulatory objective should therefore be:

security without unjustified competitive exclusion.

17. Merger Control and BGP Infrastructure

BGP infrastructure can become particularly significant during mergers.

Suppose two large networks merge.

The transaction could combine:

• autonomous systems;
• backbone capacity;
• data centres;
• IXPs;
• submarine cables;
• cloud connectivity;
• CDN infrastructure;
• and enterprise customers.

The competition authority should examine whether the merger creates:

Horizontal effects

Two transit providers become one.

Vertical effects

A backbone provider acquires a downstream cloud/CDN business.

Conglomerate effects

A platform combines routing infrastructure with several complementary digital services.

Network effects

The merged entity gains a substantially larger routing footprint.

18. Competitive Remedies

Potential remedies include:

Structural remedies

• divestiture;
• separation of network assets;
• divestiture of interconnection facilities.

Behavioural remedies

• non-discriminatory access;
• transparent peering criteria;
• fair interconnection;
• prohibition of self-preferencing;
• independent monitoring.

Technical remedies

• standardized APIs;
• routing transparency;
• route-validation requirements;
• interoperability obligations;
• audit logs.

Governance remedies

• independent technical oversight;
• competition-compliance committees;
• algorithmic auditing;
• incident reporting;
• separation of technical and commercial decision-making.

19. Key Competition-Law Tests for BGP Conduct

A useful analytical framework is:

Step 1 — Define the market

Is the relevant market:

• transit,
• peering,
• Internet access,
• CDN,
• cloud connectivity,
• enterprise connectivity,
• or a broader digital infrastructure market?

Step 2 — Establish market power

Consider:

• market share;
• AS connectivity;
• geographic coverage;
• capacity;
• customer base;
• network effects;
• switching costs;
• access to IXPs;
• and alternative routes.

Step 3 — Identify the routing conduct

Was there:

• refusal to peer;
• route withdrawal;
• discriminatory routing;
• inferior connectivity;
• excessive access pricing;
• self-preferencing;
• tying;
• or coordinated routing?

Step 4 — Establish foreclosure

Did the conduct:

• increase competitors' costs;
• reduce service quality;
• prevent customer access;
• increase latency;
• reduce reliability;
• or restrict market entry?

Step 5 — Examine justification

Possible legitimate explanations include:

• cybersecurity;
• congestion;
• reliability;
• capacity limitations;
• technical incompatibility;
• network stability.

Step 6 — Proportionality

Could the legitimate objective have been achieved through a less restrictive method?

20. Major Competitive Risks

21. The Central Legal Problem

The most difficult issue is that routing decisions are inherently technical.

Competition authorities should therefore avoid treating every difference in routing as discriminatory.

A proper investigation should ask whether:

the technical architecture has been deliberately or strategically configured so that a firm with market power can control the competitive conditions faced by rivals.

This requires technical evidence such as:

• BGP route tables;
• AS-path histories;
• routing-policy configurations;
• peering agreements;
• traffic measurements;
• latency data;
• packet-loss data;
• capacity information;
• algorithmic decision logs;
• and internal communications.

Conclusion

Global BGP governance sits at an increasingly important intersection between Internet architecture, telecommunications regulation, cybersecurity and competition law.

The strongest competition concerns arise where a network operator possesses substantial market power and uses routing or interconnection control to:

• deny indispensable access;
• discriminate against competitors;
• favour affiliated services;
• raise rivals' costs;
• restrict interoperability;
• coordinate commercially sensitive conduct;
• or create durable infrastructure-based entry barriers.

The leading jurisprudence—particularly Bronner, IMS Health, Deutsche Telekom, TeliaSonera, Slovak Telekom, Microsoft and Google Shopping—provides the conceptual tools for analysing these problems even though most were not directly decided on BGP manipulation.

The emerging principle is therefore:

BGP should remain technically decentralized and security-oriented, but the economic power embedded in routing infrastructure cannot be treated as competition-neutral when control over routes materially determines competitors' ability to access markets and consumers.