The Network Operator in the Changing Telecom Value Chain: Seeking Scarcity and Control

The vertical industry unbundling that has happened in the mobile handset OEM industry is also ongoing in the network operator industry. The clockspeed (Fine, 1998) of the mobile network industry appears to be much lower than that of the handset industry — the unbundling process is slower, and incomplete.

Whereas in the pre-liberalisation period telecom operators frequently played an important role in the research, development and design of telecoms equipment, by the 1990s they had almost completely exited from these activities, leaving them to the specialised equipment suppliers. The best-known case was AT&T which vertically integrated a telecoms equipment business and a network operation until the spin-off of Bellcore (later Telcordia, being acquired by Ericsson) in the 1980s and the equipment business Lucent (including Bell Labs, now part of Alcatel-Lucent) in the 1990s. With the break-up of the vertical integration between layers 1 and 2 of (Fransman 2002)’s value chain (link), R&D intensity was moved almost entirely to the telecoms equipment companies in layer 1, with the exception of NTT (and later DoCoMo) in layer 2.

ETSI’s Groupe Spéciale Mobile (GSM), established in 1988, was an environment suited to realize the initial vision of active coordination between business and policy stakeholders to synchronize disjointed innovations into a systemic innovation (Kano, 2000). Equipment suppliers widened technological competence to include radio communication core competences (acquired during 1G development) and digital network core competences (during 2G development). These skills could be used for developing terminal devices as well as base stations. The role of mobile network operators in the innovation process was diminishing. Their competences shifted toward service packaging, billing, and marketing and mobile operators were able to introduce a number of important complementary innovations in these fields.

Since 1999, MVNOs appeared, breaking up the boundary of the mobile operator’s firm along the distinction between facilities provision (layer 3) and network operation (layer 2) in (Fransman 2002)’s value chain (link). Firms such as Virgin Mobile, Lebara or Aldi Talk provide a mobile phone service to their subscribers but do not have their own licensed frequency spectrum, nor do they necessarily have the network and IT infrastructure required. Mobile network operators (MNOs) used MVNOs to expand geographically, retailers became telecom service providers, etc.

Different boundaries between network provision and operation may exist. The MVNO may own only billing and provisioning systems, or may own a core network including HSS/HLR as well. MVNE firms emerged to provide infrastructure services to branded resellers. The latter then owned no infrastructure at all and created value only through their brand, their marketing, sales and distribution power.

Horizontal unbundling continued when mobile network operators started outsourcing network construction and operations to service providers, typically the equipment suppliers from the infrastructure layer who were vertically integrating parts of network facilities provisioning. Today approximately 25% to 30% of telecom operations activities are outsourced and carried out under managed services contracts. Key network operator activities outsourced include wireless access, transmission and core planning, design and optimization; billing and support systems planning and design; site acquisition and construction, procurement, warehousing, installation, testing, integration; and operation and maintenance of the network and billing and support systems. Network and IT outsourcing by the mobile network operator was driven by the need to reduce operational expenditures when growth was stalling, a focus on business innovation and marketing rather than technology, and the complexity resulting from technology shifts and the corresponding inability to scale the own operation to cope with that complexity, as well as the economies of scale for companies such as Ericsson or Alcatel-Lucent able to sell outsourcing services across the industry. The latter argument reflects a belief in outsourcing as a complement and alternative to consolidation for achieving industry-scale synergies. Another driver is the preference by new industry players to source technological competence externally, given their limited operational experience (which contrasts with the technology-savvy incumbent operators).

When outsourcing, mobile network operators have to compromise with a loss of flexibility and a loss of the ability to differentiate with the network and IT systems. In the long run and with continued growth in outsourcing, new industry borderlines will be established. Cloud services will be accelerating that vertical break up.

The ability to differentiate based on network coverage or quality, generally important when introducing new network technology such as HSPA or LTE, further erodes with the advent of network sharing. The rationale for sharing is the scope for significant savings by pooling radio access assets and backhaul. In a typical setup, a joint venture company is set up or a third party is engaged to own the network and ensure that there is enough capacity in the network to handle actual usage levels and projected growth (e.g. 3GIS in Sweden is a JV set up to build and manage a 3G network on behalf of Hi3G and Telenor).

As unbundling continued, barriers to entry have fallen and competition intensified. Variable rather than fixed costs have become more significant. Operational leverage and scale economies have reduced. Small mobile network operators are able to enjoy similar costs as larger ones, and service provider competition is on relatively flat scale curves (Christensen et al. 2002). Core R&D capabilities will be less important for creating competitive advantage, while system integrators thrive. Successful firms will still need overlapping technological knowledge with suppliers to understand technological developments, select the right partners and integrate successfully.

When network operators built and upgraded networks for data communications, they initially took the traditional vertical approach to creating a user experience that they had taken for voice. The “mobile internet” user experience was developed in-house across the full layer stack using technologies such as WAP and the Japanese i-mode system. While the industry had historically been closed to third party service providers, the perceived need for mobile applications seemed to motivate at least some outsourcing and partnering with applications developers, as part of a closed, operator controlled walled garden. By contrast, fixed internet technologies were developed by non-profit institutions (unlike the commercial operators looking for new sources of revenue), which resulted in open technologies, open access and free content provision. Walled garden services largely failed due to high fragmentation (resulting in a lack of competition and dynamism, and low value creation by network effects), lack of coordination and unattractive revenue sharing models for content and application providers.

Many horizontal approaches to provisioning of new mobile services have been attempted by operators over the past decade; few have generated much traction. IMS is an operator-centric technology for horizontal services and service components on top of the network layer. It enables the encapsulation, convergence and control of traditional circuit-switched communications into a service including voice, video, web browsing, email, and internet messaging in the form of chargeable and manageable data sessions. The mobile network operator retains control.

A wide variety of service delivery platforms (SDP) that are non-IMS compliant (often SOA architectures) and even non-operator targeted have gained ground and converged into the application frameworks discussed before. These new platforms are enabling the delivery of content and services to end users in a rapid and inexpensive manner. Software using VoIP protocols such as Session Initiation Protocol (SIP) allow setting up a mobile voice or video call over the internet that behaves as a traditional call, but in a P2P fashion without needing any intervention from the mobile network operator.

These new service development and delivery environments enable mobile apps to be built and distributed easily and often without the active involvement of the operator (so-called OTT, or Over-The-Top services). A new range of web browsers are explicitly targeting the mobile phone and are attracting many developers already familiar with web service development. Session and service control is lifted to a higher layer that makes abstraction of the network. Several of the traditional operator control points are lost or at risk: non-communication needs, billing, service discovery and retailing (app stores), location (Foursquare, Google), authentication (Twitter, Facebook), voice (Skype).

These OTT services have introduced considerable disruptive dynamics. They have in common the potential to lower the barriers for mobile service development in a significant manner, and they are challenging the preconceptions about the fundamental differences between mobile and fixed internet. The OTT services and underlying technologies are the drivers of industry convergence in the sense that they allow a range of IT and internet companies to enter the mobile market, and they have (finally) opened up the mobile handset to the millions of web content providers and application developers. The quest for the killer application to generate return on the network investments of the past decade has resulted in a business model for mobile services where the key issue is the question of openness and related to that, the relation between (open) technology set-up and (horizontal/unbundled) market structure.

Whereas the loss of ground in the content provision (layer 5) and content integration (layer 4) are not generally good news for mobile operators, the trend reconfirms their core value add in connectivity, which — I will explain later — can be a very sophisticated product allowing for significant value discrimination and pricing power.

The following figure (adapted from Fine 2000) shows the oscillation in mobile network operator industry structure between vertical/integral and horizontal/modular.

Similar unbundling trends have materialized for wireline operators. Over the course of the past twenty years, regulators have forced incumbent operators to unbundle their copper-based infrastructure on layer 2 of the wireline value chain at marginal cost.

This unbundling mandated by regulation has undermined barriers to industry entry, since new competitors have had no need to secure billions in funding to roll out infrastructure of their own. It has also destroyed the incumbents’ economies of scale and their differentiation in terms of network reach. In terms of pricing, entrants had the ability to behave as the incumbent, since their cost base was similar. Without network reach and economies of scale as differentiators, the competitive battleground moved on to areas like marketing, where nimble entrants have tended to get the better of incumbent operators.

In emerging networks that mix fibre and copper (FTTx/VDSL), unbundling is intrinsically expensive as the need for hardware installations is not at a handful of local exchanges but at a huge number of street cabinets. It can even be practically impossible. Therefore, the landscape of competition in next generation access networks will likely change to one based on wholesale offerings, with improved economics and pricing conditions for the facilities provider (layer 2) as a result. This is similar to the MVNO model in mobile, discussed before. Naturally, there is the risk that regulators will impose strict net neutrality regulation and wholesale price controls on fiber-based connectivity. However, considering the digital divide that such regulation would create in less economically attractive geographical areas, regulators have an incentive not to undermine market-based pricing and hence jeopardize roll out.

One surprising example of this new vertical relation between layer 2 (facilities provision) and layer 3 (network operation) in the wireline communication value chain comes from Fujitsu, a layer 1 infrastructure vendor of long haul optical and Ethernet-based PON systems, which announced plans to vertically integrate into layer 2 of the value chain. Fujitsu wants to build a £2 billion high-speed broadband network based on optical fibre cables covering 5 million households in rural Britain. The company intends to have Virgin Media and TalkTalk as its flagship wholesale customers (Parker 2011). Another example is UK broadband provider Orange, deciding to abandon its unbundled ADSL offering and move onto BT’s wholesale platform (HSBC 2011).

The emergence of a home infrastructure (the connected home) opens up for a unbundling of previously vertical services where the home equipment necessary for consuming a specific service (such as TV) is no longer dedicated to that service. As this happens, barriers to industry entry are lowered. OTT services are becoming increasingly important particularly when it comes to paid TV services. Google TV is one example. Since even established media players are struggling in this field, the ambitions of telecommunication providers in the field of IPTV look stretched. Of course, continued activity in layer 4 with e.g. IPTV services helps retain bargaining power vis-à-vis other industry participants and can be a catalyst driving the core business of connectivity provision. Such is generally helped by the use of open platform technology. Even though most of IPTV’s revenues are passed on to the content providers [1], these services do raise entry barriers and discourage customer churn, as well as increase broadband connectivity adoption. 

A next step could be a generalized operator offered home environment where services may range from pure-play telecom services such as IPTV, video conferencing or richer communication to bundles with health services, surveillance, or energy monitoring.

Another operator opportunity is to embrace industry structure and operate as an enabler for content providers and retailers to leverage their customer relationships and brands.

Christensen C., Verlinden M., Westerman G. (2002) Disruption, Disintegration and the Dissipation of Differentiability. Industrial and Corporate Change 11(5):955-993

Fine C. H. (1998) Clockspeed. MIT Sloan, Basic Books

Fransman M. (2002) Mapping the Evolving Telecoms Industry: The Uses and Shortcomings of the Layer Model. Telecommunications Policy 26(9-10):473-483

HSBC (2011) Abundant scarcity. Pricing power is returning to telecoms. HSBC Global Equity Research. February 16, 2011

Kano S. (2000) Technical Innovations: Standardization and Regional Comparison — A Case Study in Mobile Communications. Telecommunications Policy 24(4):305-321

Parker A. (2011) Fujitsu £2bn broadband plan threatens BT. Financial Times, April 13, 2011


[1] Note the analogy with Pandora’s and Spotify’s music streaming business models on the internet. They face similar unattractive terms from content providers.

Tags: , , , ,


  1. Value Chain: Lateral Moves | - October 18, 2011

    [...] equipment players are no longer vertically integrated as explained in an earlier post. For the network equipment industry this is also the case but to a much reduced extent. The table below also illustrates how industry [...]

  2. New Network Platforms: The Telco and the SIM | - January 22, 2012

    [...] At the same time, the traditional activities of telecom operators are being pulled apart by centrifugal forces. More and more they look like a collection of increasingly unrelated businesses: retail, transport, [...]

  3. Facebook: Main Street to Wall Street | - January 31, 2012

    [...] and patents increasingly become the key assets around which firms are built. The proliferation of MVNOs and handset OEMs in the telecoms industry, companies that realize huge sales with very few physical [...]