Interconnection is the physical and logical linking of separate telecommunications networks, enabling end-users on one network to communicate with those on another through the exchange of traffic, such as voice calls or data packets.[1][2] This process is mandated by regulatory frameworks in many jurisdictions, requiring carriers to interconnect directly or indirectly with competitors' facilities to promote competition and universal service, as exemplified by U.S. federal law under the TelecommunicationsAct of 1996.[1][2]In practice, interconnection facilitates peering agreements between networks of comparable size, where traffic is exchanged without payment, or paid transit for smaller networks accessing larger ones, underpinning the globalinternet's architecture by reducing latency and costs compared to publicinternetrouting.[3][4] Key challenges include disputes over interconnection fees and terms, which have led to legal battles and network outages, highlighting tensions between incumbents and new entrants in fostering open access while protecting infrastructure investments.[5] Beyond telecommunications, analogous principles apply in electric gridconnections for renewable energy generators, where delays in interconnection queues have bottlenecked cleanpower deployment amid rising demand.[6] These mechanisms underscore causal dependencies in networked systems, where interconnection not only drives efficiency but also exposes vulnerabilities to single points of failure or regulatory capture.
Definition and Fundamentals
Core Principles
Interconnection in telecommunications rests on the foundational principle of any-to-any connectivity, which requires that customers of one network can communicate with or access services on any other interconnected network, ensuring end-to-end interoperability without undue restrictions. This principle, enshrined in international frameworks like the WTO Reference Paper on Basic Telecommunications, underpins universal service by linking disparate networks physically or logically, often through points of interconnection (POIs) at exchanges or via transit arrangements.[7][8]A duty to interconnect obligates publicnetwork operators to provide access at any technically feasible point upon reasonable request, with terms that are fair, transparent, and non-discriminatory. Non-discrimination mandates equal treatment for all interconnecting parties, including no preferential conditions for an operator's affiliates or own services, to prevent incumbents from leveraging market power to stifle entrants; this is enforced through accounting separation and regulatory oversight of dominant firms.[7][9]Pricing adheres to cost-orientation, typically calculated using forward-looking long-run incremental cost (LRIC) or total service long-run incremental cost (TSLRIC) methodologies, coveringvariable and jointcostsplus a reasonable return on capital while avoiding cross-subsidies or excessive markups. This approach promotes efficient resource allocation and competition by aligning charges with underlying network expenses, with unbundled elements allowing entrants to pay only for required components.[7][9]These principles collectively advance competition by lowering entry barriers, such as through mandatory unbundling of networkelements and infrastructuresharing (e.g., co-location or duct access), while regulators provide ex ante guidelines, reference offers, and dispute resolution to resolve asymmetries between incumbents and new operators. Operational arrangements emphasize timelyservice provision and technical standards for seamless handover, like signaling system No. 7 (CCS7) for call routing.[7][8]
Economic Rationale
Interconnection enables telecommunications operators to provide seamless end-to-end connectivity between their subscribers and those on rival networks, which is essential for realizing network externalities and expanding marketdemand. In industries where services derive value from interoperability, such as telephony, isolated networks would fragment user bases and diminish overall utility, as customers expect universalaccess rather than siloed communication. Economic models demonstrate that interconnection increases socialwelfare by allowing calls or data exchanges that would otherwise be impossible, thereby boosting total traffic volumes and operator revenues through reciprocal access to external subscribers.[10][11]From a competitive standpoint, interconnection lowers entry barriers for new providers by permitting them to interconnect with incumbents' established infrastructure, avoiding the prohibitive costs of nationwide duplication. This promotes market entry and rivalry, which empirical evidence links to reduced retail prices and improved service quality; for example, post-liberalization in markets with mandated interconnection, consumer prices for fixed-line calls fell by an average of 20-40% in OECDcountries between 1990 and 2010. Incumbents may resist interconnection due to foreclosure incentives, leading to regulatory interventions that enforce nondiscriminatory terms, as seen in frameworks balancing long-run incremental costs for access to prevent exploitative pricing.[12][13]In bilateral negotiations, particularly for IP-based peering, operators pursue settlement-free arrangements when traffic volumes are balanced, minimizing transit costs and optimizing routing efficiency; unbalanced flows, however, often result in paid transit to compensate the terminating network for incremental capacity investments. This market-driven approach aligns with efficient resource allocation, though in traditional circuit-switched networks, two-way access charges can internalize externalities like congestion, enhancing welfare under certain conditions such as calling-party-pays regimes. Overall, interconnection mitigates hold-up problems inherent in complementary assets, fostering innovation and infrastructure sharing while averting inefficient network proliferation.[14][15]
Historical Development
Early Origins and Monopoly Formation
The telephone, patented by Alexander Graham Bell on March 7, 1876, initially operated via direct point-to-point lines between subscribers, limiting scalability without exchanges. By 1877, the first commercial telephone line connected Charles Williams Jr.'s home in Somerville, Massachusetts, to his Boston office, followed by the installation of the first switchboard and exchange in New Haven, Connecticut, in January 1878.[16] These early systems were proprietary and local, with no standardized interconnection between operators, as Bell's patent monopoly until 1894 discouraged rivals and fostered isolated "telephone islands."[17]In the United States, the American Telephone and Telegraph Company (AT&T) was incorporated on March 3, 1885, by the American Bell Telephone Company to construct and operate the nation's first long-distance network, initially linking New York to Philadelphia on May 11, 1885.[18] AT&T's control of long-distance trunk lines created a choke point, enabling it to dictate terms for traffic exchange; independent operators, proliferating after patent expiration in 1894, often faced refusal of interconnection unless they relinquished local exchanges to Bell affiliates.[17] Through aggressive acquisitions—buying over 100 independent exchanges between 1899 and 1907—and exclusive contracts, AT&T achieved near-monopoly status by 1907, controlling approximately 80% of U.S. telephones and suppressing competitive interconnection in favor of a unified "universal service" network.[19]In Europe, telephony emerged in the late 1870s via private concessions, often involving Bell subsidiaries, but governments increasingly asserted control by the early 1900s, establishing state-owned monopolies that internalized interconnection within national networks.[20] For instance, by 1913, countries like France and Italy operated under full state monopolies, while others such as the United Kingdom permitted limited private competition before nationalization trends dominated, reducing cross-operator peering in favor of centralized authority.[21] This pattern mirrored the U.S., where monopoly formation prioritized economies of scale and network integrity over fragmented interconnection, though empirical data from 1892–1914 shows private competition yielded higher mainline penetration (e.g., 1.2% of population in competitive vs. 0.6% in monopoly regimes) before state interventions curtailed it.[20]
Post-WWII Expansion and Initial Regulations
Following World War II, telecommunicationsnetworks expanded rapidly amid economic reconstruction and pent-up demand for connectivity. Nationaltelephone systems, often managed by government-owned postal, telephone, and telegraph administrations (PTTs), invested heavily in infrastructure, including automatic switching equipment and expanded long-distance circuits, reviving pre-war internationallinks disrupted by the conflict. Internationaltelephony circuits proliferated, with high-frequency radio and additional submarine cables enabling higher volumes of cross-border calls, as economic growth in Europe and Asia spurred traffic from levels suppressed during the war.[22]This expansion necessitated coordinated regulations to manage interconnections between sovereign monopolies, preventing disputes over traffic routing, settlement payments, and technical compatibility. The Bermuda Telecommunications Agreement, signed on December 4, 1945, between the United States and British Commonwealth nations, established foundational principles by capping international telegraph and telephone rates at a maximum of 30 cents per word and allocating revenue shares based on originating traffic, thereby facilitating US access to the extensive imperial cable network previously dominated by British interests.[23] The accord emphasized equitable division of facilities and circuits, setting a precedent for bilateral accounting rate agreements that governed most global interconnections in the immediate postwar era.[24]The International Telecommunication Union (ITU) further institutionalized these efforts through its 1947 International Telecommunication Conferences in Atlantic City, which revised the ITU Convention, integrated the organization as a United Nations specialized agency, and broadened its scope to foster equipment development and network interoperability.[25][26] The conferences strengthened the ITU's consultative committees—particularly the International Telegraph and Telephone Consultative Committee (CCITT)—to promulgate technical recommendations for interconnecting domestic networks, including signaling standards and tariff methodologies essential for handling surging international traffic without fragmentation.[27] In practice, these standards addressed compatibility issues in an era of analog systems, prioritizing reliable call completion over competition, as most nations retained PTT monopolies.[28]In the United States, the Federal Communications Commission (FCC) upheld interconnection mandates under the Communications Act of 1934, requiring carriers like AT&T to provide access for internationaltraffic while maintaining domestic monopoly structures, though enforcement focused on common carrier duties rather than mandating competitive entry.[29] Early postwar scrutiny also emerged on interconnection costing, with regulators examining embedded costs for pricingaccess to ensure fairness amid expansion, foreshadowing debates over fully distributed versus incremental methods.[30] These frameworks preserved stability but constrained innovation by limiting third-party attachments until subsequent rulings.[31]
Liberalization Era (1980s-2000s)
The liberalizationera in telecommunications, spanning the 1980s to the 2000s, marked a global shift from state-sanctioned monopolies to competitive markets, driven by neoliberal economic policies, technological advancements in digital switching and fiberoptics, and recognition that incumbent operators stifled innovation through restricted access. Policymakers in major economies imposed interconnection obligations on dominant carriers to enable new entrants to link their networks, fostering call completion across providers and reducing consumer costs; for instance, interconnection pricing models like fully distributed costs emerged to prevent incumbents from cross-subsidizing to exclude rivals. This period saw over 100 countries liberalize by 2000, with interconnection disputes resolved through regulatory arbitration in cases where incumbents delayed or priced access prohibitively high.[32][33]In the United States, the landmark divestiture of AT&T on January 1, 1984, following a 1982 antitrust settlement, dismantled the Bell System into seven regional holding companies for local services and separated AT&T's long-distance operations, mandating equal interconnection access for competitors like MCI and Sprint to the local loops. This breakup, overseen by the Modified Final Judgment, required the regional Bell Operating Companies (RBOCs) to provide nondiscriminatory interconnection at cost-based rates, spurring long-distance competition that reduced prices by over 50% within a decade, though initial disputes over access charges highlighted tensions between incumbents and entrants. The 1996 Telecommunications Act further codified interconnection duties, compelling incumbents to unbundle network elements for rivals, which facilitated entry but led to litigation over pricing methodologies like total element long-run incremental cost (TELRIC).[34][35][36]Europe's liberalization accelerated with the United Kingdom's privatization of British Telecom (BT) in November 1984, selling 50.2% of shares and licensing competitor Mercury Communications, which necessitated BT's interconnection at fair terms determined by the newly established Office of Telecommunications (OFTEL). The European Commission advanced harmonization through Directive 90/388/EEC in June 1990, abolishing special rights for telecom monopolies and promoting competition in services, followed by the Open Network Provision (ONP) Framework Directive in 1990 requiring transparent access and interconnection conditions across member states. By 1997, Directive 97/33/EC enforced mandatory interconnection at any technically feasible point, with pricing based on long-run average incremental costs, leading to full market opening by January 1, 1998, and a decline in incumbent market share from near 100% to under 70% in fixed lines by 2000.[37][38][39]Globally, the World Trade Organization's Agreement on Basic Telecommunications, concluded February 15, 1997, by 68 governments representing 90% of world telecom revenues, committed signatories to market access and national treatment for services, explicitly requiring public telecom operators to negotiate interconnection with any willing provider at technically feasible points and non-discriminatory rates. This reference paper on regulatory principles, annexed to schedules, emphasized cost-oriented pricing and independent regulators, influencing liberalization in developing markets and resolving disputes through WTO mechanisms, though implementation varied with some countries delaying full commitments until 2004. Interconnection thus became a cornerstone of pro-competitive regulation, enabling resale and facilities-based entry, but incumbents often leveraged their control over last-mile infrastructure to negotiate advantageous terms, prompting ongoing antitrust scrutiny.[40][33][41]
Regional Variations: United States
The United States telecommunications sector historically featured a dominant monopoly under the American Telephone and Telegraph Company (AT&T), which controlled approximately 80% of the market by the mid-20th century, limiting interconnection primarily through voluntary agreements or FCC-mandated common carrier obligations under the Communications Act of 1934. This act required carriers to provide reasonable interconnection for interstate services, but enforcement favored AT&T's integrated Bell System, restricting competitors like independent telephone companies to basic access at AT&T's terms, often resulting in disputes resolved via the Kingsbury Commitment of 1913 and subsequent FCC rulings.[42] By the 1970s, emerging long-distance rivals such as MCI challenged this structure through antitrust litigation, highlighting AT&T's refusal of equitable interconnection, which culminated in the Modified Final Judgment of 1982, effective January 1, 1984.The 1984 divestiture fragmented AT&T into the long-distance-focused AT&T Corporation and seven Regional Bell Operating Companies (RBOCs) handling local services, fundamentally altering interconnection dynamics by separating local loops from long-distance networks and mandating equal access for competitors to RBOC facilities.[43] This structural remedy, overseen by the FCC, reduced access charges from $0.15 per minute in 1984 to under $0.02 by 2000 and enabled microwave and fiber-optic entrants like MCI and Sprint to interconnect directly, fostering a competitive long-distance market that grew from two major providers in 1984 to dozens by the early 1990s.[35] However, local interconnection remained restricted, with RBOCs prohibited from entering long-distance or manufacturing, preserving silos until further reforms.The Telecommunications Act of 1996 marked a pivotal liberalization, imposing affirmative interconnection duties on incumbentlocalexchange carriers (ILECs) under Section 251 to negotiate in good faith with new entrants, including resale of services, unbundled access to networkelements like loops and switches, and physical or virtualcollocation for competitors.[44] The FCC's subsequent First Report and Order in 1996 implemented these provisions, requiring ILECs to offer interconnection at cost-based rates determined via state commissions or federalpricing bands, which spurred the entry of competitive localexchange carriers (CLECs) numbering over 100 by 1998, though litigation over pricing and unbundling—such as the Supreme Court's 1999 AT&T Corp. v. Iowa Utilities Board ruling—highlighted ongoing tensions between competition and ILEC investments.[45] By 2000, interconnection agreements exceeded 3,000, but CLECs captured only about 10% of local lines due to high costs and disputes, reflecting the US model's emphasis on antitrust enforcement over universal service mandates compared to European state-led transitions.[46]
Regional Variations: Europe
In the post-World War II era, European telecommunications operated under national state-owned monopolies managed by postal, telephone, and telegraph administrations (PTTs), with interconnection limited primarily to international settlements via bilateral agreements coordinated by the International Telecommunication Union (ITU). Domestic interconnection was minimal, as each country's PTT controlled end-to-end service provision, stifling competition and innovation while prioritizing universal service in rebuilding efforts. This structure persisted through the 1970s, with investments focused on expanding fixed-line networks but yielding inefficiencies such as long waiting lists for connections and high costs, as evidenced by varying penetration rates—e.g., 30-40 lines per 100 inhabitants in Western Europe by 1980.[47][48]The shift toward liberalization began in the late 1980s amid the European Economic Community's (EEC) push for a single market, prompted by technological convergence of telecom and IT sectors highlighted in the 1987Green Paper on the development of the commonmarket for telecommunications services. Initial directives targeted specific segments: the 1988liberalization of terminalequipment and value-added services, followed by the 1990 Services Directive (90/388/EEC), which required member states to remove exclusive rights for telecom services and enable alternative operators. Open Network Provision (ONP) frameworks, starting with Directive 90/387/EEC, mandated non-discriminatory access to publicnetworks, laying groundwork for interconnection by standardizing technical interfaces and supply conditions. These measures addressed monopoly distortions, where incumbents like France Télécom or Deutsche Telekom held over 90% market share, by imposing obligations for fair access pricing.[48][38][49]The 1990s accelerated interconnection-focused reforms, with the 1994 recommendation for full competition in fixed services by 1998 and the pivotal Interconnection Directive (97/33/EC) of June 30, 1997, which required operators with significant market power—typically incumbents—to provide interconnection at cost-oriented rates, including unbundling of local loops where necessary. This directive, building on competition principles from the Treaty of Rome, empowered national regulatory authorities (NRAs) to resolve disputes and enforce carrier selection, fostering entry by new providers; by 1998, EU-wide full liberalization eradicated legal monopolies, boosting fixed-line competition with over 200alternative operators in some markets like the UK. Country variations persisted: the UK led with British Telecom's 1984 privatization and early cable competition, while France and Germany delayed until 1998, reflecting nationalresistance to privatization amid concerns over service quality in rural areas. Empirical outcomes included rapidmobilegrowth—e.g., GSMadoption reaching 100 million users by 1998—and lower prices, though interconnection disputes, such as those over accounting separation, highlighted enforcement challenges under nascent NRAs.[50][47]Post-1998, interconnection evolved with the 2002 regulatory framework, emphasizing market reviews and remedies like mandatory reference offers, but historical legacies of state dominance influenced uneven implementation—e.g., higher interconnection charges in Southern Europe correlating with slower broadband rollout. This EU-driven model prioritized harmonized competition over unilateral national policies, contrasting with fragmented pre-liberalization silos, and supported causal links between mandated access and infrastructureinvestment, as new entrants invested €50 billion in pan-European networks by 2000.[48][51]
Technical Aspects
Physical and Logical Mechanisms
Physical interconnection refers to the tangible linking of separate networks through hardware components such as fiber optic cables, Ethernet ports, routers, and switches, often facilitated in colocation data centers or dedicated facilities.[4] This setup enables direct handoffs, where one network's demarcation point connects to another's via cross-connects, minimizing latency and dependency on intermediaries.[52] In Internet Exchange Points (IXPs), networks physically attach to a shared Layer 2 switch fabric, typically Ethernet-based, allowing multiple autonomous systems to interconnect within a single location like a data center.[53][54]Logical interconnection governs the exchange of control information and routing decisions, independent of the physical medium, to ensure seamless traffic flow between networks.[55] In packet-switched environments, the Border Gateway Protocol (BGP) serves as the primary mechanism for inter-domain routing, establishing TCP-based sessions on port 179 between peering routers to advertise reachability and select paths based on attributes like AS path length and local preferences.[56][57] BGP enables policy-driven decisions, where networks agree on peering terms to exchange traffic bilaterally without payments in settlement-free arrangements.[58]In traditional circuit-switched telecommunications, logical mechanisms relied on out-of-band signaling protocols like Signaling System No. 7 (SS7) to establish, manage, and tear down voice circuits across interconnected trunks, coordinating call routing via standardized point codes and global titletranslation.[7] Modern transitions to IP networks integrate these functions through Session Initiation Protocol (SIP) for call control and Diameter for authentication, bridging legacy and next-generation architectures while preserving logical interoperability.[55] These mechanisms collectively ensure end-to-end connectivity, with physical layers handling bit transmission and logical layers enforcing routingintegrity and quality of service.[7]
Protocols, Standards, and Agreements
Interconnection between telecommunications networks primarily utilizes the Signaling System No. 7 (SS7) protocol suite for enabling call setup, routing, and management in circuit-switched environments like the public switched telephone network (PSTN). SS7 operates as an out-of-band signaling system, separating control signals from bearer channels to support efficient interconnection across operators' networks, with core components including the Message Transfer Part (MTP) for transport layer reliability and the Transaction Capabilities Application Part (TCAP) for database queries. The ISDN User Part (ISUP), a key SS7 application, handles circuit-switched connectioncontrol, specifying messages for initialaddress messages, answer signals, and release procedures to ensure seamless call handover between interconnected switches.[59] These protocols, developed in the 1970s and standardized by the International Telecommunication Union Telecommunication Standardization Sector (ITU-T) in the Q.761–Q.764 series, remain foundational for voice interconnection despite vulnerabilities like those exploited in SS7 attacks.[60]ITU-T Recommendations form the backbone of global standards for interconnection, defining technical interfaces, performance metrics, and resource designations to promote interoperability. Recommendation M.1400, last updated in 2023, outlines standardized designations for interconnection points, such as trunk groups and signaling links, to facilitate consistent network linking among operators.[61] Complementary standards include E.164 for international numbering to enable global routing and Q.703 for SS7 signaling links, ensuring reliable data exchange at rates up to 64 kbit/s per link. Regional bodies like the European Telecommunications Standards Institute (ETSI) adapt these, as in ES 200 677 for PSTN attachment requirements, while national regulators mandate compliance in interconnection offers.[62]Interconnection agreements between operators operationalize these protocols and standards through contractual terms specifying physical handoff points (e.g., E1/T1 interfaces at 2.048/1.544 Mbit/s), signaling configurations, and quality metrics like call setup delay under 5 seconds.[63] These bilateral or multilateral pacts, often approved by regulators under frameworks like the U.S. Federal Communications Commission's 47 CFR Part 51, include provisions for traffic settlement based on minutes of use and adherence to ITU-T error rates below 1 in 10^7 for signaling links.[1] In international contexts, the WorldTrade Organization's 1997 Agreement on BasicTelecommunications incorporates interconnection principles, requiring fair terms and non-discriminatory access to promote competition.[7] For emerging IP-based peering, agreements reference IETF standards like BGP for routing but traditionally emphasize SS7/ISUP for voice to maintain circuit reliability.[64]
Evolution to IP and Data Networks
The transition from circuit-switched public switched telephone networks (PSTN) to packet-switched IP-based networks fundamentally altered interconnection practices in telecommunications. In circuit-switched systems, interconnection involved dedicated physical or logical paths for voice calls, often mandated by regulation to ensure universal access, as seen in frameworks like the U.S. Modified Final Judgment of 1982 that required AT&T to interconnect with competitors. Packet switching, conceptualized in the early 1960s by researchers including Paul Baran for survivable communications and independently by Donald Davies, enabled dynamic allocation of bandwidth, prioritizing data efficiency over fixed reservations.[65]Leonard Kleinrock's 1961 paper and 1964 book formalized queuing theory for packet networks, laying groundwork for scalable data transmission.[66]By the 1980s, the adoption of the TCP/IPprotocolsuite—standardized in 1983 for ARPANET—facilitated the emergence of interconnected datanetworks, contrasting with PSTN's voice-centric model.[66] ARPANET's first packet-switched link in 1969 evolved into NSFNET in 1985, which connected research institutions and spurred academic-commercial peering.[67] The commercialization of the internet following NSFNET's decommissioning in 1995 shifted interconnection toward market-driven models: IP peering, where networksexchangetraffic directly without payment if balanced, and IP transit, where providers pay upstream carriers for broader access.[68]Peering agreements proliferated via neutral access points like MAE-East (established 1992) and commercial internet exchanges, reducing latency and costs compared to hierarchical transit.[69] This voluntary, bilateral approach diverged from regulated PSTN interconnection, emphasizing mutual benefit over compulsory access, though disputes over "fair" peering terms arose as traffic imbalances grew with content providers.[70]Telecommunications operators began migrating to next-generation networks (NGN) in the late 1990s, integrating voice over IP (VoIP) and data services on unified packet platforms to lower costs and enable multimedia.[71] ITU-T defined NGN architecture in recommendations like Y.2001 (2004), supporting IP multimedia subsystems (IMS) for seamless interconnection via session initiation protocol (SIP).[72] This evolution introduced IP-to-IP interconnection, where carriers linkcorenetworks for call routing without circuitemulation, as in early VoIP deployments by operators like NTT in Japan by 2000.[73] Regulatory challenges emerged, including debates over applying legacyaccess fees to IP traffic; for instance, the FCC's 2004 ruling exempted VoIP from traditional intercarrier compensation, promoting market-based settlements.[74] By the 2010s, widespread PSTN sunsets—such as BT's UK migration plan starting 2015—accelerated all-IP interconnection, with standards ensuring quality-of-service via protocols like RTP for real-time data.[75] Empirical data from migrations showed cost reductions of up to 50% in transport expenses due to packet efficiency, though initial investments in fiber backbones exceeded $100 billion globally by 2010.[76]
Regulatory Frameworks
Core Regulatory Principles
Core regulatory principles for telecommunications interconnection mandate that network operators link their systems to enable end-to-end communication across disparate networks, fostering competition while preventing abuse of market power. These principles, rooted in international frameworks such as the WorldTrade Organization's Reference Paper on BasicTelecommunications, require operators to provide interconnection upon reasonable request at technically feasible points, ensuring seamless servicedelivery without undue barriers.[77][7]A fundamentalobligation is non-discrimination, whereby interconnection terms offered to one party must be equivalent to those provided to others in similar circumstances, prohibiting preferential treatment that could entrench incumbents' dominance.[1][78]Pricing must be cost-oriented, reflecting the incremental costs of interconnection plus a reasonable return, rather than markup-driven rates that subsidize other services; this aligns with ITU guidelines emphasizing efficiency and avoids cross-subsidization.[7][79]Operators are required to negotiate in good faith, with terms that are transparent, reasonable, and publicly disclosed where necessary to promote market entry.[2][1] In cases of impasse, regulators intervene for bindingresolution, as seen in U.S. law under Section 251 of the Communications Act, which imposes duties on all carriers but heightens requirements for incumbents to unbundle elements and collocate equipment.[2] European directives similarly enforce access obligations to achieve an open internal market, prioritizing interoperability standards.[78]These principles extend to evolving IP-based networks, where logical and physical linking must support voice, data, and emerging services without legacy biases, though debates persist on applying traditional mandates to internet peering, which often relies on voluntary commercial agreements rather than regulation.[80] Enforcement varies by jurisdiction, with independent regulators ensuring compliance to balance innovation incentives against free-riding risks.[81]
United States Approaches
In the United States, telecommunications interconnection is primarily governed by the Federal Communications Commission (FCC) under Title II of the Communications Act of 1934, as amended by the Telecommunications Act of 1996, which mandates that incumbentlocalexchange carriers (ILECs) provide interconnection to requesting telecommunications carriers for the transmission and routing of telephone exchangeservice and exchangeaccess.[1]Section 251(c)(2) requires ILECs to interconnect directly or indirectly at any technically feasible point within their networks, using protocols compatible with their networks, and on rates, terms, and conditions that are just, reasonable, and nondiscriminatory.[44] This obligation aims to facilitate competition in local markets by enabling competitive localexchange carriers (CLECs) to originate and terminate calls on ILEC networks, with interconnection agreements negotiated between parties or arbitrated by state public utility commissions if negotiations fail.[82]Pricing for interconnection under these rules is typically cost-based, calculated using forward-looking economic cost methodologies such as Total Element Long-Run Incremental Cost (TELRIC), which estimates the cost of providing the service over the long term without including sunk costs from legacyinfrastructure.[45] The 1996 Act's "competitive checklist" further specifies that interconnections must enable competitors to access network elements on an unbundled basis, resell services at wholesale rates, and achieve parity in quality and speed comparable to the ILEC's own services.[83] For interstate services, the FCC oversees compliance, while intrastate matters fall to state commissions, though federal preemption applies where state rules conflict with competition goals.[1]In contrast to circuit-switched voice networks, IP-based interconnection for broadband and internet services operates largely on a market-driven basis without mandatory obligations, as broadband internet access service is classified as an information service rather than a telecommunications service under FCC forbearance policies.[84] Peering agreements, where networks exchange traffic bilaterally without payment, and paid transit arrangements predominate, resolved through private contracts rather than regulation, though the FCC has authority to investigate disputes under Section 201(b) if they involve unjust or unreasonable practices.[85] Recent FCC actions, including a July 2025 notice proposing forbearance from time-division multiplexing (TDM) interconnection mandates under Sections 251(c)(2) and (c)(6), seek to accelerate the transition to all-IP networks by reducing legacy regulatory burdens on ILECs while preserving competition through voluntary IP interconnection options.[86] This shift reflects empirical evidence that mandatory unbundling has sometimes discouraged investment in fiber infrastructure, with studies showing higher deployment rates in forborne markets.[84]
European Union Policies
The European Union's policies on telecommunications interconnection form part of a harmonized regulatory framework designed to promote competition, ensure non-discriminatory access to networks, and facilitate the development of a single market for electronic communications services.[87] This framework originated with early directives such as Directive 97/33/EC, which established initial rules for securing interconnection between public telecommunications networks, emphasizing fair and reasonable terms to prevent anti-competitive practices by dominant incumbents.[78] Subsequent reforms, including the 2002 regulatory package, refined these obligations through Directive 2002/19/EC on access to, and interconnection of, electronic communications networks, which mandated that national regulatory authorities (NRAs) impose specific duties on operators with significant market power (SMP), such as meeting reasonable interconnection requests, providing necessary technical information, and ensuring service interoperability.[88]Under the current regime, codified in the European Electronic Communications Code (EECC) via Directive (EU) 2018/1972, interconnection obligations apply symmetrically where necessary, but primarily target SMP operators, requiring them to negotiate in good faith and offer interconnection at any technically feasible point in their network, including unbundled access to local loops. Pricing must be cost-oriented, typically based on long-run average incremental cost (LR AIC) or long-run incremental cost (LRIC) methodologies to reflect efficient provision, with NRAs empowered to approve reference offers that detail terms, charges, and quality standards.[87] The EECC, transposed by member states by December 21, 2020, introduces enhancements like obligations for co-investment in next-generation networks and greater regulatory symmetry to encourage infrastructure sharing, while the Body of European Regulators for Electronic Communications (BEREC) oversees consistent implementation across the EU to mitigate fragmentation.[89] Dispute resolution remains a core mechanism, with NRAs able to intervene swiftly in interconnection disagreements, imposing remedies if operators fail to comply, as reinforced by amendments in Directive 2009/140/EC to the Framework Directive 2002/21/EC.[90]These policies prioritize ex-ante regulation in markets prone to SMP, such as wholesale call termination and fixed access, with periodic market reviews every three to five years to assess competition and lift obligations where effective rivalry emerges.[91] Interconnection extends to mobile networks via similar SMP-based duties, ensuring number portability and fair transit arrangements, though enforcement varies by member state due to national implementation, leading BEREC to issue guidelines on SMP assessment since 2021 to promote uniformity.[89] While aimed at fostering investment and consumer choice, the framework has evolved to address IP-based transitions, mandating fair access to internet exchange points and peering arrangements where imbalances threaten competition, without imposing universal service obligations on interconnection itself.[92]
Global and Emerging Market Practices
The International Telecommunication Union (ITU) establishes foundational principles for global telecommunications interconnection through the International Telecommunication Regulations (ITRs), which promote interoperability and non-discriminatory access among networks worldwide.[93] ITU-T Recommendation M.1400 specifies standardized designations for interconnection points between operators' networks, facilitating technicalcompatibility and operational efficiency across borders.[61] These frameworks emphasize transparent agreements, cost-oriented pricing, and dispute resolution mechanisms to supportcompetition, as outlined in ITU guidelines that prioritize fairaccess over market dominance by incumbents.[7]In emerging markets, interconnection practices often adapt ITU principles to local contexts, mandating obligations on dominant operators to enable new entrants amid uneven infrastructure development. The World Bank's Telecommunications Regulation Handbook highlights that effective interconnection reduces entry barriers, boosts investment, and expands service coverage, though enforcement varies due to regulatory capacity constraints.[94]In India, the Telecom Regulatory Authority of India (TRAI) enforces the Telecommunication Interconnection Regulations, 2018, requiring service providers to negotiate agreements within specified timelines, provide bank guarantees for compliance, and adhere to technical standards for points of interconnection.[95] TRAI's framework mandates cost-based interconnection usage charges (IUC), with recent consultations in 2025 proposing amendments to streamline dispute resolution and address broadcasting-cable interconnections, aiming to curb delays that hinder smaller operators.[96][97]Brazil's NationalTelecommunicationsAgency (ANATEL) permits negotiated interconnection fees under ResolutionNo. 600 of 2012, which defines five classes of interconnection including directnetworklinks for voice and data services, with ANATEL intervening in disputes to ensure non-discriminatory terms.[98][99] This approach supports competition in a market where fixed and mobile operators invested BRL 34.6 billion in 2023, though regulatory reforms in 2025 focus on modernizing rules to align with 5G rollout without altering core interconnection mandates.[100]In China, the Ministry of Industry and Information Technology (MIIT) mandates interconnection among state-dominated operators like China Telecom and China Mobile to prevent redundant infrastructure, as required since at least 2018 policies promoting efficient resource use.[101] However, limited private entry and centralized control result in fewer disputes but potentially reduced incentives for innovation, contrasting with more liberalized models elsewhere.Across African emerging markets, regulators like those in the Southern African Development Community (SADC) align interconnection rates using bottom-up long-run incremental cost (BU-LRIC) models, as implemented in Tanzania to promote fairpricing amid low broadband penetration.[102][103] Cross-border challenges persist, with initiatives like the African Union's policies urging harmonized standards to lower traffic flow barriers, though uneven enforcement in regions with sparse infrastructure often favors incumbents.[104] These practices reflect a balance between fostering competition and addressing developmental gaps, with empirical evidence showing interconnection mandates correlating to higher mobile adoption rates in liberalizing economies.[105]
Economic Impacts
Competition and Consumer Benefits
Network interconnection in telecommunications enables competing providers to exchange traffic efficiently, thereby lowering barriers to entry for smaller operators who can leverage established infrastructure rather than duplicating costly last-mile facilities. This access promotes market entry, as evidenced by U.S. data from 1995 to 2004, where approved interconnection service requests correlated with increased competitive entry into local markets, fostering a more fragmented provider landscape.[106] By facilitating peering and transit arrangements, interconnection reduces the dominance of incumbents, allowing content providers and edge networks to route traffic without prohibitive fees, which in turn intensifies rivalry among backbone and access providers.Consumers derive direct benefits from heightened competition, including reduced prices and expanded service options. In mobile markets, interconnection regimes have structured tariffs to minimize double marginalization—where each network adds markup on terminated calls—resulting in equilibrium retail prices closer to marginal costs and increased total surplus.[107] Empirical analyses of interconnection benchmarking across jurisdictions indicate that aligning termination rates nearer to forward-looking economic costs stimulates demand through lower end-user pricing, as observed in studies projecting price reductions from rate adjustments implemented between 2024 and 2029.[108] For internet services, peering policies that balancetrafficexchange without paid imbalances enhance connectivityquality, reducing latency and enabling broader access to diverse content, which indirectly pressures providers to innovate and lower broadband fees amid competitive pressures.[109]These dynamics extend to infrastructuresharingvariants of interconnection, which yield cost efficiencies passed to consumers via improved coverage and quality without proportional price hikes. For instance, mobilenetworksharinghas been linked to accelerated investment returns and lower tariffs, benefiting users in underserved areas by expanding effective competition without full-scale duplication.[110] Overall, while interconnection does not eliminate all market power asymmetries, its role in enabling scalable rivalry consistently correlates with tangible gains in affordability and choice, as liberalized regimes post-1990s deregulation demonstrate through widespread tariff declines and serviceproliferation.[9]
Investment Incentives and Market Distortions
Mandatory interconnection requirements and access pricing rules, such as those mandating unbundled network elements at regulated costs, often diminish incentives for infrastructureinvestment by allowing entrants to leverage incumbents' facilities without incurring equivalent build-out expenses. This shifts competition toward resale models, where providers arbitrage regulated access rates rather than deploying their own networks, leading to less efficient resource allocation and reduced innovation in network upgrades. Empirical evidence from telecommunications markets indicates that such policies create a trade-off: while they facilitate short-term entry, they systematically lower total industryinvestment levels by undermining the returns needed to justify high-risk capital expenditures.[111][112]In the United States, the Telecommunications Act of 1996 required incumbents to provide interconnection and unbundled elements at Total Service Long Run Incremental Cost (TSLRIC), which critics argue distorted markets by enabling competitive local exchange carriers (CLECs) to focus on leasing rather than facilities-based entry. Post-enactment data show initial CLEC investments peaking around $10 billion annually by 1999, but much of this involved resale of unbundled loops rather than new infrastructure, culminating in widespread CLEC bankruptcies after 2000 when access pricing reforms limited such arbitrage. This resale-heavy approach contributed to underinvestment in broadband facilities, with U.S. fixed-line investment declining relative to unregulated segments post-1996.[112][113][114]European Union policies emphasizing local loop unbundling under the 2002 regulatory framework similarly fostered service-based competition but empirically reduced incentives for next-generation access (NGA) deployments, such as fiber-to-the-home. Cross-country analyses reveal that countries with stricter unbundling obligations experienced 10-20% lower NGA investment shares of GDP between 2005 and 2010 compared to those with lighter regulation, as alternative operators delayed self-provisioning in favor of leasing aging copper infrastructure. Ownership unbundling mandates further exacerbated this by increasing incumbents' cost of capital, leading to aggregate investment shortfalls in fixed networks.[111][115][116]Comparisons of facilities-based versus resale entry underscore these distortions: facilities-based models drive higher capital expenditures and service quality improvements, as operators internalize upgrade costs and capture full returns, whereas resale relies on incumbents' investments without proportional contributions from access-seekers. In mobile markets, low regulated termination rates—intended to curb calling party pays distortions—have been linked to reduced network capacity expansions in some regimes, though bill-and-keep pricing can mitigate this by aligning incentives better with traffic volumes. Overall, these regulatory interventions risk chronic underinvestment, constraining capacity amid rising data demands and elevating future upgrade costs borne by ratepayers.[117][118][119]
Controversies and Criticisms
Pricing Disputes and Compensation Models
Pricing disputes in telecommunications interconnection typically arise from disagreements over the allocation of costs for traffic termination, origination, and transit between networks, particularly when traffic volumes are imbalanced. Compensation models vary by jurisdiction and service type; common approaches include bill-and-keep arrangements, where neither party compensates the other for terminating calls or data, and usage-based fees such as termination charges, where the originating network pays the receiving network a per-minute or per-unit rate.[7] Cost-oriented models, like TotalService Long Run Incremental Cost (TSLRIC), aim to recover only the incremental costs attributable to interconnection, as recommended by bodies like the ITU and implemented in reforms by the U.S. Federal Communications Commission (FCC).[120] These models seek to promote efficiency but often spark contention when operators dispute the underlying cost calculations or traffic ratios.In voice telephony, disputes frequently center on access charges, which historically favored incumbent networks through high termination fees subsidized by long-distance callers. The FCC's 2011 intercarrier compensation reform phased down these rates from an average of $0.0007 per minute to bill-and-keep for most traffic by 2018, addressing arbitrage schemes where carriers routed calls to exploit high payouts, but this transition led to litigation over recovery mechanisms for affected rural carriers.[121] Internationally, similar issues prompted WTO disputes; for instance, in 2004, the U.S. prevailed against Mexico over Telmex's anticompetitive international interconnection rates, which exceeded costs by up to 300% due to monopoly control, resulting in mandated nondiscriminatory pricing.[122] Regulators often intervene via arbitration, as in Jordan's 2003 case where the Telecommunications Regulatory Commission resolved fixed-to-mobile termination fee disputes by imposing cost-based rates after operators failed to agree.[123]For IP-based datanetworks, pricing tensions intensify due to asymmetric traffic flows from contentdeliverynetworks (CDNs) and edge providers, challenging traditional settlement-free peering norms. Operators receiving disproportionate inbound traffic, such as ISPs serving end-users, demand paid peering or transit fees to covercapacity costs, while content providers advocate bill-and-keep to avoid subsidizing infrastructure. A prominent 2024 example occurred in Germany, where Meta Platforms severed its peeringlink with Deutsche Telekom in September, citing refusal to compensate for the heavy one-way data volumes from services like Facebook and WhatsApp, which strained DT's network without reciprocal flows; DT countered that Meta'sscale necessitated payment for utilized resources, highlighting broader debates on "fair share" contributions in internet ecosystems.[124] Such breakdowns can degrade service quality until renegotiated, underscoring causal links between traffic imbalances and incentives for networks to extract value from high-demand senders, often resolved through private settlements or regulatory oversight rather than mandated models. Empirical analyses indicate these disputes correlate with rising datademands, with peering conflicts increasing 20-30% annually in major markets since 2020 due to streaming and cloud growth.[125]
Free-Riding on Infrastructure
Free-riding on infrastructure in network interconnection arises when parties, particularly large over-the-top (OTT) content and application providers (CAPs), are accused of leveraging the capital-intensive investments of access network operators—such as broadband and mobileinfrastructure—without bearing a proportionate share of the associated costs, especially amid asymmetric traffic volumes where CAPs originate far more data than they terminate.[126] Telecom operators contend that this imbalance stems from explosive growth in data-intensive services like video streaming, which necessitate ongoing upgrades to last-mile networks funded primarily through consumer subscriptions, while CAPs capture most monetization from end-users.[127] For instance, European telecom firms have highlighted that a handful of dominant OTT platforms generate the bulk of internet traffic, exacerbating network strain without direct contributions to deployment costs for fiber and 5G infrastructure.[128]Proponents of the free-riding narrative, including industry groups like the GSMA, argue that settlement-free peering arrangements—common in interconnection—fail to account for these disparities, allowing high-traffic generators to externalize upgrade expenses onto access providers facing thin margins amid regulated pricing and competition.[129] A prominent example is the 2014 interconnection dispute between Netflix and Comcast, where Netflix's surging video traffic led to congestion on Comcast's network via upstream transit providers like Cogent and Level 3; Netflix ultimately entered a paid peering deal directly with Comcast on February 23, 2014, to bypass bottlenecks and ensure quality of service, effectively compensating for the disproportionate inbound traffic load.[130] Similar tensions have surfaced in Europe, where operators like Deutsche Telekom and Orange have lobbied for "fair share" mechanisms to mandate contributions from CAPs toward gigabit network rollouts, citing data volumes from platforms like YouTube and Netflix as drivers of capex needs exceeding €500 billion EU-wide by 2030.[131]Critics, including EU regulators, counter that no empirical evidence supports systemic free-riding, as CAPs invest substantially in content delivery networks (CDNs) and edge caching to localize traffic and minimize backbone strain, while consumer demand—not unilateral imposition—drives usage patterns.[132] The Body of European Regulators for Electronic Communications (BEREC) has repeatedly affirmed in reports, such as its October 2022 analysis, that large traffic generators also bear interconnection costs through private investments and negotiated agreements, rejecting claims of exploitation as overstated given the mutual benefits of open internet ecosystems.[133] In the U.S., post-dispute resolutions like Netflix-Comcast have leaned on commercial negotiations rather than regulation, with the FCC declining to classify such paid peering as net neutrality violations, preserving market-driven outcomes despite telco assertions of imbalance.[134] These debates underscore tensions between infrastructure funders seeking cost recovery and innovation advocates wary of distorting voluntary interconnection, with unresolved EU proposals as of 2024 deferring mandates amid concerns over higher consumer prices and reduced access.[135]
Regulatory Overreach and Innovation Suppression
The Federal Communications Commission's 2015 Open Internet Order reclassified broadband internet access service under Title II of the Communications Act, subjecting providers to common carrier regulations that encompassed interconnection obligations and heightened regulatory scrutiny over peering arrangements. This shift introduced uncertainty and compliance burdens, correlating with a decline in U.S. broadband capital expenditures from $78.9 billion in 2014 to $75.1 billion in 2015, with further drops to $74.5 billion in 2016, before stabilizing amid signals of repeal in 2017.[136][137] Industry analyses attribute this to diminished incentives for network upgrades and innovative peering models, as providers anticipated price controls, unbundling mandates, and interventions in private traffic exchange agreements.[138] For instance, AT&T suspended nationwide fiber optic investments in November 2014 amid regulatory threats, citing the chilling effect on long-term planning.[139]Such mandates disrupted market-driven interconnection, where networks negotiate terms based on traffic volumes and costs, potentially reimposing outdated tariffs that favor content providers over infrastructure owners. Critics, including free-market policy experts, contend this overreach stifles innovation by discouraging specialized services like usage-based pricing or prioritized peering, as seen in Google's 2012 reconsideration of expanding voice services under similar rules due to regulatory hurdles.[140] Empirical reviews link Title II's framework to reduced entry by new competitors and slower adoption of advanced technologies, as bureaucratic approvals replace agile commercial decisions.[141]In the European Union, interconnection requirements under the 2002 Framework Directive and subsequent updates have imposed ex ante obligations on dominant operators, mandating cost-oriented access that fragments national markets and deters cross-border scaling. A 2025 analysis identified at least 10 divergent implementation approaches across member states, complicating the deployment of innovative services such as 5G network slicing and low-latency peering for edge computing.[142] This regulatory heterogeneity raises compliance costs and uncertainty, contributing to Europe's lag in fiber rollout— with only 48% coverage by 2024 compared to higher U.S. rates— as operators withhold investments amid fears of mandated free-riding by competitors.[143] Proponents of deregulation argue that such overreach supplants voluntary bilateral agreements with administratively determined terms, reducing incentives for infrastructure upgrades and favoring incumbents who exploit access rights without reciprocal contributions.[144]Globally, interconnection overreach manifests in compelled settlements that distort traffic economics, as evidenced by disputes where regulators intervene in private peering pacts, potentially embedding legacy models ill-suited to IP-based networks. Economic models demonstrate that access regulations can yield socially suboptimal outcomes by underincentivizing upstream investments, with empirical studies showing reduced entry and innovation when interconnection is not left to market negotiation.[145] While intended to promote competition, these interventions often entrench inefficiencies, as seen in historical U.S. telecom arbitrations under Section 252, where state-mandated terms prolonged litigation and delayed network enhancements.[146] Truth-seeking assessments prioritize data from industry filings and econometric analyses over advocacy claims, revealing a pattern where regulatory expansion inversely correlates with dynamic efficiency in interconnection ecosystems.[147]
Modern Developments and Future Outlook
Integration with 5G and Mobile Networks
5G networks introduce virtualization and features like network slicing, which partition physical infrastructure into isolated logical networks for diverse applications, necessitating updated interconnection protocols to ensure end-to-end servicecontinuity across mobile operators.[148]Network slicing supports use cases ranging from enhancedmobile broadband to ultra-reliable low-latency communications, but cross-operator implementation demands standardized definitions for slice attributes, such as latency and throughput guarantees, to facilitate peering and roaming.[149]The GSMA's Generic Slice Template (GST) and Network Slice Type (NEST) provide frameworks for describing slice characteristics, enabling operators to negotiate interconnection and roaming agreements that maintain quality of service (QoS).[149] For instance, NEST documents are used in bilateral roaming pacts to align slice parameters, addressing the complexity of 5G Standalone (SA) cores where signaling and data paths must interconnect seamlessly.[150] ITU-T Recommendation Q.4073 (June 2024) outlines definitions for 5G access networks in interconnection contexts, including roaming interfaces aligned with GSMA specifications.[151]Challenges in 5G interconnection include preserving slice isolation during traffic exchange to prevent interference or security breaches, as denial-of-service attacks could target specific slices without affecting others. End-to-end slicing across networks requires incentive-compatible commercial agreements, given the variability in operator capabilities and the need for coordinated orchestration.[152] Security protocols, such as those in GSMA's Network Equipment Security Assurance Scheme (NESAS) developed with 3GPP, aim to mitigate risks in interconnected 5G environments.[153]In Europe, 5G interconnection integrates with EU policies emphasizing spectrum harmonization and infrastructure sharing to accelerate deployment, with the GSMA forecasting 57% of mobile connections on 5G by 2030.[154] Regulatory frameworks like the European Electronic Communications Code indirectly support slice-aware interconnection by mandating fair access, though specific adaptations for 5G slicing remain under development amid calls for international alignment.[155] As of 2025, trials and commercial 5G SA roaming demonstrate progress, but fragmented national implementations pose hurdles to ubiquitous interconnection.[150]
Internet Peering and Exchange Points
Internet peering refers to the direct, typically settlement-free interconnection between autonomous systems, enabling the mutual exchange of traffic without reliance on upstream transit providers. This arrangement optimizes routing by allowing networks to bypass intermediaries, thereby reducing latency and transit costs. Internet exchange points (IXPs) serve as the physical infrastructure facilitating this peering, consisting of shared switching fabrics in data centers where multiple networks connect via common ports to exchange data efficiently.[156][157][158]In modern internet architecture, IXPs play a pivotal role in enhancing interconnection resilience and performance amid surging data demands from cloud services, content delivery networks, and edge computing. By localizing traffic exchange, IXPs minimize round-trip times and dependency on distant transit paths, which is critical for applications requiring low latency, such as real-time video streaming and financial transactions. For instance, peering at IXPs has been shown to reduce data exchange costs by up to 60% in regions like Europe, according to analysis of large-scale network data. This efficiency supports scalable interconnection, with global IXP traffic volumes continuing to grow as networks prioritize direct connections over paid transit to handle exponential increases in internet usage.[159][160][161]Recent developments underscore IXPs' evolution as critical infrastructure, particularly in addressing vulnerabilities exposed by events like the 2025 Portugal blackout, where traffic dropped by up to 90% due to IXP disruptions. In emerging markets, such as Africa, the number of IXPs has expanded rapidly, with a 2024 study noting over 50 operational points promoting local traffic retention and reducing international bandwidth expenses. Trends include a shift toward enterprise-focused IXPs tailored for AI workloads and 5G integration, enabling dynamic peering for edge AI processing, where the market grew from $49.3 billion in 2024 to $53.5 billion in 2025. However, challenges persist, including risks from traffic concentration at major hubs and "zombie" IXPs operating without innovation, prompting calls for resilience enhancements like diversified routing and recognition as essential facilities.[162][163][164]Looking ahead, IXPs are poised to underpin future interconnection by supporting decentralized architectures that mitigate single points of failure and foster AI-driven network optimization. Expanded deployment in underserved regions and integration with software-defined networking could further democratize peering benefits, though economic pressures favoring consolidation may heighten outage risks without proactive diversification. Sustained growth in IXP participation remains essential for accommodating projected internet traffic surges, ensuring robust, cost-effective global connectivity.[165][166][167]