Module 7.

New Technologies and Impacts on Regulation
Downloaded Tuesday, April 06, 2010
This document is a downloadable report from the infoDev/ITU ICT Regulatory Toolkit. Please note that the version that appears on the website at www.ictregulationtoolkit.org is the most updated version and has extra functionality (e.g., links, practice notes, reference documents etc). If you find any problems, please contact Feedback@ictregulationtoolkit.org.

Module 7. New Technologies and Impacts on Regulation Executive Summaries 1 Technology Trends 1.1 Fundamental Technological Changes 1.1.1 Digitalization Compression Modulation Forward Error Correction (FEC) 1.1.2 Computerization 1.1.3 Packet based switching 1.2 The Internet 1.2.1 Internet Protocol (IP) 1.2.2 Internet design principles Separation of networks and services End- to- End architecture Scalability Distributed design & decentralized control 1.2.3 QoS 1.2.4 Security 1.2.5 Mobility and Nomadicity 1.2.6 IPv6 1.2.7 Peer 2 Peer 1.3 Mobile Communication 1.3.1 Mobile Standards First Generation (1G) Second Generation (2G) Evolution of 2G (2.5G) Third Generation (3G) 1.3.2 Mobile Services Voice Services Location Based Services Multimedia Services

Corporate Services Mobile Internet Access 1.3.3 Future Technologies Software Defined Radio (SDR) Cognitive Radio 1.4 Next Generation Networks (NGN) 1.4.1 Next Generation Core Networks (NGCN) 1.4.2 Next Generation Access Networks (NGAN) ADSL ADSL2, ADSL2+ & RE-A D S L 2 VDSL & UDSL Cable TV PLC FTTx WiFi WiMAX Satellite Digital Broadcast Infrastructures Wireless Mesh networks 1.5 Convergence 1.5.1 Mobile Broadcast Convergence 1.5.2 Fixed Mobile Convergence 1.5.3 Converged Services VoIP IPTV 1.6 Information Society Technologies 1.6.1 The role of ICTs in other sectors Marketed Services Virtual Organizations 1.7 Disruptive Technologies 2 Market and Regulation 2.1 Direct and indirect technology implications 2.2 Implications of technology for the economics of the overall Market Structure 2.2.1 Service innovations Internet Key points and recommendations Mobile communication 2.2.2 Network innovations Copper- based telecom network Cable TV networks (HFC) Optical fibre networks Power line communication Mobile networks Wireless networks (other than mobile) Satellite Key points and recommendations 2.2.3 Vertical Separation – the Scope for new Business Models Internationalisation Key points and recommendations 2.2.4 Horizontal integration Key points and recommendations 2.3 Price regulation 2.3.1 The objective of price regulation 2.3.2 Current approaches to price regulation Discretionary Price Setting Rate of Return Price Cap C o s t - Based Price Setting 2.3.3 Technology implications for price setting

2.3.4 Key points and recommendations 2.4 Interconnection 2.4.1 The objective of interconnection regulation 2.4.2 Interconnection products/service offerings Circuit switched interconnection products Unbundled network components Packet switched interconnection products 2.4.3 Comparison of existing regimes for interconnection 2.4.4 Regulatory implications Mobile communication Development of next generation access network infrastructures Increasing use of IP and other packet switching infrastructures Convergence Implications caused by in market structure 2.4.5 Key points and recommendations 2.5 (Re) Licensing 2.5.1 Objectives and current regulation 2.5.2 Implications of technology trends Mobile communication Development of next generation network infrastructures Increasing use of IP and other packet switching infrastructures Convergence Implications caused by changes in market structure 2.5.3 Key points and recommendations 2.6 Universal Access 2.6.1 Objectives and current regulation Universal Service Obligation Universal Service Fund 2.6.2 Implications of technology trends Mobile communication Development of next generation network infrastructures Increasing use of IP and other packet switching infrastructures Convergence Implications caused by changes in market structure 2.6.3 Key points and recommendations 2.7 Spectrum Management 2.7.1 Objectives for spectrum management 2.7.2 Regulatory framework for spectrum management Allocation Allotment Assignment 2.7.3 Implications of technology trends Development of next generation of network infrastructures Mobile Communication Convergence Increasing use of IP and other packet switched infrastructures Convergence Implications mediated via changes in market structure 2.7.4 Key points and recommendations 2.8 Numbering 2.8.1 Objectives of numbering 2.8.2 Regulatory framework for allocation of numbers 2.8.3 Implications of technology trends 2.8.4 Key points and recommendations 2.9 Summary of regulatory implications 2.9.1 Regulatory implications from mobile communication 2.9.2 Regulatory implications from NGN 2.9.3 Regulatory implications from IP 2.9.4 Regulatory implications from convergence

2.9.5 Regulatory implications caused by changes in market structure 2.10 Key points and recommendations 3 New Regulatory Paradigm 3.1 The development so far and the new paradigm 3.1.1 The telecom reform process 3.1.2 The natural monopoly thesis 3.1.3 Transition phases 3.1.4 Consolidation and further development 3.1.5 Key points and recommendations 3.2 Technology and market trends 3.2.1 Regulation of convergence and divergence Convergence Divergence Key points and recommendations 3.2.2 Changed relations between national and international level International tariffs Internationalization of telecom markets Key points and recommendations Public sector influence 3.2.3 Technology leapfrogging Factors affecting leapfrogging Key points and recommendations Technology examples 3.3 Regulatory implications 3.3.1 Sector specific and/or general competition regulation Key points and recommendations Present discourse Regulatory areas 3.3.2 Technology neutrality Implications Interpretations of the concept Key points and recommendations 3.3.3 Infrastructure vs. service competition Access competition Key points and recommendations New access infrastructures The ‘ladder’ theory 3.3.4 Cost- based regulation C o s t - based pricing Costs as a basis for technology choice Key points and recommendations 3.3.5 Alternative business models Diversifying participation Focus of regulation Key points and recommendations 3.3.6 Quality of Service (QoS) Key points and recommendations Regulation Transition to packet- switching 3.4 Policy Integration 3.4.1 Integration of different categories of public policy intervention Key points and recommendations Modes of intervention The broader policy context 3.4.2 Focus on innovation Open the sector for innovations Competition and innovation Networks and services Key points and recommendations

3.4.3 Standardization Application and service standards Intra- and inter- standard competition Key points and recommendations Network standards 3.4.4 Public- private partnership Key points and recommendations PPP in communications The PPP discourse 3.4.5 Network and information security Importance of security issues Key points and recommendations Protection measures 3.5 Organizational aspects 3.5.1 Organizational dimensions 3.5.2 Organizational issues 3.5.3 Organizational leapfrogging 3.5.4 Key points and recommendations 3.6 Key Points and Recommendations 3.6.1 Diversification of participation 3.6.2 Policy integration 3.6.3 Convergence and regulation 3.6.4 International outlook 3.6.5 Organizational change 4 Hot Topics 4.1 NGN 4.2 Wi- Fi 4.3 WiMAX 4.4 VoIP 4.5 IPTV 4.5.1 What is IPTV? 4.5.2 Classification of IPTV 4.5.3 Applying content regulation to IPTV providers 4.5.4 Licensing IPTV providers 4.5.5 Network unbundling 4.6 Mobile TV 4.6.1 What is mobile TV? 4.6.2 Classification of mobile TV 4.6.3 Applying content regulation to mobile TV 4.6.4 Licensing mobile TV providers 4.6.5 Spectrum issues 4.7 Other legal and regulatory issues impacting IPTV and mobile TV 4.7.1 Legal issues related to acquiring content 4.7.2 Vertical integration in content markets 4.7.3 Standards 4.7.4 Quality of service issues 4.7.5 Ownership issues 4.7.6 Regulatory authorities responsible for IPTV and mobile TV 4.7.7 Checklist for regulators introducing IPTV and mobile TV Abbreviations

Module 7. New Technologies and Impacts on Regulation
The purpose of this module of the study is to develop an intellectual framework and innovative content exploring issues related to new technologies and regulation policies. The module is organized in three main parts: s Technological trends. This part examines the main technological trends and their impacts on regulation.
s

s

Market and regulation. The analysis of the technology implications on the current regulation and market structure in this part is focusing on traditional regulatory areas, like Interconnection, price regulation, etc. and market structure aspects, like horizontal and vertical integration. New regulatory paradigm. Based on the technology trends and regulatory implications a new regulatory paradigm and its constituent elements are discussed in this part.

RELATED INFORMATION Telecommunications Regulation Handbook 2000

Executive Summaries
Module 7. New Technologies and Impacts on Regulation
Executive Summary

1 Technology Trends
The technological development within ICT can be illustrated in three phases/levels: ‘The first wave of Technological Changes’, which are the fundamental technological changes that ICT has undergone and that has enabled the second phase; called ‘Second waves of Technological Changes’, which are further developments and deployments of the first wave of changes in the ICT industry and market. The ‘Third wave of Technological changes’ goes beyond the ICT industry and covers the broader aspect of application of ICT in the information society services. In this part of the study these technological trends are described in detail. The first wave of technological changes:
s s s s

Fundamental Technological Changes Digitalization Computerization Packet Switching

The second wave of changes:
s s s s

Internet Mobile communication Next Generation Networks (NGN) Convergence

The third wave of changes:
s

Information Society Technologies

1.1 Fundamental Technological Changes
Digitalization, Computerization and Packet based switching are the fundamental technological changes that have massively revolutionized the communication landscape during last two decades. As seen in the following, these technologies have massively improved resource utilization and increased the bandwidth capacity in the communication networks. Furthermore, these technologies have enabled possibilities for creation of new services and created conditions for gaining synergy in the technological development. These changes have directly influenced the communication markets and the regulation framework, as they are the basis for the IP revolution, the convergence process and emergence of Next Generation Network (NGN) technologies, which in turn have reshaped and restructured different communication sectors. This section is organised in three main parts:
s

Digitalization. The development from analogue to digital is by far the most fundamental precondition for any other technological changes we have witnessed in recent years. Digitalization enables the integration of different services in the same network and enables synergy to be reaped in the whole value chain of service- production, distribution and consumption. Furthermore, digitalization enables expansion of resources in the access and core networks in a technical and cost efficient way. Computerization. Another vital fundamental development has been the deployment of computers in the production and consumption parts as well as within the network infrastructures. The role of computers in production and consumption parts is quite obvious, but seen in the light of the objective of this Toolkit, it is important to emphasize the role of computers in the development of network infrastructures, including the deployment of computers in the network nodes as a replacement for switches and as devices adding intelligence to the network nodes. Furthermore, the processing power affects the spectrum use and management. Packet based switching. Packet switched technologies have had an important role in the more efficient utilization of the available resources in different network infrastructures and the creation of platforms enabling multi-service delivery in the same network, enabling real convergence. Different packet technologies have been developed with different advantages/disadvantages. Internet Protocol (IP) is the most successful packet-based technology and the dominant paradigm of today’s ICT infrastructures.

s

s

1.1.1 Digitalization

Digitalization is the technological foundation for the modern convergence process. In the beginning the concept digitalization was used synonymously with the simple conversion of an analogue signal to digital. However, even though the Analogue Digital conversion is the precondition for digitalization, the concept of digitalization, as it is used today, goes beyond this and encompasses the whole digital platforms & standards. Three main technologies have been important to make ICT digitalization become a reality: 1) Modulation, 2) compression and 3) Forward Error Correction. These are described in separate sub-sections. The following is a definition for digital signal: In the beginning of the 1960s, a radical transformation was introduced, as analogue telephone signals were digitized and multiplexed digitally using Pulse Code Modulation (PCM). A telephone signal has a bandwidth of 3.1 KHz. The signal is sampled, using sampling frequency of 8 KHz and quantised by 8 bit per sample, resulting in a digital bandwidth of 64 Kbit/s for a telephone signal. Two major digital transmission systems and multiplexing schemes, PDH and SDH, were developed to transmit these PCM codes within the backbone network. Three main technologies that have been important to realise digitalization of communication technologies and infrastructures:
s s s

Compression Modulation Forward Error Correction (FEC)

COMPRESSION

Compression denotes the techniques and protocols that reduce the bandwidth necessary for transmission of a given signal. For example, there is a huge amount of redundant information in the analogue audio and video signals. These can be removed and, consequently, the amount of bits per second that must be transmitted will be reduced substantially. Compression technologies determine the digital bandwidth by making a trade-off between how much capacity is available and the quality of service that is needed. Compression standards have been a vital factor for enabling distribution of audio/video services on the IP networks. Furthermore, compression techniques are the main pillars in the digital broadcasting standards. A number of different standards are defined for audio/video compression. ITU, ETSI, and a number of other standardization organizations have been involved in the development of compression standards. Moving Pictures Expert Group (MPEG) has developed three audio/video compression standards, widely deployed in development of audio video services:
s s

MPEG-1. Primarily intended for applications like computer images and graphics. MPEG-2 is used in digital broadcasting. MPEG-2 is intended to be generic in the sense that it serves a wide range of applications, bit rates, resolutions and services. MPEG-2 covers different picture resolution from Low Level (352X288) pixels to a very high resolution of 1920X1152 pixels, also called High Definition TV (HDTV) resolution. MPEG-4. Contrary to the MPEG-1 and MPEG-2, which are frame-based, MPEG-4 is object-based. MPEG-4 supports two-dimensional arbitrarily shaped, natural video objects as well as synthetic data. Synthetic data includes text, generic 2D/3D graphics and animated faces, enabling content-based interaction and manipulation. MPEG-4 is an important standard for the distribution of Digital TV to Handheld devices, and also for broadband IPTV and Video on demand (VoD).

s

MODULATION

Modulation technology is used to transmit information, including audio and video signals, over different transmission media; the information is modulated on the carrier waves when transmitted, and demodulated at the reception point. In principle, the technology is used for both analogue and digital transmission, though the techniques used in digital modulation – where a stream of binary numbers must be transmitted – are different from those used in analogue modulation. The modulation technologies are developed based on the characteristics of the transmission medium, so that the data is transmitted in the most efficient way. The modulation efficiency in digital transmission can be measured in bit per symbol or bit per second per Hertz (bit/s/Hz). The modulation efficiency depends on the deployed modulation technology, which again depends on the type of the delivery network deployed. There is always a trade-off between the error performance required and minimum data payload needed, which makes some modulations more attractive to a particular broadcast media. Modulation technologies have expanded the transmission resources in all infrastructures, and particularly in the radio spectrum modulation and other technologies which improve the spectral efficiency or include new spectrum to be deployed in the ICT sector, have challenged the scarcity argument, which has been an important pillar for the regulatory design in the ICT market. In this relation, a number of other technologies like Software defined radio, Cognitive radio, smart antenna and the technologies which use new spectrums (like the frequency range at 60 GHz and above) are important.

FORWARD ERROR CORRECTION (FEC)

The signals that are received at the end users' site are often erroneous, particularly in the wireless environment, due to noise and multipath interference in the transmission medium, among other things. These errors are experienced by the end-user as signal degradations, and consequently degradation of the quality of service. In the two-way communication networks, the problem of errors is often solved by retransmission of the signal. Another technology that can be used, when there is no return path to send commands up-stream and ask the source to retransmit the signal, or the timing requirements of the signal does not permit retransmission, is Forward Error Correction (FEC). FEC is implemented so that some overhead information is calculated and added to the signal prior to transmission. The FEC overhead information is then used in the decoder to detect and, if possible, correct the errors in the signal. An important issue regarding FEC is that a part of the transmission capacity is ‘sacrificed’ to reduce errors and increase quality, and the greater the capacity that is ‘sacrificed’ for the FEC overhead, the more secure the transmission. In practice, the level of FEC is determined by the characteristic of the transmission medium.

1.1.2 Computerization
Development of computers has had vital influence on the effective organization and operation of network infrastructures. As end devices, computers act as intelligent terminals. The use of computers in the network nodes has reduced the cost of technology, network management, operation and maintenance. The processing power of computers and the new applications have had a radical impact on the ICT sector. On the one hand, the expensive and complex functions in the network, such as switching and Intelligent Network services, are done to a large extent by computers. On the other hand, computers have diffused in practically every function necessary for operation of an ICT network, such as billing and Human Resource Management.

1.1.3 Packet based switching
Development from circuit switched to the packet switched paradigm is another important technological development. In a circuit switched network a dedicated connection (circuit or channel) is set up between two parties for the duration of the communication. The connection and therefore the network resources are occupied during the whole session. The plain old telephone service (POTS) network is an example of a circuit switched network. The problem with a circuit switched network is that the network resources are occupied even when they are not in use. For example, with regards to POTS, the network resources are occupied in the silence periods, resulting in inefficient utilization of network resources. Packet switched technologies, on the other hand, are designed to use the network resources only when meaningful data is subject to transport. Hence, packet switched networks utilize network resources more efficiently through bandwidth sharing. Another aspect of packet switched networks is their capabilities in carrying different types of services. Many modern packet-based technologies like ATM and IP are designed to be able to carry different types of services; however, specific technologies/protocols must be implemented for different services. Different packet switched networks are designed to support variable or fixed packet size and to operate in connection-oriented or connection-less modes. To be able to handle the packet loss, retransmission is implemented in some packet-based technologies. In the beginning 2 major packet switched standards dominated the market: X25 and Frame relay. Another important packet based technology is Asynchronous Transfer Mode (ATM). ATM was developed by the telecom industry and was telecom’s solution for integration of all services in the same network. ATM is cell-based technology consisting of packets with a fixed size: 53 bytes with 5 bytes of control information and 48 bytes of data segment. ATM is connection-oriented. Because of the fixed (small) cell size the delay can be controlled. ATM can be used both at the transmission and switching layer. It is possible to have full control of the end to end connection and guarantee Quality of Service (QoS). The most important and widespread packet technology in the ICT platforms is the Internet Protocol (IP), which is described in chapter 1.2 on the Internet.

1.2 The Internet
The emergence of the Internet, which interconnects billions of IP-based devices like computers to each other, may be seen as one of the most important changes in the ICT sector in recent times. The internet was in the beginning primarily used for data services. E-mail and World Wide Web (WWW) were the most important services on the Internet. In further development, however, the number of services over the Internet has expanded, and today these include a variety of audio/video services like Internet radio and IPTV, Internet telephony (VoIP), blogs and computer games as well as various ICT applications (e-education, e-government, e-health, e-commerce, etc.). The next development we are witnessing is the emergence of ‘Internet of things’, which is mainly connected to the development of RFID technology and ‘sensor networks’. While a number of issues related to the organization of the general Internet are in place, there are a number of

unsolved problems and challenges related to the ‘Internet of things’ which will be on the political agenda in the coming years. Even though the Internet itself has not been regulated in many countries, it has however had massive implications on the regulatory framework, as the Internet in different levels of development has been able to facilitate the offering of regulated services like voice telephony and TV/radio. The resource organization and IP Interconnection, for example, are becoming more and more important as the Internet development goes beyond advanced countries and becomes a part of daily day life in developing countries. This section describes different important aspects of the Internet and relevant issues connected to the its development: RELATED INFORMATION Design principles of the Internet Separation of networks and services End-to-End architecture Scalability Distributed design and decentralised control

1.2.1 Internet Protocol (IP)
Internet protocol (IP) was first developed in the mid-1970s, when the Defense Advanced Research Projects Agency (DARPA) became interested in establishing a packet-switched network that would facilitate communication between dissimilar computer systems at research institutions. With the goal of heterogeneous connectivity in mind, DARPA funded research by Stanford University and Bolt, Beranek, and Newman (BBN). The result of this development effort was the Internet protocol suite, completed in the late 1970s. TCP/IP (Transmission control protocol/Internet protocol) was later included with Berkeley Software Distribution (BSD), UNIX, and has since become the foundation on which the Internet and the World Wide Web (WWW) are based. The IP packets contain all the addressing information, which is necessary to be routed in IP networks. The IP routers transmit the IP packets within the network based on the destination address available in the IP packet in a connection-less manner. This reduces network complexity immensely. However, to provide services in the IP network, connection oriented protocols like TCP and UDP (User datagram protocol) must be implemented to establish a session and make sure that it functions properly.

1.2.2 Internet design principles
IP technology is designed in a way that enables a radically different environment for service development, innovation and competition when it comes to infrastructure platforms and service development platforms. The wide spread success of the Internet is based on four important characteristics:
s s s s

Separation between network technology and services End-to-End architecture, and extension of intelligence from the core to the edge of a network Scalability Distributed design and decentralized control

These characteristics of the technology create good conditions for development and competition where several actors can be involved in service creation and provision. The general Internet is the major IP network in the world, but it is far from the only IP network. In recent years, several private IP networks have been established and utilized for both corporate and residential services, and the future of communication platforms like the Next Generation Network architecture is based mainly on IP technology. However, when it comes to NGN, the level of competition or monopolistic characteristics depends heavily on the chosen architecture for the deployment of NGN.

SEPARATION OF NETWORKS AND SERVICES

The separation between the underlying network technology and the services removes entry barriers for the service providers. The only precondition for service provision is access to the network. This has created a huge dynamic in the service development within the Interne, but it also creates a problem of revenue sharing between the owners of the network infrastructures and the service/content provider. This is more obvious in the broadband IP infrastructures, which are mainly provided by the telecom operators. The development in value proposition is obviously mainly connected in service provision, particularly because flat rate billing for connectivity has become the dominant business model.

END-T O -END ARCHITECTURE

End-to-End architecture and extension of intelligence from the core to the edge of a network is another factor that moves the development and innovation activities to the edge of the network. The concept was first introduced in a paper named: ‘End-to-End argument in system design’. The main argument here is that an efficient network design can be based on ‘dumb core network’, where processing is moved to the edge of the network.

SCALABILITY

Scalability is another main feature of the IP design. One of the barriers for further scalability is the shortage of address room in the current IP version 4 (IPv4) systems. As discussed in the section on IP version 6 (IPv6), the shortage of address room is a big problem for developing countries, mainly due to uneven allocation of the IPv4 addresses.

DISTRIBUTED DESIGN & DECENTRALIZED CONTROL

Distributed design and decentralized control is another characteristic that has obviously improved conditions for the development of services, innovations and creations of new businesses. Different networks can easily connect to other IP networks, including the Internet, and can obtain added value from network effects etc.

1.2.3 QoS
QoS denotes the capability of the network infrastructure, client applications and the end user terminals to deliver a service living up to certain quality levels. QoS requirements vary from service to service and depend directly on the specific services. In POTS, for example, there are detailed recommendations on QoS from ITU on maximum delay, blocking rate, MOS (Mean Opinion Score) etc. QoS on the Internet is affected by a number of factors, including:
s s s s s

Delay Bit Error & Packet loss Speech compression Echo Firewalls

Different methods can be used to improve QoS. One can provide the necessary capacity in the backbone and access networks by ‘over provisioning’. QoS can also be implemented using one or more of following technologies:
s s s s s

Diffserve, ToS, RSVP, etc. Using priority schemes in the IPv6 Using appropriate speech codes Buffer size optimization Packet size optimization

The main deployment of QoS is nevertheless connected to the introduction and development of IP version 6 (the advanced or next generation IP), which allows for end-to-end QoS provision. In the managed IP infrastructures it is possible to provide measurable QoS, but this is more difficult in the best effort infrastructures like the Internet; however, in both cases regulatory measures may be necessary. An important issue is the facility-based operators' willingness to offer access to QoS provision to non-facility based operators. For example, a major debate in Europe and other regions is the lack of QoS provision in the wholesale Bit stream access products offered by the PSTN incumbents.

1.2.4 Security
In regular telephony services the security and consumer protection standards have been defined and are generally found adequate. With regard to the IP services, there is no one-to-one relation between the service and the physical infrastructure. In the IP networks, anyone with access to the network can tap the signal and actively damage the integrity of the message and the signal. For example, to ensure privacy in VoIP application, the VoIP provider can implement end-to-end encryption, which is not 100% secure, but can establish security levels comparable to those of regular telephony. The end-to-end encryption will, on the other hand, prevent the authorities from lawfully tapping the VoIP signal, i.e. from using ‘Lawful Interception’.

1.2.5 Mobility and Nomadicity

Generally we can distinguish between two types of mobility:
s s

Terminal mobility: A mobile terminal can move around the network without disrupting the service; Personal mobility (nomadicity): A user can move to different terminals and networks and still be connected.

Terminal mobility requires a wireless connection, while personal mobility can be implemented without necessarily having wireless connections. What is available now on the Internet could be called personal mobility or portability; one can move to different places, connect to the Internet and check e-mails etc. Through their advanced services, mobile operators are attempting to provide terminal mobility. Mobility can be implemented at different levels:
s s s

At link layer At application layer At IP layer

When it comes to e-mail application, the mobility (nomadic use) is implemented at the application level. This type of mobility is highly relevant for the VoIP. The VoIP service can be offered like e-mail, i.e. so that the only precondition for service accessibility is the availability of an IP connection. Here it is extremely complex to determine the location of the caller, which creates regulatory problems for emergency calls, discussed in the section about VoIP.

1.2.6 IPv6
The current Internet Protocol, which is primarily based on IPv4 (IP version 4) has had rapid growth both when it comes to the number of IP enabled devices and when it comes to applications and services. IPv4 suffers from major weaknesses when it comes to dealing with the rapid growth in the number of devices connected to the Internet and the new applications and services. This has resulted in the standardization of a new version of Internet Protocol, IPv6 (IP version 6), to cope with the shortcomings of IPv4. One of the main weaknesses of IPv4 is the number of IP addresses available globally. The IPv4 address consists of 32 bits, meaning that there are about 4 billion addresses available. On the one hand, it is obvious that 4 billion addresses are not enough in a world where more and more devices and terminals become IP enabled. On the other hand, even the current addresses available are allocated so unevenly that many of the developing countries lack IP addresses to develop their ICT infrastructures. For example, according to a consultation paper on ‘issues relating to transition from IPv4 to IPv6 in India'[1]: ‘India has merely 2.8 million IPv4 addresses compared to 40 million acquired by China’. Here it is important to note that any common US university has more IP addresses than the total of India, and that a US ISP, Level-3, alone has more IP addresses than China. The distribution is much worse when it comes to the least developed countries; for example, Bangladesh has about 150,000 IP addresses. IPv6 extends the address room to 128 bits, meaning that the number of IP addresses will not be any problem in the foreseeable future. This allows for the allocation of more addresses to different countries and regions. The allocation of IPv6 addresses can be done more evenly, as it does not suffer from the historical matters that resulted in the uneven allocation of IPv4 address room. In future development, where we are surrounded by the ‘Internet of Things'[2], there will be an even greater need for IP addresses. The other issues that are dealt with in IPv6 are the QoS and security issues. QoS is important in relation to real time services, and security at IP level will generally be required by a number of services in the future. ENDNOTES [1] TRAI: Consultation paper no. – 8/2005, TRAI, ‘issues relating to transition from IPv4 to IPv6 in India, August 26, 2005. [2] See amongst others the ITU Internet report 2005: the Internet of Things.

1.2.7 Peer 2 Peer
The Internet is traditionally based on a client-server approach. There are a number of servers in the networks doing specific tasks, such as e-mail server and web server. The end-users install clients on their IP terminals, such as computers, mobile phones and PDAs, and connect to the servers for specific services. There is, however, another approach that is used more and more, where the end-user's IP terminals act as both a client and a server. In this approach the IP terminals connect directly to each other and share information such as files. This approach is called Peer 2 Peer to indicate that the peers communicate directly with each other. The Peer 2 Peer concept is as old as the Internet. The first Internet was Peer 2 Peer, and it was in later development that the Client Server approach was invented and developed to offer services like E-mail and WWW. However, the large deployment of Peer 2 Peer is mainly connected to the applications designed for sharing music and later movies. The Peer 2 Peer technology is by no means limited to the sharing of music and movies, and can be used in any network applications which need information sharing. These are, for example, multiplayer games, signaling systems in telecommunication and corporate applications. One of the main challenges to making Peer 2 Peer networks work efficiently is the locating of information. This can be compared to the role of signaling in the telephony networks to locate the parties who want to communicate with each other. Through the last 5-6 years a number of Peer 2 Peer applications have been developed; some known examples are: 1) Napster, 2) Gnutella, 3) FreeNet and 4) BitTorrent. The aim of these applications has been to facilitate document sharing and mainly the sharing of music and movies. Many of the current applications, which are based on Peer 2 Peer, are based on one of these technologies, or a combination of them. Furthermore, the VoIP application Skype has based its signaling on the Peer 2 Peer approach. Skype is discussed in a separate section in this report. In the following, Napster and Gnutella are described.

Napster was the first large Peer 2 Peer application for music sharing. Napster was developed in 1999 by a college student to share his music with his friends. Soon afterwards, Napster became popular among the users and very unpopular among the record industry and the content owners. Napster was closed down by a US court in mid 2000. The locating of information in Napster was implemented by deploying a central index server, keeping track of available content on the net, so the approach was Peer 2 Peer, but it still relied on a central server. Gnutella, on the other hand, was designed with the requirement of not relying on a central server to index and locate of information. In Gnutella, every peer has connections to a few neighbor peers. In order to locate content, the peer will ask its neighbor peers, if they don’t have the content they will ask their neighbor peers, and in this way, finally, the content will be found and a connection to the content owner will be established. This approach doesn’t have the weakness of Napster with regards to centralism, but consumes a great deal of bandwidth, by sending requests in several directions. As described in the section on Skype, Skype deploys an approach between Napster and Gnutella, where the index server is distributed in the network so that some of the nodes are appointed to be super nodes.

1.3 Mobile Communication
The development of mobile technologies and services in the last two decades has had massive implications on the ICT landscape. Mobile technologies enable mobility and flexibility in the use of ICT services. Mobile technologies have primarily been driven by voice telephony but in their development, they embrace the whole portfolio of converged services, particularly when it comes to wireless standards and the new generation mobile technologies. The emergence of mobile communication has influenced the telecom regulation at all different levels. Licensing and frequency management have been the main regulatory issues for the introduction of mobile services. Furthermore, the regulatory design related to interconnection and tariff regulation, pricing, numbering etc. have been important in making a competitive and innovative mobile market to develop. Due to its ‘time to market’ and flexibility, mobile communication has been important in offering telephony to developing countries. Taking off in about 1990, mobile communication has gone from being a rare and occasional service based on expensive pieces of equipment used by businesses to an integrated part of daily life based on a pervasive low-cost personal item. In many countries, mobile phones now outnumber land-line telephones. Today there are more mobile than fixed lines in every single region of the world. However, a distinction must be made between the European phenomenon (and growing North American phenomenon) of every man, woman and child having a mobile phone, and the developing country experience of there being more mobiles than fixed-line phones, but primarily just for adults, because there are so few fixed lines and mobile has been cheaper to deploy. The latest indicators from ITU show that “By the end of 2004, the world counted some 1.8 billion mobile subscribers (including both second and third generation mobile subscribers), or 28 per cent of the world’s population. Some 58 per cent of these mobile subscribers were located in developing countries.” This means that after only 15 years of roll-out, a new communication mode based on its own networks and a new set of services has been globally established. Mobile communication has a history stretching back to the 1940s with radio phones, with hand-held cellular radio devices based on analogue technology being available since the beginning of 1980s. Due to their low establishment costs, rapid deployment and cheap terminals, mobile phone networks based on digital cellular technology have spread rapidly throughout the world. The digital mobile phone has become ubiquitous because of the interoperability of mobile phones across different networks and countries. This is to a large extent due to a general acceptance by operators and equipment manufacturers of the GSM standard which was designed for Europe-wide interoperability. All European nations and some Asian nations chose it as their sole standard, and according to the GSM Association (GSMA), the global trade association representing mobile network operators, it is active across 210 countries of the world. Countries such as Japan and South Korea have selected another standard, CDMA, as their sole system, whereas in the US both standards have been introduced. This has implied an explosion in connectivity. In the developed world, the connectivity options available have multiplied enormously. Once, there was only the fixed-line telephone and people had to wait to more than six months before the telephone company installed the phone. Then, from 1990, the mobile phone changed the way in which people live and communicate. In the lesser developed world, mobile has for the first time brought communications to many living in remote and rural locations. Particularly in these countries, the availability of prepaid services, where the subscriber does not have to commit to a long-term contract, has helped fuel the take-up of mobile. Developments in device technology are also playing a key role in this development. Moore’s Law, associated with the first wave of technology, has enabled dramatic increases in the computing and processing capability of portable devices, whilst ensuring that costs and prices fall significantly. Information, music, video and data can be stored in large amounts on small form factor devices and transported easily and quickly, making content virtually ubiquitous. Today, the mobile phone can play MP3 files and handle MPEG4-tasks. This represents a total shift in the communications paradigm.

1.3.1 Mobile Standards
In this section the three main generations of mobile standards (1G, 2G, 3G) as well as the 2.5G are described.

FIRST GENERATION (1G)

The first generation mobile standards were based on analogue technology. The mobile market in this era was fragmented, where a variety of standards were developed and used in different countries. The Nordic Mobile Telephone (NMT) is one of the earliest 1G-standards. NMT was developed jointly in Denmark, Finland, Iceland, Norway and Sweden. NMT operated originally in the 450 MHz band and later also in the 900 MHz (NMT-900). A number of other 1G-standards were used in different countries/regions. Some important examples are Total Access Communication Systems (TACS) in the UK and Ireland, NMT-F and RC 2000 in France, NTT in Japan, Advanced Mobile Phone System (AMPS) in the US, C450 in South Africa and C-Nets in Germany and Austria. In Japan, the first commercial 1G service was provided by NTT Public Corporation (NTTPC) in 1979, where they introduced the ‘automobile telephone’. Soon the device became detached from automobiles and was called ‘shoulder phone’. Between 1985 and 1988 a number of new carriers entered the market. A number of EU and other European countries introduced their first analogue cellular phone systems in the 1980s, starting in Sweden in August 1981. In the majority of European countries, government-owned incumbent telecom operators were in charge of operating the mobile service. However, among the EU member states, U.K. and France were the only countries that permitted a second mobile operator to provide cellular phone services. In the US the AMPS standard developed by the Bell labs was in use in 1983. Today the analogue mobile systems are not in use any more, and many countries are reallocating the frequency resources to be used by 2G and other mobile/wireless systems/standards. Because of the level of fragmentation of the market, efficient harmonization and interoperability/roaming was either a non issue or at best a very complicated process. This was especially seen as a huge problem in Europe. Hence one of the requirements for the next generation mobile was the use of common standards and the creation of a single market for mobile. Another main requirement for the new standards was more optimal utilization of frequency resources. This requirement has been fulfilled by selecting digital technology as the foundation for all the 2G standards and beyond.

SECOND GENERATION (2G)

The second generation mobile standards are based on digital technology. Digital technology utilizes the transmission resources in an efficient way, both due to advances in audio compression standards and also due to advances in digital modulation technologies. Another important characteristic of the 2G mobile is the less fragmented mobile market. This is particularly due to Europe’s decision to use a common standard and the creation of a single mobile market, but also because the European standard, GSM, has had enormous success beyond Europe and is used in a number of other countries. However, there are a number of competing standards to GSM, mainly: s TDMA IS-136 is the digital enhancement of the analogue AMPS technology. It was called D-AMPS when it was fist introduced in late 1991 and its main objective was to protect the substantial investment that service providers had made in AMPS technology. It is mainly used in North America. s CDMA IS-95 increases capacity by using the entire radio band, each using a unique code (CDMA or Code Division Multiple Access). It is a family of digital communication techniques and South Korea is the largest single CDMA IS-95 market in the world. s Personal Digital Cellular (PDC) is the second largest digital mobile standard although it is exclusively used in Japan, where it was introduced in 1994. Like GSM, it is based on the TDMA access technology. s Personal Handyphone System (PHS) is a digital system used in Japan, first launched in 1995 as a cheaper alternative to cellular systems. It is somewhere between a cellular and a cordless technology. It has inferior coverage area and limited usage in moving vehicles. GSM Group Special Mobile within the CEPT started to develop GSM in 1982. Later it was standardized by the ETSI and branded as a Global System for Mobile (GSM). One of the most important conclusions from the early tests of the new GSM technology was that the new standard should employ Time Division Multiple Access (TDMA) technology. This ensured the support of major corporate players like Nokia, Ericsson and Siemens, and the flexibility of having access to a broad range of suppliers as well as the potential to get product into the marketplace faster. After a series of tests, the GSM digital standard was proven to work in 1988. GSM operates mainly on 900 and 1800 frequency bands. However, in North America it operates on 1900 MHz. There is also a version which uses the 450 MHz band (GSM400), which can be used in the less populated areas, and can be relevant for the less populated rural areas in the developing countries. In the transition from 2G to 3G a number of standards have been developed, which are categorized as 2.5G. These are add-ons to the 2G standards and mainly focus on deployment of efficient IP connectivity within the mobile networks.

EVOLUTION OF 2G (2.5G)

In the 2G and 2.5G mobile, several technological developments have been introduced to increase the capacity bandwidth of the networks and to enable provision of new services in these platforms. Standard bandwidth for data services in GSM networks is 9.6 Kbps per time slot. However, many providers offer 14.4 Kbps per time slot using more efficient modulation technologies. To increase the available capacity at the end user’s site in GSM networks, two approaches are used:
s s

Deployment of several time slots. This is called HSCSD (High Speed Circuit Switched Data). Deployment of packet oriented IP based technologies like GPRS and EDGE.

When using HSCSD technology, a maximum capacity of 38.4 Kbps will be achieved if 9.6 Kbps per time slot is used (and 57.6 Kbps in the case of 14.4 Kbps per time slot). In both cases, the assumption is that all 8 time slots are used: 4 time slots for uplink and 4 for downlink. GPRS, on the other hand, is packet-based and is optimized for IP traffic. In GPRS, the capacity per time slot depends on the deployed technology: CS1: 9.05 Kbps per time slot; CS2: 13.4 Kbps per time slot; CS3: 15.6 Kbps per time slot; CS4: 21.4 Kbps per time slot. In theory, using 8 time slots and CS4 technology, a maximum capacity of 171.3 Kbps can be achieved. EDGE can be seen as a technology with the same characteristics as GPRS, but with more efficient modulation techniques and, consequently, higher capacities per time slot. Theoretically, it is possible to achieve 59 Kbps per time slot, providing a maximum capacity of 472 Kbps. The capacity will depend on the deployed technology (MsC1 to MsC9), and a maximum capacity per time slot of 48 Kbps is considered realistic in mature EDGE networks giving a maximum overall capacity of 384 Kbps. One important issue here is that even though GPRS and EDGE are capable of offering high bandwidth connectivity to the end users, the amount of frequency resources in the GSM network is far below the resources necessary to cope with the ever increasing demand of the end users for data services. The technological evolution path towards 3G networks and the standards that will be deployed in different markets depend primarily on the current 2G markets. The natural consequence of this has been the definition of a variety of variants of IMT-2000 standard, that can be chosen by different operators based on parameters like reusability, interoperability, etc.

THIRD GENERATION (3G)

The main development in the mobile networks has been the development from 2G to 3G and beyond. This has been primarily driven by the lack of frequency resources in 2G to cope with the rapid development and penetration of mobile services and the need for new mobile services with varying demand on bandwidth. The 3G platforms on the one hand include new frequency bands for the provision of mobile services, and on the other hand deploy more efficient technologies than 2G, resulting in increased spectral efficiency. Furthermore, the 3G technologies have been developed due to their potentials in meeting universal access goal. This has been one of the arguments at ITU for backing the development of 3G standards. In order to generate a more efficient usage of available frequency, IMT-2000 had the job of standardizing the new air interface that would enable this. The WCDMA air interface was selected for paired frequency bands (FDD operation) and TDCDMA (TDD operation) for unpaired spectrum. CDMA2000 standard was created to support IS-95 evolution[1]. W-CDMA (Wideband Code Division Multiple Access). W-CDMA is the access scheme defined by the ITU to be the main technical platform for UMTS or 3rd Generation Mobile services. W-CDMA services are to operate within the following frequency bands: 1920 MHz 1980 MHz and 2110 MHz - 2170 MHz. NTT DoCoMo’s FOMA network uses a version of the W-CDMA standard in its network deployment. The ITU had selected W-CDMA as one of the global telecom systems for the new IMT-2000 3G mobile communications standard. In W-CDMA interface different users can simultaneously transmit at different data rates and data rates can even vary in time[2]. W-CDMA is capable of delivering up to 384 kbps in outdoor environments and up to 2 Mbps in fixed indoor environments. CDMA2000 (Code Division Multiple Access 2000). CDMA2000 (with the ITU name IMT-2000 CDMA Multi-Carrier) represents a family of technologies that includes CDMA2000 1X and CDMA2000 1xEV:
s

CDMA2000 1X can double the voice capacity of CDMAOne networks and delivers peak packet data speeds of 307 kbps in mobile environments. CDMA2000 1xEV includes:
r

s

CDMA2000 1xEV-DO. CDMA2000 1xEV-DO delivers peak data speeds of 2.4Mbps and supports applications such as MP3 transfers and video conferencing. CDMA2000 1xEV-DV. CDMA2000 1xEV-DV provides integrated voice and simultaneous high-speed packet data multimedia services at speeds of up to 3.09 Mbps.

r

1xEV-DO and 1xEV-DV are both backward compatible with CDMA2000 1X and CDMAOne. As a more advanced development from the previous CDMAOne system, CDMA2000 supports data communication speeds ranging from 144kbps to 2Mbps, with 144kbps being the maximum possible speed outdoors and 2Mbps being the maximum speed in an indoor environment. Because CDMA2000 has evolved directly from the previous generation of proven CDMA systems, it provides the fastest, easiest, most cost-effective path to 3G services. While all 3G technologies (CDMA2000, WCDMA and TD-SCDMA) may be viable, CDMA2000 is much further ahead in terms of product development, commercial deployment and market acceptance[3]. The world's first 3G (CDMA2000 1X) commercial system was launched by SK Telecom ( Korea ) in October 2000. Since then, CDMA2000 1X has been deployed in Asia, North and South America and Europe , and the subscriber base is growing at 700,000 subscribers per day. CDMA2000 1xEV-DO was launched in 2002 by SK Telecom and KT Freetel. The commercial success of CDMA2000 has made the IMT-2000 vision a reality.

TD-CDMA (Time Division- Code Division Multiple Access). UTRA TDD is planned to operate in the unpaired spectrum. TDD uses a combined time division and code division multiple access scheme. It is based on a radio access technique proposed by ETSI Delta group and the specifications were finalized in 1999[4]. ENDNOTES [1] http://www.umtsworld.com/technology/technology.htm [2] http://www.umtsworld.com/technology/tdcdma.htm [3] http://www.qualcomm.com/cdma/3g.html [4] http://www.umtsworld.com/technology/tdcdma.htm

1.3.2 Mobile Services
Mobile services in the 1G and 2G are dominated by regular voice services, offered primarily in circuit switched network architecture. In the 2G, however, the SMS service has also been an important service. Furthermore, IP connectivity and Internet access have been the drivers of the development towards 2.5G and 3G. It is generally accepted, particularly in 3G, that the data and Internet type of services will dominate the 3G markets. Furthermore, the voice services will be further differentiated and will not remain as a unique and coherent set of services as we know them today. The mobile services can be divided in two categories:
s

Inter-personal communication services: These are the main services in the current mobile networks, with voice services as the absolute dominant one. Data and other communication services: These are primarily communication services between a service provider (or a workplace, machine or application) and the end-user. These services are developed rapidly on the Internet and the majority of them are based on IP protocols provided on the Internet and /or other IP based networks.

s

VOICE SERVICES

Voice services are one of the major inter-personal communication services and will remain so in mobile networks. Three categories of voice services are:
s s s

Regular voice services, as we know them from 2G networks. Premium voice services: Voice services with higher Quality of Service, targeted at business users. Voice over IP: VoIP is used for enabling voice communication over packet switched IP networks. The possibility of using the Internet as a backbone particularly creates possibilities for provision of long distance cost-efficient voice services. In the mobile networks, on the contrary, it is not necessarily utilization of Internet as a backbone but efficiency of establishment and maintenance of one switching technology, namely packet-switched technology, that will drive the development towards using VoIP.

IP protocols will be used to a higher degree in the future mobile networks and, as stated in the UMTS report number 11: “It is unlikely that a 3G network operator will want to operate both circuit-switched and packet-switched networks to cater for different data types for very long. The most likely scenario is that the network operator will want to migrate quickly to a single multi-service packet-switched network”. Different standardization organizations are working on ‘all IP’ 3G networks. Among them can be mentioned organizations such as 3GPP, 3GPP2, and IETF, as well as different industry forums such as 3G.IP and MWIF. Other voice services will also be offered in close connection to non-interpersonal communication services. It will, however, be difficult to separate voice services from data services. Voice is an essential component in many "data-oriented” services. Voice communication, speech recognition and options such as Web-initiated voice calls and voice-activated WEB access are increasingly being added as options in the fixed Internet world.

LOCATION BASED SERVICES

Any service that is created by using geographical location information as one of the components of the services can be categorized as a location-based service. Location based services are business and consumer 3G services that enable users or machines to find other people or machines, and / or enable others to find users as well as enabling users to identify their locations. Navigation is a good example of a location-based service. Location-based services are especially essential regarding the 3G networks. These services are only interesting in the mobile and wireless networks, and in the beginning it was predicted that location-based services would represent a high proportion of services available in the 3G networks. However, the widespread success of location-based services is still to be seen.

MULTIMEDIA SERVICES

The provision of multimedia services, known primarily from Internet, will seriously begin by the introduction of 3G mobile networks. This is due to higher capacity available for data services in 3G networks. Multimedia services are also relevant in GPRS and especially in EDGE, where the available capacity facilitates the provision of services containing, for example, video, audio and text components. Multimedia Messaging System (MMS) can be mentioned as a successor for the Short Message System (SMS) that has been a successful service in the 2G markets in European countries. High expectations on MMS services are evident in different analysis, such as the above-mentioned UMTS reports.

CORPORATE SERVICES

Remote access to corporate networks and services (intra/extranet) and tele-working has increased rapidly during recent years. Mobile access to corporate IP networks has been underlined by the analysts as one of the vital services aimed at the business sector.

MOBILE INTERNET ACCESS

One of the important services in the future mobile networks is data (including Internet) services. These service types will be a subset of the current fixed Internet service types like information, interactive and entertainment services. Some examples of information and interactive services are news, weather, stock prices, horoscopes, sports scores, train times, restaurant guides, dictionaries, stock trading, banking, ticket purchasing, m-commerce, insurance and car rental. Examples of entertainment services can be video/audio services, gambling, chat or jokes.

1.3.3 Future Technologies
So far in this section the development within the PLMN (Public Land Mobile Networks) has been analyzed. Another important development is related to the new wireless systems and standards, which are partly fixed and partly mobile. Development of the new wireless infrastructures is analyzed in the chapter on NGN. Another important technological development which relates to PLMN as well as the new wireless systems is related to Software Defined Radio and Cognitive Radio.

SOFTWARE DEFINED RADIO (SDR)

Software Defined Radio (SDR) and Cognitive Radio are future technologies that provide a more flexible design for the wireless and mobile industry, and at the same time these technologies enable the efficient utilization of frequency resources. However, correct deployment of the technologies requires radical changes in the regulatory framework of frequency management. Software defined radio (SDR) is a flexible radio architecture programmed through software, which is reconfigured depending on the usage scenario. SDR consists of a programmable hardware base that is controlled through software, where different parameters, like power level, frequency band and modulation are changed/configured depending upon the environments in which users move. In Wireless Telecom, Issue 1 2004, called ‘Defining a Radio Revolution'[1], a good analogy is used between SDR and the emergence of computers: ‘Before the commercialization of the PC, the typical business had many specialized devices. A typewriter was used for correspondence. An adding machine handled various math functions. A drafting board and associated equipment might be required for creating drawings. And so on. Each performed its assigned task very well, but for the most part each was a single-function device. No matter how hard one tried, a typewriter would not add a column of figures, ... . As we all know, PCs changed that by creating a hardware platform and an operating system, then allowing users to mount various programs on these as required. A single computer could become a typewriter, a calculator, a drafting board, ... .’ In the current Hardware Defined Radio (HDR) era all the parameters in the radio access interface are fixed parameters that cannot be changed. A GSM interface is designed for the GSM band, so a GSM phone cannot be reconfigured to work in a CDMA environment[2]. Therefore, using the analogy of the pre-PC era to the current HDR era, there are a number of specialized wireless devices for specialized tasks. The SDR platform will, like PCs, create the possibility of changing the character of the device depending on the application and by downloading different software modules. SDR creates a number of regulatory challenges, especially when it comes to frequency allocation and management. For regulators, SDR has the potential to bring radical changes to how spectrum is used, and therefore to the regulations that apply to radio communication systems. In the HDR era it is relatively easy to maintain frequency management, because ‘the manufacturer sets power limits and other restrictions based on the regulations that govern the type of device being built, and the device must be certified before it can be sold. But in Advanced

Wireless Technologies and Spectrum Management…. the ITU suggests that if a device can change its function simply by loading a new application onto it, the current, hardware-based approvals process will not be sufficient. (According to ITU) "This raises some difficult questions for regulators since rather than focusing on the hardware elements of electronic devices... the key approval must lie with the software that controls the radio'[3]. Furthermore, there are important issues related to the way the SDR market will emerge. In the SDR era (like PCs) one can acquire hardware from one firm, software from another and the operative system from a third firm. Which one of these must be regulated? What if somebody modifies the software, for example, to change the power level? These questions and uncertainties show that even though there are many possibilities and flexibilities connected to the SDR development, there are also a number of challenges from the regulatory side to make this revolution take place.

ENDNOTES
[1] http://www.crc.ca/en/html/crc/home/research/satcom/rars/sdr/news_room/wireless_telecom_sdr_apr04 [2] Another approach is to implement different hardware standards in the same terminal, like dual band, etc. [3] INTERNATIONAL TELECOMMUNICATION UNION, WORKSHOP ON RADIO SPECTRUM MANAGEMENT FOR A CONVERGING WORLD, Document: RSM/08, February 2004,GENEVA — ITU NEW INITIATIVES PROGRAMME — 16-18 FEBRUARY 2004.

COGNITIVE RADIO

Cognitive Radio is a technology that could make efficient use of unused spectrum, potentially allowing large amounts of spectrum to become available for future high bandwidth applications. Most of today’s radio systems are unaware of their spectrum environment; they are designed to operate in a specific frequency band. A Cognitive Radio system senses and understands its local radio environment to identify a temporarily vacant spectrum to operate in. Cognitive Radio would hop into unused bands of the radio spectrum and hop out again if a primary user[1] of a band required that spectrum.[2] Ofcom’s studies in this area are just commencing and no firm conclusions have yet been reached. The work undertaken into SDR suggests that flexible, multi-protocol, multi-band Cognitive Radio systems are some way off. Handsets employing this technology are unlikely to emerge before 2010. In the interim there is the possibility of specific band sharing technologies emerging, which would provide a stepping stone towards the full Cognitive Radio vision.[3] ENDNOTES [1] For example a license holder to that part of spectrum [2] Ofcom Technology Research Program [3] Ofcom Technology Research Program

1.4 Next Generation Networks (NGN)
The NGN concept is mainly used in two ways: 1) A broad concept encompassing the whole development of new network technologies, new access infrastructures and even new services, and 2) A focused concept of specific network architecture and related equipments, with one common IP core network deployed for the entire legacy, current and future access networks. The first definition is so broad that in a sense it covers the whole current chapter on technological trends. The second definition relates to the transition path towards a converged IP-based core and access network. In this report, the concept of NGN denotes the second definition. Here we distinguish between the Next Generation Core Network (NGCN) and Next Generation Access Network (NGAN). The NGCN is about the new switching, gateways and transmission equipments in the core network, enabling several access networks to use the same core network. The NGAN is about new access networks, like deployment of optical fibers, and the challenges derived from that. ITU defines NGN as "a packet-based network able to provide telecommunication services and able to make use of multiple broadband, QoSenabled transport technologies and in which service-related functions are independent from underlying transport related technologies. It enables unfettered access for users to networks and to competing service providers and/or services of their choice. It supports generalized mobility which will allow consistent and ubiquitous provision of services to users".[1] In this definition there is a major emphasis on one of the main characteristics of IP platforms namely the separation of network and service layers. NGN is about the transition from current dedicated voice (and radio/TV) networks to the IP-based networks. From a technology efficiency point of view this is a natural development of all network technologies. However, there are a number of problems connected to the overall organization of the NGN platforms, which are subject for discussions amongst regulators today. One of the main issues is the interconnection model which will be used in the NGN. Will this, for example, be dominated by the IP interconnection models like peering and transit, or will the PSTN interconnection and tariff regime be modified and used in the future NGN platforms? In this section first the NGCN and then the most important NGANs are described and organised in two major categories, Fixed and Wireless: Fixed:
s

ADSL

s s s s s

ADSL2, ADSL2_PLUS & RE-ADSL2 VDSL & UDSL Cable TV PLC FTTx

Wireless:
s s s s s

Wi-Fi WiMAX Satellite Digital Broadcast Infrastructures Wireless Mesh Networks

ENDNOTES [1] ITU-T Recommendation Y.2001

1.4.1 Next Generation Core Networks (NGCN)
Figure 1 illustrates the difference between today’s telecom networks and tomorrow’s NGN platforms. Today, the PSTN, mobile networks, Cable TV networks and Wireless networks use several dedicated metro and core networks. Figure 1: Today's Networks and Next Generation Networks

Source: Ofcom In the NGN platform all of these different access technologies share the same IP core network. The main arguments for transition to the NGN architecture are:
s

It is not efficient to maintain several core networks for different access networks. Substantial cost savings can be achieved due to the economy of scope inherent in a single converged network. BT predicts[1] a reduction of costs by £1 billion per annum by 2008/2009 as a consequence of migrating to NGN. According to BT, NGN enables improved time to market for new services and improves customer experience. NGN enables the continued offering of services in the legacy access networks. For example, the analogue PSTN access line/service does not need to be changed in transition to NGN. The main changes here are the efficiencies gained in the core network, especially when one operator owns and operates several parallel core networks. The latter is the case for a majority of incumbent operators. So the operator on the one hand utilizes the backbone efficiency gains and on the other hand continues to make profit from the investments in the access networks. NGN enables the provision of value added innovative services using the possibility that one core network is connected to and manages different access networks. For example, a SMS can be sent to a mobile subscriber to inform the users if there are problems with the operation of DSL.

s s

s

These arguments show that the implementation of NGN is a radical change in the network architecture of incumbent telecom operators.

This raises the question of the role of regulation in this process: Should the regulators get involved in the practical implementation of the NGN? The answer to this question is no, as it contradicts the new regulatory doctrine of telecom development, where the decision of technological changes is taken on the market and by the market players/industry. However, the regulator must make this clear in setting the constraints within which the industry should design their networks. The role of regulation regarding NGN is on the one hand to make sure that effective competition can take place in the NGN era, and on the other hand to make sure that the consumers and the level of services they receive are not affected in a negative way in this transition. ENDNOTES [1] Ofcom: Next Generation Networks: Further Consultation, Issued: 30 June 2005, Closing date for responses: 12 August 2005.

1.4.2 Next Generation Access Networks (NGAN)
Due to the wide spread of the installed base of the PSTN physical infrastructure, PSTN has been the basis for fast and efficient development and penetration of the Internet. In the pre broadband phase this was implemented by modulation of data signal in the same frequency spectrum as regular voice in the copper access lines. The data capacity in this frequency bandwidth is small and it can reach the maximum of 56 kbps. The next phase was introduction of ISDN, which improved the capacity and could offer 128 Kbps to residential households. While dial up PSTN modems and ISDN have had enormous impacts on the development of Internet access, today the Internet connectivity is mainly influenced by different broadband technologies. The new IP broadband infrastructures are in the literature denoted as the New/Next Generation Access technologies (NGAN). NGAN covers both the fixed (wired), wireless and mobile infrastructures, which enable IP connectivity to the households and companies. The terms NGAN and broadband can be used interchangeably, however the important thing is that NGAN is not only characterised by higher capacity networks but also by the two other main characteristics, namely: 1)'always on' provision model, and 2)the flat rate business model. With regards to the fixed networks the NGAN is mainly dominated by xDSL and cable modems. The advantages of the xDSL and cable modems are the availability of the physical infrastructure and by that the low deployment cost. Other fixed technologies like PLC have been on the market for some time, however it is difficult to see how this technology can compete on the NGAN market other than when it comes to home networking, which is at best peripheral to NGAN. Another fixed network with huge potentials is the fibre access networks. NGAN is also affected by the development of wireless and mobile technologies. Wi-Fi and WiMAX and their combination seems to have huge potentials. Also satellite, broadcast and new mobile infrastructures are important in development of NGAN. In this section the most important NGAN are described and organised in two major categories, Fixed and Wireless: Fixed:
s s s s s s

ADSL ADSL2, ADSL2_PLUS & RE-ADSL2 VDSL & UDSL Cable TV PLC FTTx

Wireless:
s s s s

Wi-Fi and WiMAX ADSL2, ADSL2_PLUS & RE-ADSL2 VDSL & UDSL Cable TV

ADSL

ADSL (Asymmetric Digital Subscriber Line) is standardized such that the frequency bandwidth of regular telephony (below 4 KHz) on the access lines remains for telephony service. Broadband is transmitted on two other frequency bands; one of them is allocated to the low speed upstream channel (25 KHz to 138 KHz) and the other band is allocated to a high speed downstream channel (139 KHz to 1.1 MHz). The theoretical maximum bit rate of 8.1 Mbps is defined by the standard, but the bit rates, which can be achieved in practical implementations, depend on different parameters, for example, the distance between the household and the central, as the high frequency band of the copper line gets strongly attenuated as the distance increases. This distance dependency results in a situation whereby some households can simply not be reached by ADSL, even though they have access to PSTN infrastructures. Even in a country like Denmark, which has quite an advanced PSTN infrastructure, in mid-2004 about 5% of households could not be reached by any ADSL services and

only 70% of the population could access a 2 Mbps connection. As seen in the following, the new generations of DSL standards try to overcome these limitations and enable the DSL platform to be competitive in the future broadband market.

A D S L 2 , A D S L 2 + & R E -A D S L 2

In the ADSL2 standards advanced technologies are implemented to improve the capacity/bit rate, to establishing QoS, and also to a lesser degree to improve the coverage. Furthermore, ADSL 2 standard introduces radical innovations when it comes to power consumption, monitoring etc. The extended possibilities for monitoring and control give the operators a tool to adjust the utilization of resources, so it becomes possible to deliver reliable capacity in spite of external degrading parameters like ‘Cross talk’ and noise. Furthermore, improvement of capacity is achieved by utilizing more efficient modulation technologies, reduction of overhead, deployment of efficient coding algorithms and a number of other techniques. Over short distances it is possible to achieve bit rates of about 12 Mbps downstream and 1.2 mbps upstream. Another way of achieving higher bit rates in ADSL2 is by bonding several lines. Here at the ends of the connection, multiplexing and de-multiplexing is deployed to split a connection to several parallel connections at one end and reassemble them at the other end. ADSL2 enables the implementation of QoS by, for example, splitting the bandwidth into different channels with different characteristics, and reserving these channels for different applications. For example, it is possible to allocate a 64 Kbps of the bandwidth for the transmission of regular telephony. This makes it possible to establish a transparent communication path for the transmission of PSTN services (the so called CVoADSL) without IP conversion or conversion to other protocol. This is a technology that enables PSTN operators to deliver efficient voice and data services. This is, however, not in line with the general VoIP development, which requires interoperable technologies across different platforms. In ADSL2+ the bit rate is increased by doubling the deployed frequency bandwidth, i.e. by including the frequency band between 1.1 to 2.2 MHz. As mentioned earlier, the high frequencies get strongly attenuated as function of distance, which implies that the increase in bandwidth is only valid for short distances of under 2.4 km. Doubling of capacity can be achieved for distances less than one km. RE-ADSL2 (Reach Extended ADSL2) is designed to optimize the coverage by increasing the power used in the lower part of frequency spectrum in the upstream and downstream channels. Here it is possible to achieve coverage extension of about 900 meters, which increases the potential market for PSTN operators considerably. The coverage problem is however not solved totally, as there will still be areas where the PSTN operator cannot reach without changing of the current network.

VDSL & UDSL

VDSL enables capacities of about 52 Mbps, which are higher than the ADSL family. This is implemented by including more high frequency bandwidth in the copper cables and by deploying more efficient modulation. Furthermore, VDSL enables high speed symmetrical connections. The coverage of VDSL is on the contrary very short and is kept below 1.3 km. However, due to the aforementioned characteristics of copper lines, the capacity is very much dependent on the distance, and at the maximum distance the achievable capacity is about 13 Mbps. This implies that only in the very last part of the network (from street cabins to households) the current installed infrastructure is used, and a backbone network infrastructure must be established to supply these street cabins. This backbone network will mainly be based on optical fiber technology. Furthermore, this implies that compared to ADSL the cost of deployment is very high. There are some limitations to VDSL, for example, the co-existence between VDSL and ADSL is a challenge due to interference. Interference from AM radio is also a source of problems for VDSL. Furthermore, there is a problem of supply of electricity to the street cabins, as in regular telephony it has not been necessary to have a power supply to these cabins. When it comes to interoperability, there are some advantages to using VDSL (at least in the short term), as it will be possible to offer symmetrical 10 Mbps Ethernet connections. VDSL2 is under standardization and the aim is to offer bit rates of up to 100 Mbps. UDSL developed by Texas Instrument is the newest variant of DSL, which tries to utilize the un-utilized resources in the Copper network, and to give the PSTN operators the possibility to be competitive on the broadband market. UDSL promises aggregated bit rates of up to 200 Mbps, including 100 mbps symmetrical connections. Uni-DSL comprises the whole DSL family: ADSL, ADSL2, ADSL2+, VDSL, the coming VDSL2 standard and UDSL. Hence the platform gives the operators a flexible possibility of offering a number of different connections to their customers. However, it is important to mention that offering high bit rate connections cannot be done using the current PSTN infrastructure and, as discussed in relation to VDSL, it requires the establishment of a new infrastructure and the use of the part of PSTN networks close to the households.

CABLE TV

Cable TV infrastructure is another infrastructure with a huge installed base and with great potentials for delivery of broadband connections. The penetration of cable TV networks varies from country to country. In Denmark, for example, cable TV penetration is about 70%. A cable TV system is a distributive system, where the resources are organized as a number of 8 MHz channels for broadcast TV distribution. Cable TV systems have a huge capacity; however, the total capacity depends on how modern the system is and, consequently, on how much frequency bandwidth of the coax is utilized. When cable TV infrastructure is used for broadband provision, a number of 8 MHz channels are allocated to broadband provision. In an 8 MHz channel, it is possible to transmit between 27 and 56 Mbps depending on the deployed modulation technology and some other parameters, such as level of error correction. To enlarge the IP/broadband capacity in the cable TV system, several solutions can be used: · Using new standards with more efficient modulations technology · Modernize the cable TV system and utilize more frequencies (channels) in the system · Reallocate more channels from TV to broadband · Digitize the cable TV distribution system. Consequently, one TV service will occupy fewer frequencies and in this way it is possible to free some resources for IP services Or more radically: · Remove dedicated TV transmission and use the whole capacity for IP and also deliver TV over IP. This solution is however strongly dependent on the development of IPTV technology Cable TV infrastructure is optimally positioned in the future broadband market due to its capabilities in offering triple/multi-play services. This is because the network is optimized for TV distribution and capable of delivering broadband. Many other broadband infrastructures face a huge challenge in delivering broadcast TV. One of the weaknesses of Cable TV network in relation to broadband is that it is a shared medium, i.e. a number of users share the capacity in a network segment. Another problem which is connected to the current structure is that it is not a simple task to open the cable networks to a third party operator and establish competition. This is both due to the ‘shared medium’ aspect and because the cable TV networks are not standardized. An important element in the utilization of Cable TV structure for broadband is the introduction of VoIP with QoS support. Particularly in DOCSIS 1.1, there are specific procedures for establishing prioritization to minimize delay and jitter which are highly necessary for VoIP. The problem is, however, that because of the aforementioned problem of opening the network to third party operators, the general ‘best effort’ VoIP operators cannot take advantage of these QoS improving measures.

PLC

For the promotion of competition, the establishment of new access infrastructures has been very important (”Several pipes to the home”), and in connection to this, broadband over power lines has been discussed, especially in Europe, for many years. PLC utilizes the high frequency part of spectrum in the existing power line infrastructures. Electricity supply is provided in the low 50-60 Hz band and the frequencies over 1 MHz can be used for broadband. With regard to the capacity, PLC has been able to match DSL technologies in recent years. One of the important arguments for utilizing PLC as IP infrastructures has been the ubiquity of the physical infrastructure. Power line infrastructures have very high penetration and the idea has been that one can use this infrastructure to offer broadband in an easy way, and without establishing a totally new physical infrastructure. Another aspect is that all rooms in a household are connected to the power line infrastructure, and this allows for new and innovative services within the ‘intelligent home’ technologies paradigm. PLC has suffered from several problems with noise and interference, which are solved today to a certain degree in the low voltage part of the power line infrastructure. At EU level, the EMC-directive is the only regulatory tool which can be used to assess the level of interference, and there is no agreement for harmonizing the interference requirements within power line (and other fixed infrastructures like xDSL). The EU recommends that the member countries remove any barrier to the development of services over PLC. Even though the major technical obstacles like noise and interference are being solved, there are not that many market players (within and outside the power line business) that see any future for this technology as a means to deliver IP/broadband services. For more discussions about techno-economics of PLC, please refer to part III. In Denmark, for example, only one power Line Company has commercial trials with PLC but they are planning to out-phase it and replace it with FTTH (Fiber To The Home) connections. Other power companies that have a very solid broadband business have decided not to put any efforts for PLC technology from the very first day and solely deploy FTTx technology in their networks. This is, however, the main tendency in power companies’ involvement in the IP business and what is evident is that the power companies get more and more involved in the IP/broadband business, but mainly by focusing on fiber technology. PLC is deployed in a niche market with low scale in connection to some FTTx solutions. For example, when a big apartment complex is supplied by fiber to the cellar, connecting the apartments to this fiber requires that either the suppliers extend the fiber to all apartments, they establish a new electrical cabling to the apartments, or they use PLC to deliver broadband to the apartments. Here it has been shown that PLC is a good solution when the number of interested apartments is low. The strategy of the provider in this case is typically that the PLC is used only for Internet connectivity, and real triple/multi-play may wait until new broadband infrastructures are established.

FTTX

Optical fibers are broadband infrastructures with huge potentials. Here the physical capacity is not indicated by Mbps but by Gbps, and with regard to coverage we talk about distances of around 10 km from the central points. Even though it is possible to offer capacities of Gbps, these capacities are not implemented at the end user's site. Different reasons for this are, amongst others, cost of termination and the resource planning, as well as pricing issues at the service provider side. Optical fiber infrastructures are implemented using different architectures which can be denoted commonly: FTTx (FTTHome, FTTArea, FTTCabinet, FTTCurb etc.). The cost of deployment of the optical infrastructures is higher than other broadband technologies, but the broadband products which can be offered in the fiber infrastructures are incomparable with the traditional broadband. The development in the last couple of years shows that the implementation of fiber infrastructures is becoming more and more viable, and that power companies in particular have been very active in the area. This is mainly due to the decreasing cost of fiber, the decreasing cost of termination equipments, general liberalization and the possibilities for offering triple/multi-play.

WIFI

The wireless network standard 802.11, which has gained a lot of attention, was published by the Institute of Electrical and Electronics Engineers (IEEE) in 1999. Several variations of the standard have been published since, of which the best known is IEEE 802.11b, better known to the public as Wi-Fi (Wireless Fidelity). The 802.11b standard uses the unlicensed Industrial, Science and Medical (ISM) band. In the absence of licensing barriers, and because of the simplicity of the technology and its cost effectiveness, Wi-Fi networks have developed rapidly in both industrialized and developing countries. Indoor coverage of 50 to 100 meters is normal and depending on the standard, bit rates of 11 to 54 Mbps (in some proprietary versions even more) are possible. It is, however, important to mention that the net data capacity is far below these figures, as depicted in the following figure:

Actual capacities in Wi-Fi, IEEE802.11b (Source: Measurements carried out by Lars Staalhagen, one of the researchers

at COM center, Technical University of Denmark.)
Furthermore, the capacity in a WLAN is shared and the available capacity per user depends on the number of users connected to an access point. Wi-Fi coverage can be extended using outdoor antennas, and point to point connections can also be established using Wi-Fi.

WIMAX

WiMAX is the popular name of IEEE802.16 standard, which may become the international FWA standard. Other FWA standards have shown not to be competitive in the access networks. In some cases FWA is used for business users and in the backbone network. FWAs lack of success in the access networks is due to different reasons; among others the lack of open standards and the requirement for Line of Site in the installations. WiMAX is also developing to become mobile. It may be fixed wireless access now, but is expected to go mobile in 2008. The term FWA is also changed to BWA (Broadband Wireless Access), which encompasses both FWA and mobile wireless access.

WiMAX is, like Wi-Fi, becoming a standard, which is supported by several market actors. WiMAX is forecasted to be a simple and cheap technology with long coverage and high capacity. Coverage of 50 Km and capacity of around 70 Mbps is a reality in this technology. It is, however, important to note that the capacity offered over long distances is only a fraction of the maximum capacity. And WiMAX as access technology is offered in distances of 5 to 10 Km. WiMAX will then be a good complementary/competitive infrastructure to traditional broadband. Another important aspect is that 70 Mbps will only be achieved if frequency bandwidth of 20 MHz is allocated and assigned by the local authorities. Many regulators will probably assign smaller frequency bands to the potential WiMAX operators. A competing technology to the mobile version of WiMAX (IEEE.802.16e) is IEEE.802.20 or MobileFi.

SATELLITE

The history of satellite communications can be traced back to an article by the British science writer Arthur Clarke in the British Magazine “Wireless World”. He pointed out that if a satellite could be flung into an orbit 36,000 km above the earth, it would travel at the same speed as the rotation of the earth. It would thus appear to be stationary and would be in line of sight for sending and receiving stations at about 40% of the globe. Clark concluded that only three satellites would be needed for global communication. The first satellite (Sputnik) was sent up by the former Soviet Union in October 1957, but not in the geo-stationary orbit. The first geostationary satellite was sent up by the US in 1963. The satellites were primarily designed to enable a high-capacity transmission medium for the increasing international telephone traffic. The development of optical fiber technology was also intensified at that time. The optical fibers proved to be competitive with satellite communication in handling international telephone traffic. One of the alternative uses of the satellites that turned out to be a success was broadcasting, both as a means of distribution to the transmitters and relay stations, but also as direct broadcasting to end-consumers. It is possible to use satellite for the provision of broadband connections. Here the return path must be established through other networks, like PSTN. The technology used for implementing down-stream IP connectivity is IP data Cast (IPDC). IPDC can also be used in terrestrial broadcast networks and is seen as a viable candidate to offer broadband services to mobile devices in combination with the regular mobile networks. Another implementation of satellite networks, which is highly costly, is implementation of two-way satellite link using VSAT technology. This is mainly used as backbone technology and as access technology to business users, mainly in the parts of the world that are far from fiber backbones. Developing countries are the main users of VSAT.

DIGITAL BROADCAST INFRASTRUCTURES

Digital broadcasting denotes a broadcasting system where the broadcasting signal is digitalized. The digital signal at the end users’ site can be fed directly into the integrated digital receivers, or in a transition period, e.g. regarding TV, through a set-top-box to a regular analogue TV receiver. Digital broadcast denotes a set of standards that aim to distribute broadcast signals in digital form in a specific and standardized way. In this set of standards room is also created for the transmission of data services. Data services in Digital broadcast standards are either stand-alone data services or program related data. Digital broadcast standards are not worldwide standards and different markets apply to different standards: European DAB &DVB, US ATSC or Japanese ISDB standards. The main block of the Digital TV standards, namely video compression standard MPEG-2, is deployed in the majority of standards. Because of specific characteristics of different infrastructures, different standards apply to different infrastructures. In the European DVB standard, different sets of standards are developed for all current infrastructures (Cable, Satellite and terrestrial). DVB standards are widely used all over the world in terrestrial, cable and satellite digital TV. In some markets, combinations of different standards are also used. For analogue broadcasting each TV program requires its own set of frequencies. A transmitter distributes the program and as a transmitter; for example, terrestrial broadcasting has a limited coverage area countrywide, and regional distribution requires the deployment of several bigger and smaller transmitters. The different transmitters must use different frequencies to avoid noise or disturbance. The result is that a set of frequencies is necessary to cover a country with one TV program. For users, digital broadcast will offer many advantages over analogue broadcasting, such as better technical quality, more programs and services on a given set of frequencies, and the option of multimedia and interactive services. This development is an expression of converging media: digital broadcast platforms will integrate elements from several different media, computers, telecommunications and broadcasting. The shift to digital broadcasting is not simple, however, as it introduces a range of interrelated political, economic and technical challenges. Some of these challenges are specific to the mode of distribution: satellite, cable or terrestrial, with the latter having special problems and potentials.

The services that will be feasible in the digital broadcasting networks will go beyond traditional TV and radio and encompass a range of new services like:

· · · · ·

Enhanced text TV. By using graphical tools, hypertext etc. the text TV in digital version can be more advanced and usable. Download of software. The broadcasting networks are mostly used in the day and evening hours. The transmission capacity at night can be used to download, for example, new versions of software to the set-top-boxes. Download of newspapers. In the same way newspapers can be downloaded to the set-top-boxes. E-commerce. The products can be ordered by remote control, e.g. during advertisement. Internet on TV. Access to the Internet as known in the communication networks will not be possible because of capacity-per-user problems of digital TV networks. Here the solution can be to broadcast a limited version of Internet, e.g. sites that are seen relevant from a political/societal perspective.

As seen, there are a number of possibilities to utilize the potentials of digital broadcast networks to offer data service and therefore to use it as a tool to reduce the digital divide. As the basic technologies are now ready, solutions to two sets of regulatory issues are pertinent for the development and diffusion of terrestrial DVB. One set of issues is related to the concept of Public Service Broadcasters. In almost all countries, cultural policy considerations have given rise to privileges and obligations for a few broadcasters. There is general acceptance of the need for the continued existence of Public Service Broadcasters, but to what extent are they needed in the context of new services? There are some parallels to be drawn in the discussion of Universal Service in telecommunications, but the emphasis on content in broadcast regulation adds new dimensions to the discussion. These regulatory problems are, however, not the focus for this paper. The other set of regulatory issues is related to new facilities such as multiplexing (management of frequency sharing), Electronic Programming Guide (EPG), and Conditional Access (CA). The organization of the multiplexing function becomes crucial, as digital broadcasting allows a number of content providers to share frequencies traditionally allocated to one channel. The EPG represents users' entrance to the digital services. Especially for small language areas, the strong presence of national programs makes cultural policy and some regulation a high priority in relation to these functions. Conditional Access regulates entry to services typically via an entrance code on a PCMIA-card. From a user’s perspective, it is essential that entry control is standardized and does not require different hardware for each content provider.

WIRELESS MESH NETWORKS

In a Mesh network topology all the terminals connected to the network also act as a repeater and relay information to other users in the network. Mesh networks have been developed for a number of years for military purposes. Recently we can see signs for commercial use of the concept, especially in the Wi-Fi Mesh networks. A number of different applications can use Mesh networks. Ofcom has listed a number of applications in a recent report; however, these are only some examples: s Military, emergency and disaster relief activities which cannot depend on infrastructure. s Sensor networks s Hotspot extension of urban public wireless access s Rural Community Networks Lecture hall or convention hall networks s Integrating PAN (Personal Area Networks) devices (for example PDA phones) into the WAN s Home networks

1.5 Convergence
The traditional broadcasting and telecommunication industries have co-evolved with the developing Internet, but the technological development is making this current sectoral distinction un-sustainable. Content and service provision has already taken place across the traditional sectoral boundaries for some time. Different services can be carried on different infrastructures and the end users’ access equipment will be designed to communicate with different services. This process of fusion of content, service, infrastructure and end user equipment is denoted as convergence. This convergence process is illustrated in the Figure 1: Figure 1: Convergence Process

s

according to the European Commissions Green Paper on convergence, it can be expressed as: 'The ability of different network platforms to carry essentially similar kinds of services; and the coming together of consumer devices such as the telephone, television and personal computer'.

s

In the Green Paper it is furthermore stated that 'Convergence is not just about technology. It is about services and about new ways of doing business and of interacting with society'. This approach is interesting regarding the objective of this project to identify broader implications of the technological parameters on market development.

1.5.1 Mobile Broadcast Convergence
One of the main challenges that the mobile industry faces is the demand for increased broadband capacity that is necessary to distribute video, music, games and other digital content optimally to many mobile users at the same time. Parallel to this, the broadcast industry faces a decisive challenge in personalizing content and segmenting channels towards a still more fragmented market that, apart from digital TV and radio, includes the Internet, which must be accessed through mobile terminals. In particular, young people have their requirements for content and communication through the Internet and mobile services covered, while their consumption of the traditional TV-media is correspondingly strongly reduced. This has caused broadcasters to look for new ways to target this mobile segment by offering streaming of video and music over the net and at the same time to integrate mobile SMS services, thereby creating interactivity in relation to existing radio and TV program platforms. The ability to distribute a large number of programs and other digital content to many mobile users at the same time and to combine this with the possibilities that lie in the 3G mobile network for new interactive services and business models are all conditions that create a corresponding/congruent interest within the broadcast and mobile industry. Mobility has been propagated as one of the strengths of some of the digital terrestrial broadcast standards, including the DVB-T standard. This has, not least, been an important argument that promoters of terrestrial digital TV networks have always advocated to legitimize the use of this standard, even in countries where other multi-channel infrastructures (cable and satellite) have been reasonably well developed. The argument being that terrestrial DVB (DVB-T) in contrast to satellite and cable makes mobility and nomadic use (indoors as well as outdoors) possible. This is a valid argument as long as it concerns nomadic and mobile use of TV as we know it, which is typically used in camping vans, cars, buses and trains. However, when we discuss mobile use of personal (hand-held) terminals, there are restrictions in the DVB-T’s possibilities of supporting mobility. One of the major problems is the power consumption which will be handled when watching TV on a train or bus for example, but which will be a practical problem when using a mobile phone or PDA to watch TV. This problem has been taken care of in new mobile broadcast standards. The DVB group has standardized a new standard, DVB-H, which is based on DVB-T but reduces the power consumption and other limitations of DVB-T when it concerns mobile use. At present, there are 3 competing standards for mobile broadcast (DVB-H, DMB and MediaFLO), which create the conditions for true mobile/broadcast convergence. DVB-H is designed in the framework of the European DVB program, DMB (Digital Multimedia Broadcast) is based on DAB (digital Audio Broadcast) standard and is promoted by the Korean mobile/broadcast industry, and MediaFLO is a proprietary standard from Qualcomm. Convergence between digital broadcast and mobile services can be described as a paradigm shift, which will change radio and TV from being a broad ‘push’ media, in future, to delivering a large amount of segmented channels with targeted ‘pull’ services customized to mobile users’ constant changing demands and uses. Mobile and nomadic application will result in decisive new behavior patterns that carry great potentials for research and innovation.

1.5.2 Fixed Mobile Convergence
Fixed Mobile Convergence or Integration, FMC or FMI is a broad concept that covers various ways of integration of mobile and fixed (wired/wireless) technologies and services. FMC is not a new development and several FMC services have been on the market for the last five to seven years, but new technological and market developments have created new incentives for further development of FMC services and creation of new types of FMC services. There are different reasons for the emergence of these services:
s

A high proportion of mobile calls are made from the home and office environment.

s

At the same time the fixed operators are losing voice minutes and want to reallocate some of their traffic from mobile to their fixed network. When it comes to data capacity mobile networks are lagging far behind the fixed networks. Hence it is much more efficient to connect to the fixed network when it is possible. VoIP is gaining momentum and many broadband operators offer VoIP services. Integration of mobile telephony and VoIP over broadband opens up new possibilities for competition in the Voice market.

s

s

Due to these reasons FMC is being developed massively in the near future. However, the efficient provision of FMC is connected to the maturing of technologies deployed in the backbone network and is highly connected to the development of NGN.

1.5.3 Converged Services
Services like VoIP and IPTV are important drivers of future ICT infrastructures. The following two sub-sections describe these two services.

VOIP

VoIP is an application that uses IP infrastructures, including the Internet to transmit voice telephony from point A to point B. For a long time, POTS (Plain Old Telephony Services) was seen as a natural monopoly. In the new regulatory paradigm, it is generally accepted that the networks must be opened up for competition through unbundling and interconnection regulation. However, within the traditional telecom paradigm, competition will at best exist between a few actors in an oligopolistic market. The central reason for this has its roots in the technological architecture of infrastructure and service development platforms. The POTS network is a dedicated network, which is optimized for voice communication. Because of the deployed technology and the way POTS services have historically been organized, a centralized structure has been implemented to offer POTS. Two network layers are deployed in parallel in order to establish a network connection and to transmit services between points A and B, the so called transport and signaling/control layers. Consequently, service creation and provision require access to both the control/signaling layer and the transport layer of the network, which in turn requires access to the whole telecom infrastructure. Even though interconnection to the POTS networks is possible, there are still large entry barriers for newcomers to offer services in the POTS networks. The precondition for service provision in POTS is access to all infrastructure and services development platforms, which requires huge investments. Using VoIP has gradually changed this situation and, through the convergence process, has opened up new conditions for service development. Using VoIP technology and the general Internet as backbone, new providers can offer competitive prices, particularly for long distance and for international calls. The transmission of the service over long distances within the Internet is much cheaper than keeping the service within POTS, with its distance-related cost structure and interconnection pricing schemes. The entry barriers for these service providers are lower and the number of them is increasing, contributing to the overall competition in the public voice market.

IPTV

For a number of years, Video and Audio services have been distributed within the IP-based network, including the Internet, using streaming technologies or by downloading the video/audio materials. IP TV/radio denotes the delivering of TV/radio over IP protocol. IPTV can be transmitted in different networks that are based on the IP protocol. One of the major IP networks is the Internet, but Internet is not the only IP network. Dedicated/Managed IP networks can be established, and in current networks the IP protocol can be used widely without having specific relations to the Internet. One of the examples for dedicated networks is cooperative networks or Local Area Networks in the firms and residential areas. Managed IP networks have an obvious advantage for the distribution of IPTV compared to the Internet, because it is possible to maintain certain levels of QoS in these networks. In Managed networks, the network provider can allocate resources such that different services/applications perform in the most optimal way, where in the Internet it is a complex issue to guarantee specific level of service quality. RELATED INFORMATION Hot Topics: IPTV

1.6 Information Society Technologies

The third wave of technologies builds on the technologies of the first and second waves and implements these technologies, broadly resulting in the use of ICT in other socio-economic sectors with decisive influence on efficiency and quality in the production processes. This is called “Information Society Technologies”. Examples on the deployment of ICT in private and public sectors include E-banking, Ehealth, E-government, E-learning and a range of other E-based processes/activities. These implementations are then likely to give rise to the further advanced development of infrastructure networks, including ubiquitous networks, the portable internet and the automated Internet of things, rather than people. Furthermore, many new technologies are expected to be smaller scale and cheaper to deploy, so this will change investment cycles and patterns. Smaller players will be able to enter markets and fuel network expansion with relatively small scale investments. The key elements of both the second wave and the third wave of technological changes thus impact the techno-policy environment. Pursuing the overall more multi-faceted and complicated objectives will, however, also influence the technological trends. This interrelationship implies that the overall problem area of regulation includes a list of parameters resulting in multi-dimensional success criteria, compared to the more simple success criteria of securing competition and universal service in the first bundle of reforms. This is not to argue that the first bundle of reforms has succeeded in introducing full competition as the general market structure. This is still an important issue in most markets, but the second and third wave of technology introduces new challenges that have to be addressed by regulators based on national policies for the development of ICT.

1.6.1 The role of ICTs in other sectors
New organizational forms developed in interaction with ICTs (Information and Communication Technologies) are generally seen as carrying the real socio-economic benefits of these technologies, as illustrated in the "Third Wave" above. It relates clearly to the discussions in WSIS on providing a roadmap to bring the benefits of ICT to underserved economies.[1] The WSIS activities aim at bringing the benefits of information and communication technologies to everyone. They cover a very broad area including almost all socio-economic activities with major categories of the project including:[2]
s s s s s s s s s s s

national e-strategies; multi-stakeholder projects involving the private sector in partnership with public bodies; infrastructure projects; broadening access to ICTs; international and regional cooperation; access to information - create online libraries and provide open access to public and research data; policy and regulation; capacity-building; online education, medicine, business, government; security; and cultural issues.

This very broad approach to promoting the benefits of ICTs can be seen as an illustration of the idea behind the concept of the "Third Wave": ICTs basically have qualities as one of the emerging generic technologies, the others being biotechnology and nano-technology. There are two main dimensions in the generic aspect of the ICTs; One is related to the effects on the potential commercialization of services, and the other to the effects on the internal structures of companies and associated transaction costs, often discussed as "virtual organizations."

[1] Proposed at the World Summit on the Information Society, Phase I, Geneva, December 2003 and confirmed at Phase II, Tunis, November 2005. [2] http://www.itu.int/wsis/tunis/newsroom/background/wsis-stocktaking.html

MARKETED SERVICES

It has, however, become clear that even if ICTs are truly generic,[1] their impact have been most profound on knowledge and informationintensive services like entertainment/broadcasting, consultancy and banking, publishing and education. Before the development of new ICTs, services typically required simultaneousness in production and consumption, even in the informationintensive services mentioned. This situation has completely changed now that services can be digitalized and transmitted on communications networks. For informationintensive service elements, the introduction of computerization creates possibilities for dividing the processes of service production into

different components, i.e. division of labor, and for transporting and assembling the components at various places, i.e. international division of labor and trade. This opens a window of opportunity for developing countries to enter the international production process at a new level as actors in a global service sector. India has been the prime example, with back office functions, call centers, airline ticketing and software production, but other countries are following. This is, however, not a straightforward, fully technically determined process. The technical possibility of delivering a service to a party located at a distance is a necessary element and a precondition for the division of labor and trade in the classic and narrow sense of the word. However, the sheer technical possibilities are not sufficient to develop and realize the potentials. The economic and institutional framework must accommodate for and allow the new possibilities enabled by technology. Very basically, the electronic processing of information presupposes a codification and division of information-intensive working procedures that hitherto have been embodied in individuals[2]. Once divided into components, the introduction of telecommunication technologies enables the transporting of information-intensive service elements. Having established a technical possibility for transportation of service elements, the transaction procedures and legal factors will influence the realization of the potentials for transport. This means that the ongoing liberalization processes, for example, in the WTO, that cover various attempts to increase the level of competition by means of deregulation of the markets for services, are crucial for the actual impact of the potentials for increasing international division of labor enabled by ICTs. Even if all the technical, economic and political/cultural factors are positioned to enable the international division of labor, trade is not automatically happening. At least one of the parties involved must usually benefit in an economic sense, and they must be interested and capable of agreeing to trade.

The introduction of ICT in the service industries as sketched out above begins a process that is similar, but not identical, to the industrialization of the material producing industries in the 1700- 1800 years. As it was the case in the traditional industrialization process, the socio-economic implications are not limited to the division of labor and industrialization. The development in ICT communication networks and services also influences the organization of firms and their relations to the market.

[1] Discussed in the literature as, e.g., 'pervasive computing.' [2] Jagdish Bhagwati discusses the process of disembodiment of services in a well-known article: "Splintering and Disembodiment of Services and Developing Countries", The World Economy, Vol.7, no.2, 1984, pp. 133-143. It is true that electronic data processing allows for many functions to be joined together. However, this presupposes a division of production processes as in goods production where equipment/ machinery can integrate different processes, but where a prior disintegration is the basis.

VIRTUAL ORGANIZATIONS

The possibilities for new organization forms are attracting growing interest from business managers and researchers. These new organization forms – often referred to as “virtual” organizations – are posited as radical departures from the traditional, hierarchic, bureaucratic and co-located modes of organizing that dominated the twentieth century. In contrast, the characteristics of the new, virtual organization forms are seen to be dynamic, networked, distributed, digital, flexible, collaborative and innovative. There is a strong focus on the ability to operate in an increasingly distributed and networked manner, with people, resources and activities located in different parts of the country or across the globe. The growing interest in these new virtual and networked organization forms is motivated by fundamental macroeconomic, legal and cultural changes associated with the ongoing processes of globalization. All these changes challenge the traditional modes of organizing, and force corporations and their managers to rethink and revise their ways of organizing work and conducting business in the new, global economy. This creates new opportunities for developing countries, as discussed above, but it is also challenging the possibilities for creating national strategies – e.g. for ICT-development – in a global economy.

Pervasiveness of ICT services

1.7 Disruptive Technologies
The term disruptive technology (or disruptive innovation) has reached increasing popularity during the past 10 years since the publication in 1997 of ‘The Innovator’s Dilemma’ by Clayton Christensen. He systematized a conception of the term, where it came to mean technologies that, at an initial phase, may have a lower utility than existing technologies but which eventually will win the market because of their unfolding advantages and displace existing technologies. If a technology is to be disruptive, it will have to be a technology that incumbents in the market will not be inclined to introduce – not because they don’t have the technological competences, but because the new technologies are not sufficiently profitable and/or do not fit into the product portfolio of the incumbents. If a new technology, on the other hand, builds on and adds to exiting technology solutions used by incumbents and, consequently, will be introduced by these incumbents, the term sustaining technology is used. The regulatory implications are obvious. If new technologies are taken up by the market but not by the incumbents, more competition will be introduced and less regulation is likely to be necessary. Not only can regulation be scaled down once new and disruptive technologies gain important market shares, but regulation should also pave the way for the introduction of new technologies that will have a disruptive effect on the market and no only constitute sustaining innovations. However, the ability of incumbents to take up new technologies is often underestimated. Incumbents may not be the first in the market to introduce new technologies that do not build on existing technology solutions. But once it becomes clear that new technologies gain sizeable market shares or have obvious potentials, it is often seen that incumbents are able to include new technologies in their product portfolio – even though they may, at first, have been considered as potentially disruptive. Wi-Fi is up for discussion as an example. In some cases Wi-Fi operators disrupt the market; in other cases Wi-Fi is offered by incumbents. When Wi-Fi first hit the markets, there was a vivid discussion as to the potential disruptive character of this technology. It could eventually substitute for existing mobile solutions and thus constitute a threat to existing mobile operators. However, Wi-Fi has also been taken up by many incumbent mobile operators and complements the existing mobile services offered. This does not mean that Wi-Fi or other wireless solutions do not, in different cases. substitute for existing mobile solutions, but it means that Wi-Fi does not necessarily disrupt the market. Wi-Fi often complements the services offered by mobile incumbents.

There are, nevertheless, technologies that turn out to disrupt the markets. History has documented this. However, it is difficult ex ante to determine whether a technology is disruptive or sustaining. This will have to be determined ex post. The issue of disruption is not only a technology issue. It cannot be determined whether a technology is disruptive just by looking at its technology characteristics. Disruption is a market issue and the degree of disruptiveness will be determined on the market by the market players. Still, public policy and regulation should seek to pave the way for the introduction of as many new and potentially disruptive technologies as possible. Though incumbents will be able to adopt the new technologies, new technologies will also open the market to new and alternative players. The end-result may eventually be major disruptions through cumulative innovations even though it did not seem that way at the beginning. It will, at any rate, be an advantage to the users to have as many alternative offers as possible, technologically as well as with respect to supplying market players. RELATED INFORMATION VoIP IPTV Wi-Fi WiMAX

2 Market and Regulation
This section analyzes the implications of new technology trends on specific market and regulatory issues. Following technology trends are considered: 1. 2. 3. 4. Internet Mobile communication Next Generation Access Networks (NGAN) Convergence

The impacts of technology trends on regulation are either direct or mediated via their impact on market conditions. Therefore both market and regulatory implications are included. The section is organized as follows:

Direct and indirect technology implications Market Structure
s s s s

Service innovations Network innovations Vertical Separation – the Scope for new Business Models Horizontal integration

Price regulation Interconnection (Re) Licensing Universal Access Spectrum management Numbering Regulatory implications by technology trend is summarized here. Link to Key points and recommendations by topic

2.1 Direct and indirect technology implications
Regulation in the areas listed above is affected by technological development in three different ways. First, there is a direct impact: New technologies lead to the development of new services and modes of delivery unforeseen by existing regulation. Use of IP telephony, for

instance, raises new issues with regard to numbering and emergency services. Second new technologies affect the overall market structure and the level of competition by changing conditions for supply, which again affect the need for regulation. An example of this is facility-based competition enabled through the use of cable broadcasting networks for the provision of Internet access. Third, new technological opportunities create a demand for new types of services, which again affect the overall market structure. The two most prominent examples are the introduction of mobile services and the introduction of the World Wide Web. The demand for mobile services created an entirely new market and enabled a number of new entrants providing their services in competition with the incumbent operators. The Internet has been the driving force behind the growing demand for broadband connections, and this has made it more attractive for alternative network providers, such as cable and electricity companies, to enter the telecom business. It is, however often difficult to draw a sharp distinction between these three different kinds of impact, as the same technology may affect market structure and regulatory needs in several different ways simultaneously. However, it is necessary to analyze the technology impact on the overall market structure before the implications for specific regulation areas can be discussed. Figure 1: Direct and indirect regulatory implications

Table 1: Examples of Direct and Indirect implications of technologies on regulation areas Technology Direct Impact on Impact on Indirect regulatory regulatory supply/demand market implications implications structure Cellular Mobile Licensing, Roaming Demand for a type of voice service Drives demand for broadband Establishment of new competing operators Alternative solutions for Internet access become economically viable and create market opportunities for new kinds of providers Price regulation, Interconnection a.o. Interconnection, spectrum management a.o.

WWW

IP numbers, Domain names

VoIP

Numbering

New business models for voice communication

Separation of network and service provision

Funding of USO

2.2 Implications of technology for the economics of the overall Market Structure
Technology impacts the telecom market in many different ways. A distinction is often made between product innovations and process innovations. Product innovations are innovations creating opportunities for the delivery of new products to the customers, and process innovations creating more efficient and cost effective processes of production. In the telecom sector this distinction may not be as clear as in many other sectors. Innovations in network technologies may be seen as process innovations, as they enable expansion of the network capacity, i.e. more production of the same product, but at cheaper prices. On the other hand, expansion of capacity allows for a wide range of new applications of the network and therefore creates a completely new type of infrastructure. In this context it is therefore more relevant to distinguish between innovations in network technologies and innovations related to service development. Furthermore, it is relevant to study how innovations affect the vertical relationship between networks and services, and horizontal relationships between different sub-sectors of the telecom industry. This leaves us with four different types of technology implications on market structure: 1) Service innovations Up to the mid-90’s telephony was the absolute dominating telecommunication service. As late as 1994, the revenue from PSTN constituted more than 75% of the total service revenue, and in developing countries the figure was even higher. Since then, high growth in the demand for Internet and mobile services has contributed to the overall growth and changed the telecom market from a single service to a multi-service market, where fixed telephony contributes less than 40% of the total service revenue. 2) Network innovations Innovations in network technologies have not only made substantial cost savings possible, but have also created new possible alternatives to the fixed copper based telecom network operated by the incumbent operators. While telecom networks a few decades ago were seen as natural monopolies with a limited scope for competition, it has become feasible to establish competition in most segments of the network. However, regulatory intervention ensuring unbundling and open access to essential facilities may be necessary if this opportunity for real competition is to be realized. 3) Vertical separation Digitalization of both networks and services has made it easier to separate network and service provision. This enables the development of a market structure with a vertical separation of network operators and service providers. New business models focusing on the provision of a particular service or network component are developed as alternatives to the end to end concept, where all service components are provided by the same operator. 4) Horizontal integration Both service and network innovations have caused a blurring of the boundaries of the telecom sector. A wide range of new telecom service products have been created. Some of these products incorporate service elements from other sectors, such as IT or broadcasting. At the same time digitalization and expansion of network capacities enables transmission of IT, telecom and broadcasting services on the same networks (network convergence). These implications correspond to a certain extent to the four technology trends, mobile communication, next generation access networks, IP and other packet switched infrastructures, and convergence. They are however not completely identical. For instance, there are other important service innovations in addition to mobile services, and vertical separation involves packet switched as well as circuit switched networks. Figure 1: Relation between technology trends and market implications

Like the technology trends, the four implications are interrelated. The horizontal integration and vertical separation affect the boundaries of the telecom market and redefine the boundaries of the sub-sectors within the ICT area. Furthermore, the impact of these implications depends on both capacities and cost profiles for the various ICT infrastructures, while the economic viability of the various networks depends on what types of services they are being used for in the future.

2.2.1 Service innovations
The technology trends described in section 1 of this module include development of a host of new services in addition to fixed telephony (PSTN). In spite of their importance for electronic banking and other business services, none of these services was able to generate revenues comparable with those generated from PSTN before the emergence of mobile communication and the Internet.

INTERNET

In spite of impressive growth rates of the Internet with respect to the number of users and number of subscribers, the revenue collected from provision of Internet services is far lower than that of mobile communication, and the penetration is also substantially lower, particularly in developing countries. The digital divide between developed and developing countries is much wider with regard to access to the Internet than for mobile communication. While the penetration of mobile subscribers is four times higher in developed countries than in developing countries, the penetration of Internet subscribers is eight times higher. In addition to this, the bandwidth available is higher in developed countries.

Several low income countries have, however, achieved comparable high penetration rates. These include Bulgaria, Romania, Belarus, Guyana, São Tomé and Principe and Moldova. Countries like Jamaica, Chile, Barbados, Latvia, Estonia, Slovenia, Taiwan, Malaysia, Singapore and Hong Kong have also obtained higher penetration rates than other countries with comparative income levels. According to

OECD, a major cause to their success is their implementation of regulatory reforms, including market liberalization, competition and an independent regulator (OECD: Communications Outlook 2005). The importance of the Internet as a telecom service goes far beyond its share in the telecom revenue. The Internet has provided the infrastructure for an entirely new way of doing business and has in this way facilitated the creation of a new business service sector. In addition to this provision of Internet services is the driving force behind the growth of demand for broadband. This contributes to more facility-based competition as the demand for broadband has driven the market for alternative access network infrastructures, such as hybrid cable TV networks and Wi-Fi. The economics of such infrastructures are further explored in the section below on network innovations. The Internet is also instrumental to both convergence and vertical separation. The Internet provides a common platform for a wide range of services and thereby facilitates way convergence between different services. Provision of IP/Internet services can be done through the use of several different types of network infrastructures, and many Internet service providers do not own their own physical network infrastructure. One of the most debated services, with a profound impact on the whole structure of the telecom sector, is VoIP, although in its infancy, this service poses a threat to the traditional business model for voice telephony. VoIP is further discussed in the section on vertical separation.

KEY POINTS AND RECOMMENDATIONS
s

s

s

The telecom market has developed from a single service to a multi-service market. This raises a number of questions for regulators and a need for reformulation of the concept of network access. Is access to fixed telephony more important than access to mobile or Internet services? Can barriers to access be reduced for all services? Should different services be regulated differently? Mobile services have become more widespread than fixed services. Mobile markets are generally more competitive than those for fixed services. Mobile services are often provided through the use of business models different from those used for provision of fixed telephony services. This raises new set of issues with regard to consumer protection and competition. Internet is less widespread than mobile services in particular in developing countries. The socio-economic impact of Internet services can hardly be overestimated. The Internet provides a common platform for a wide set of services such as World Wide Web and email, services widely used by almost any type of business. Ensuring wide access to this service is therefore extremely important for both economic and social development.

MOBILE COMMUNICATION

Mobile communication services have from the mid-90’s developed from being a supplement to fixed telephony reserved for a selected few to become an industry in its own right, and partly independent of the fixed telecom operators. In 1994 the revenue from mobile services constituted less than 10% of the global telecom service market, and the number of subscribers was around 50 million, compared to more than 600 million fixed-line subscribers. In 2002 the number of mobile subscribers exceeded the number of fixed-line subscribers, although the revenue was still somewhat lower than that coming from fixed services. Particularly in Africa, mobile communication has led to a significant improvement in access to telecom services, as the penetration of mobile phones is almost three times as high as the penetration of fixed telephony. This is compared to the rest of the world, where the number of mobile subscribers is only 50% higher. Mobile has also played a remarkable role in improving access to telecom facilities in low income countries in Asia. Density of fixed and mobile services in African Region

The rapid take-up of mobile communication services, particularly in developing countries, is due to two different factors: 1. 2. Establishing mobile communication networks is less costly and the necessary investments in infrastructure are lower than for fixed networks. Mobile markets have from the beginning been more competitive and less regulated than markets for fixed services.

As discussed further in the section on network innovations, mobile networks can be established at much lower costs than fixed networks. This relates to the fact that a substantial part of the infrastructure investments in a fixed network lies in the local loop. The combination of lower costs and the fact that mobile communication in many countries has been introduced after liberalization of the telecom sector, has led to a more competitive market structure right from the beginning. It follows from the figure that in particular in the African region, mobile markets are far more competitive than fixed-line markets. In many countries, mobile services were first introduced by a new entrant, often with international backing. The national interests in maintaining a monopoly market for mobile services have therefore been smaller than for fixed services. Although tariffs are still higher than for PSTN, mobile communication provides an alternative to fixed lines and contributes therefore to a more competitive environment in the entire telecom sector. Another positive impact relates to a high degree of internationalization among mobile operators. A number of mobile operators, first of all Vodafone and Orange, are established in several countries. International mobile operators with access to international capital markets account for a substantial part of mobile network investments in developing countries. In Africa, 26 out of 52 million mobile subscribers in Africa are served by foreign operators[1]. Finally, the introduction of the prepaid business model seems to have been a key for stimulating the demand in the least developed countries[2]. Prepaid services are more affordable for low volume costumers, and the service can be provided without rating of consumers’ creditworthiness. Status of competition for fixed lines and mobile by region

Source: OECD Communications Outlook 2005, OECD 2005 http://ocde.p4.siteinternet.com/publications/doifiles/932005011P1_ch11e.xls ENDNOTES [1] This figure is calculated on basis of J. Whalley & P. Curwen (2005), on the assumption that Vodacom (partly owned by Vodafone) is seen as a foreign company, 2003 figures. [2] ITU: The Application of Information and Communication Technologies in the Least Developed Countries for Sustained Economic Growth. Geneva 2004.

2.2.2 Network innovations

Techno-economic characteristics of infrastructures Network innovations are constantly reducing costs and expanding the capacities of networks. However, innovations do not only affect total cost and capacities in general, they also affect the relative costs of different network technologies and of each network element. Both are affecting the overall trends in the structure of the supply side of the market for provision network access. Cost characteristics of telecom networks are affected by new technologies in two ways:
s s

New technologies are being implemented in existing networks, affecting cost structures of these networks New types of networks are being established through the use of new technologies. These networks may be entirely new or based on existing facilities in other networks.

This section provides an overview of trends in the cost structure of existing and new types of communication networks. Copper Availability High Cable Varies (mainly in cities) Telcos/ cable operators/ municipalities Medium Broadcast Distance/ connections Long High fixed costs Very high High Medium/ high Optical fibre Low Power line Low Mobile High Wireless Medium/ low Often SMEs Satellite High

Ownership

Incumbent Telco

Capacity Major Services Costs drivers Lifetime Fixed/ variable costs Share of access costs Economies of density Economies of scope (bandwidth)

Medium/ low POTS Distance/ connections Very long High fixed costs Very high High Medium/ high

Telcos/ power companies/ municipalities High VPN/ Internet Distance/ connections Long High fixed costs High High Very high

Power companies

Telcos

Satellite operators

Low Internet Distance Short Low fixed costs Medium/ high Medium/ high Low

Low Mobile com. Area/ capacity Short Medium fixed costs Low Low/ medium Low/ medium

Medium/ high Internet/ VoIP. Area/ capacity Short Medium/ low fixed costs Low Low/ medium Low/ medium

High Broadcast Capacity Medium High fixed costs Very low Low Low

These infrastructures impact the market in three different ways: s They can extend network reach and facilitate increased network access to basic services s They can facilitate growth in availability and penetration of broadband services. s They can facilitate real competition and lower prices An important techno-economic issue here is how these different infrastructures can support not only basic telephony services, but also TV broadcast and broadband services, and how this will affect cost profiles as well as total costs per subscriber. The results are highly dependent on the architectures used in existing network facilities, density of subscribers and the size of the uptake. With regard to basic telecom services wireless networks solutions are particularly relevant. Optical fibers may however also be relevant in the core network. Copper-based networks are usually not cost effective when it concerns green field investments, but may be considered for adding new subscribers in areas already covered. It follows from the box below that both DSL and Cable modem are attractive solutions if the right facilities are available. Comparative capital costs per subscriber for various broadband technologies The EU funded research project BREAD- Broadband for All in Europe has made a cost comparison of different broadband access technologies. Comparative capital expenditures per subscriber Urban Suburban Rural ADSL 1.0 1.1 2.2 Cable 2.4 7.3 20.1 FTTH 12.8 16.0 23.2 WIMAX 3.7 5.5 6.7 FWA 11.8 11.7 12.1 Note: 25% uptake is assumed. 1.0 represents the cost per subscriber for provision of ADSL service in urban area. Source: BREAD – Broadband for All in Europe

It follows that wireless technologies are most advantageous in low density areas, while solutions using existing copper-based facilities are the most advantageous, particularly in urban areas. The trends in network cost structures affect the supply structures on the markets for telecommunications networks and services in a number of different ways. First of all the establishment of high capacity core networks based on optical fibers and with very low traffic unit costs has almost eliminated distance as an independent cost driver. The costs of telecom services depend only to a very little extent on distance. The costs may however still depend on location, as these high capacity networks are not equally accessible from everywhere. The cost of a high capacity fiber connection varies by a factor of 39 within Europe, and prices may vary even more if the more remote regions in Africa are included in the comparison. Access to global network facilities may still constitute a substantial part of the total costs in some areas. This implies that markets for services and applications are becoming more international and more competitive. Technology does also foster more competition for some network services. It has become more feasible for new entrants to develop their own core network facilities, particularly if dark fibers are available. The costs of the local loop will still account for a major part of the costs, and facility-based competition in this segment of the network will remain too limited. However, these costs also decrease with the availability of wireless access services. In particular, low-density areas can benefit from this development. An important question is what role the competitive advantage of ownership of the existing, mainly copper-based, network infrastructures will play in the future. Clearly the emergence of wireless alternatives is putting the incumbent operators under pressure. Two 2G networks are offering an alternative to fixed voice telephony but have not yet been able to compete on prices, and their ability to provide Internet access is limited. Demand for Internet access will become an important driver for competition between different network structures in the future, and the success of the various network options will depend on their ability to match future demands for bandwidth on IP services. Dial-up services are increasingly being replaced by broadband connections. Demands for very high bandwidth favor the use of optical fibers, as these, once installed, can offer very high capacity at very low traffic unit costs. Low demands for bandwidth are more easily fulfilled by wireless services, where the initial investments can be made at far lower costs. Upgrade of the existing copper-based network facilities is particularly relevant in high density regions. xDSL and cable are at present the most widespread technologies used for provision of broadband (figure III 8). Incumbent operators dominate provision of xDSL, although the market is becoming more competitive. New entrants have a much larger market share in cable and other technologies such as Wi-Fi and optical solutions, and the level of competition on the market for broadband services is in general much more competitive than the PSTN market. Internet access by type 1998-2003 and broadband by technology 2003

Note: Other includes Wi-Fi, Metro Ethernet, Fixed Wireless access, Satellite etc. Source: ITU: Trends in Telecommunication Reform 2004/05. Geneva 2004 Broadband Market Shares of incumbents and new entrants by technology in EU (Jan. 2005)

Note: 79.1% of the BB connections were using xDSL Source: European Commission: Broadband access in the EU: situation at 1 January 2005. COCOM05-12 FINAL Brussels, 1 June 2005.

COPPER-B A S E D T E L E C O M N E T W O R K

The major advantage of this network is that it is widely available. In industrialized countries there is almost universal access to this network, and in all countries the network can be accessed in the major cities. In those areas, use of existing infrastructure facilities is still very competitive in providing most types of services. However, in areas not covered by copper-based networks, use of other network technologies is likely to offer a cost efficient alternative. The network is usually operated by the incumbent operator, which in many cases is fully or partly owned by the public. The copper-based networks are established in markets with monopoly, and are therefore designed for covering the entire market. Efforts have been made to introduce competition through demands for unbundling of facilities and interconnection with other networks. To establish a copper-based network demands substantial long term investments in particular in the access network. Existing copperbased networks have gradually been expanded during several decades and their architectures are not optimized with regard to use of current technologies. If an entirely new network were to be built today, it would not be based on use of copper-based technologies, and the design would therefore be very different from those of today’s copper-based networks operated by the incumbent operators. One problem is that networks are designed mainly for carrying POTS, while a growing share of the traffic is based on IP or other data communication protocols, and in some areas there are problems with capacity and quality of service. Network costs depend on the kind of network design applied. The most recent detailed cost analyses are those made for calculation of interconnection charges using forward-looking cost accounting methods, such as long run average incremental costs (LRAIC), used within the EU, or total element long run incremental costs (TELRIC), which are used in the US. These types of cost analyses are to a certain extent based on existing basic network designs, but with use of the most recent transmission technologies. It should be noted that the outcome of such analyses highly depends on the assumptions made, and very different results may be obtained for different network architectures and geo-types. A survey of cost analyses made shows that access costs constitute 35-50% of the total network costs. Here the major cost driver is total cable length, which again depends on the number of connections and the density of customers. Therefore it may cost as much as five times more to connect customers in rural areas than in metropolitan areas. A major part of the costs are related to the laying of cables underground. Here substantial savings can be obtained through the use of ducts that can be shared between several cables. The digging costs are highly dependent on the geo-types. Here it should be noted that digging costs per km often are much higher in metropolitan areas than in the open land. The costs of copper-based networks are affected by the following technological advances:

s

s

Today the copper-based trunk network is replaced by an optical network, while the access network is still based on copper lines. Installation of fibers has reduced the cost of capacity in trunk networks considerably. Digitalization of switching facilities and use of packet switched transmission technologies has reduced switching costs.

Implementation of Next Generation Access Network technologies will reduce transmission costs even further. Altogether these trends imply that the cost of the copper-based access network constitutes a still larger share of the total network costs. However, technological advances are also taking in this part of the network. As discussed in other sub-section, alternative access networks offering lower costs or higher capacity have been developed, but in areas where investments in copper-based access networks have already been made, the development of technologies offering more capacity on existing access facilities is at least as important. It is possible to upgrade the copper-based access networks to carry high-speed services through the use of xDSL technologies; the possible capacity depends on the length of the copper cables and the quality of the network. The bandwidths offered here range from 128 kbps to 10 Mbps. xDSL is the most widespread access technology for broadband access as 57% of all broadband connections use xDSL (end 2003). Provision of higher bandwidth will often, but not always, require more investments in the access network, and of course more capacity in the core network. By all means, the capacity is much lower than in optical networks. On the other hand, the additional investments needed for upgrading the network are only a fraction of what is needed for the establishment of an optical network.

Indicative cost per line in a fixed telecom network
Subscriber connection Switching Transmission Land and building Total Cost in % 50 30 20 10 100

CABLE TV NETWORKS (HFC)

Cable TV networks were first established to provide broadcast services. The availability of cable TV networks varies from country to country. In some countries, like the Netherlands and Belgium, this service is available to 80-90% of the households. In some countries this service is only available in major housing complexes, while other countries are virtually uncovered. The ownership structure is also very different in different countries and in different areas. Sometimes the networks are owned by local communities, sometimes by the municipalities and sometimes by private companies, including the incumbent operators. The networks are built for distribution of TV signals and the capacity in the local loop is therefore higher than in the copper-based telecom networks. However, the networks are designed for carrying distributive services only and have no switching capabilities. Furthermore, the capacity is not dimensioned to carry different signals to every user. If upgraded, cable networks are able to provide a higher capacity, but in contrast to xDSL users, they do not have their own channel all the way to the local exchange. This implies that the speed will be reduced if many users are connected at the same time. The total costs for cable TV networks are somewhat lower than for a copper-based telecom network. The costs for an upgrade in order to provide data services are comparable with those for upgrade of telecom networks to xDSL. However, the share of fixed costs is lower. In particular in high density areas, cable networks can offer a cost efficient alternative to xDSL services. In 2003, cable provided 37% of all broadband access connections. As the availability of this service is much lower than that of xDSL, the market share is higher than for xDSL in those areas where this technology is available.

OPTICAL FIBRE NETWORKS

Optical fiber networks have replaced copper cables in large parts of the core network, but up to now optical fibers have not been widely available in the local loop. Optical fibers are today not more expensive than copper cables, but as with copper cables, laying the fibers

underground constitutes a substantial part of the costs. Therefore, the replacement of copper-based cables with fibers is a costly affair. In addition to this, the end equipment converting optical signal to electrical signals and vice versa is also expensive. Deployment costs including components, civil works and installation have in Europe decreased from 1250-1500€ per connection in 2001 to 600-800€ in 2006 [1]. The most expensive component is labor costs related to laying down fibers. This component will be far cheaper in countries with low unit costs for labor. Due to the very high capacity, the unit costs per bit are very low, and much lower than in copper lines and in wireless systems. Optical fibers have first been installed in the core network. This has reduced the costs of long distance communication to less than 1% of the former costs per traffic unit. This indicates that the costs of fixed network services are concentrated in the local loop and convergence between the costs of local and long distance communication. Incumbent operators are gradually replacing their copper lines with optical fibers, beginning with the trunk network and gradually approaching customer premises. However, other types of telecom operator are also installing optical fibers, particularly in the core network. It is also common to install fibers in connection with construction work related to other network infrastructures such as electricity, water and gas. In contrast to network cables made of copper, optical fibers can be integrated in power cables, as the transmission of electricity power does not interfere with the light signals in the optical fibers. Public utility providers, in particular power companies, will therefore become new actors on the market for provision of communication network facilities (see practice note on NESA). As the fibers themselves constitute only a small fraction of the total costs, it is common to lay down many more fibers than necessary when an optical network is constructed. These fibers will not be connected to the network and no transmission equipment will be installed in the end of the fibers. This implies that once optical fibers have been installed, it will be possible to upgrade capacity. As the maximum capacity per fiber increases constantly, it is unlikely that additional cables will be needed in the foreseeable future. It follows that the installation of an optical fiber network involves substantial long-term fixed costs, and that there are very high levels of economies of scale and scope. However, the costs of end equipment are to a large extent variable, as the costs of transmitters, converters etc. depend on the capacity. The wide availability of dark fibers has created a market for such facilities (see practice note What is a dark fibre?). It is therefore in certain areas possible to buy or lease fiber facilities. As the additional costs of upgrading existing dark fibers to a working communication network are much lower than building a new network, it becomes possible for new entrants to establish their own networks at very competitive prices. This is however possible only at destinations where fibers are readily available. Therefore, the economics of fibers imply that unit costs highly depend on the supply of fiber capacity. Transmissions between major destinations, such the metropolises of Western Europe and the US may be far cheaper than connecting two small neighboring cities. A study of the European market for broadband connections shows prices which are largely independent on distance but highly dependent of the level of competition. The price for a fiber broadband connection in the most expensive monopoly markets were up to 39 times the price in the most competitive markets[2]. In conclusion, the techno-economic characteristics of optical fibers imply an increasing price gap between major routes and less developed areas. This trend is exacerbated by the supply of dark fibers offered at very competitive prices, not even covering investment costs.

[1] Albert Grooten: Reducing CAPEX and OPEX in FTTH networks . Europe at the speed of Light, Vienna 25-26 January. [2] Report on present status of international connectivity in Europe and to other continents. SERENATE deliverable no. 6, 2003.

A technical description of optical fibre access networks can be found at FTTx.
POWER LINE COMMUNICATION

Power line communication is mainly used for in-house cabling and may in this case become a cost effective alternative to other wired local loop services. Power lines are even more widespread than telecom lines. The potential for using power lines for telecom access is therefore enormous. More than 100 trials have been implemented in at least 40 different countries within the past decade. In spite of this, the penetration of this technology for use in the provision of local loop is still very limited. In Europe in 2004 fewer than 20,000 actual users, and in the US the number was even lower[1]. One of the problems has been high costs of equipment. Power line communication is today mainly used for inhouse cabling and may in this case become a cost effective alternative to other wired solutions.

The potential for using power line communication in developing countries depends on whether a critical mass can be reached at the global market. This is necessary in order to achieve sufficient economies of scale in production in order to enable supply of low-cost equipment. See PLC for a technical description.

[1] (PLC Utilities Alliance and PLC Forum) http://europa.eu.int/rapid/pressReleasesAction.do? reference=MEMO/05/119&type=HTML&aged=0&language=EN&guiLanguage=en

MOBILE NETWORKS

Wireless networks include a wide range of technologies both in the access and the core network. Of these wireless cellular networks, supporting mobile communication is the most widespread. Mobile networks have become widely available all over the world. As it follows from the page on mobile services, the number of mobile subscribers has exceeded the number of fixed network subscribers, and in some countries they have become even more available than fixed telecom facilities. In many countries the most important actor is the incumbent operator of fixed network services, but there are other operators which also have their own networks. Often they will however need to rely on the incumbent operator in order to connect their base stations to the core network. The cost factors depend rather weakly on the basic type of radio technology employed[1]. The costs of establishing a new network are considerably lower than for a fixed network, as the costly last mile can be completely bypassed. In particular, 2G networks have proven to be a cost effective viable alternative for provision of telephony services. Compared to a fixed network the investment costs constitute a much lower share of the overall costs. Still a major share of the investments costs are in the access network as the site build out constitutes 30-50% of the investments costs.

Cost structure for US

mobile operators
Source: Tim Giles a.o.: Cost Drivers and Deployment Scenarios for Future Broadband Wireless Networks – Key research problems and

directions for research Proceedings IEEE Vehicular Technology Conference, SpringDate: May 2004. http://www.s3.kth.se/radio/Publication/Pub2004/TimGiles2004_1.pdf.

Operator investment structure
Large country Small country Core network 30% 10% Site build out 30% 50% Radio access network 40% 40% Source: Klas Johansson, Anders Furuskär, Peter Karlsson, Jens Zander: Relation between cost structure and base station characteristics in cellular systems. IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, September 2004 The capacity of 2G networks is too limited to provide an alternative to the fixed networks as an infrastructure for provision of most types of data services. 3G networks are able to provide some data services, but the capacity is still much lower than what can be offered through the use of xDSL, cable or optical fibers. Establishment of 3G networks is also more costly than establishment of 2G networks, in particular in low density areas, as maximum cell size is much smaller than for 2G services. In high density areas, however, the costs per bit may in some cases even be lower than in 2G networks. The major cost driver for 2G and 3G networks is geographical coverage. When density of use increases beyond a certain point, usage becomes the major cost driver. The fixed costs in mobile networks constitute a lower share of the total costs than the fixed costs in fixed networks. The lifetime of the investments is also shorter. Both 2G and 3G networks operate in designated frequency bands, which are licensed. Paying of license fees is an additional cost which must be paid. In countries using auctions for assignment of these frequency bands, this can be a considerable extra cost. For 3G services this type of cost has in some countries reached a level comparable with the costs of the entire network investment.

WIRELESS NETWORKS (OTHER THAN MOBILE)

Other wireless networks include a wide range of technologies both in the access and the core network. In this context we will only discuss the cost implications of use of wireless technologies in the access network, as these are most important for the techno-economics of provision of telecom services. The wireless access technologies include:
s s s

Wi-Fi/Wi-MAX Wireless Local Loop e.g. Fixed Wireless Access Terrestrial broadcasting networks

Wi-Fi is mainly used for provision of Internet access and is available from so-called hot spots installed at public places like hotels, airports etc. Here customers can subscribe to Internet access either for a limited time slot of 2-4 hours, or they can they have a monthly subscription. A monthly subscription may Include access from a large number of access points see IPASS ).

Wi-Fi is also used as a substitute for internal cabling at customers’ premises. This includes provision of access to neighbourhood networks operated by local communities. Such networks provide Internet access to residents in an apartment complex or in a neighbourhood through a combination of Wi-Fi and a centralised data connection using the fixed network or FWA (see below). Provision of Wi-Fi access was from the beginning dominated by a large number of small operators, often without any background in the telecom market, but as the service has matured more telecom operators, including the incumbents, are providing this service as well. A Wi-Fi network can be set up at very limited costs and demands no long term investments. Wi-Fi is often seen as an alternative to 3G, as it offers higher bandwidth at lower costs. At present the limited coverage of up to 300 meters implies that Wi-Fi cannot be used as a substitute for the entire local loop in rural areas. However, the WiMAX standard provides similar functionality as Wi-Fi, but with a much higher range (up to 50 km). In contrast to Wi-Fi, WiMAX uses in some case licensed frequency resources which may add to the costs and make it more difficult for small local operators to use this technology.

Wireless local loop is a fixed wireless service, where the copper-based local loop is replaced by a wireless connection. In this way new entrants can bypass the part of the incumbent’s network facilities which is most costly, and therefore most difficult part of the network, to establish for a new entrant. Wireless local loop has so far mainly been used in two particular cases: 1) In areas beyond the reach of the copper-based network

2)

For high speed connections (FWA)

The first case relates first of all to rural areas in developing countries. Here WLL is a cost effective alternative to extension of the wired network. WLL is for instance used in Ghana by Capital Telecom. One the advantages of WLL is that it can be used to cover an entire area. This is in contrast to a fixed network connection. It may economic feasible to extent network facilities to connect a small village, but this will still leave the surrounding habitations unconnected. In the second case WLL is used to deliver a higher capacity than it is possible to deliver via an ordinary telephone line. In a number of countries a number of licenses for FWA have been issued in order both to improve coverage of high speed services and to increase competition on the high-speed service market. Terrestrial broadcasting networks are designed to carry broadcasting signals only, but the digitization of broadcasting signals allows integration with other digital services. Digital broadcasting networks may therefore in some cases prove to be an alternative to other telecom infrastructures. How this opportunity will affect the market structure remains to be seen.

[1] Source: Tim Giles a.o.: Cost Drivers and Deployment Scenarios for Future Broadband Wireless Networks – Key research problems and
directions for research Proceedings IEEE Vehicular Technology Conference, SpringDate: May 2004. Cost in Wireless networks

SATELLITE

Satellite networks are particularly cost effective for broadcasting and for transmission over long distances in remote areas. A major advantage is that use of satellite enables a complete bypass of regional and even national network facilities. Satellite is therefore an attractive alternative for business customers, for instance banks, demanding reliable data services outside the major cities.

The major cost component is the satellite itself. The major cost driver is therefore traffic volume, while distance does not affect costs. Satellite is therefore cost-effective in low demand areas without a reliable communication infrastructure
See Satellite for a technical description.

KEY POINTS AND RECOMMENDATIONS
s

s

s

Innovations in network technologies have enabled construction of a number of alternatives to the traditional PSTN network. In areas with an existing telecom infrastructure, these networks can be used to offer new services and more bandwidth. In other areas without adequate telecom facilities, alternative network structures can be used to extend coverage and improve accessibility of existing services. Cost characteristics of wireless networks are very different from those of fixed networks. The cost profiles of wireless networks make them in particular suitable for provision of network access in low density areas. The economic viability of different network technologies will, among others, depend on the regulatory environment. Particular types of regulation such as license fees and administration of universal service funds may benefit some types of infrastructures at the cost of others. Regulators must be aware of this in their design of regulation in order to avoid favoring of suboptimal solutions.

2.2.3 Vertical Separation – the Scope for new Business Models
Vertical separation implies that each layer of the entire service can be provided by different sets of actors. These may be telecom companies, companies from other industry sectors or public organizations. Service providers are able to offer their services to their customers without having their own network facilities, and infrastructure providers do not need to engage in service provision to endusers, but can lease facilities to network operators or service providers. This allows for the use of a wide range of new business models in the telecom sector. The vertical structure of telecom markets has undergone considerable changes since the 1980s. Before that time the telecom sector was dominated by national or regional monopolies controlling all levels of the value chain and with close ties to the national equipment industry. In some countries the operators even produced parts of their own equipment. Since then the market has become much more disintegrated, and different levels in the value chain may be controlled by different companies. One reason for this is that the use of digital transmission technologies and of the IP protocol have made it easier to separate the various functions. As it follows from the figure below there, is a certain correspondence between the layers defined in the OSI model and in the TCP/IP protocol, and in the way the market for telecom services has been segmented.

Network layers and Market segmentation

Basic network facilities include ducts as well as fibers or other physical network facilities. These facilities can be provided in many different ways. For instance, fibers can be offered, including all the auxiliary services needed to provide a telecom network, simply as dark fibers or including some basic services. In the same way, wireless network infrastructure may include or may not include various types of passive and active components. This has implied a market structure with different levels of competition within the different market segments, and a supply structure with a wide range of companies representing different degrees of vertical integration.

Categorisation of suppliers according to business models applied
Suppliers grouped according to business models applied 1) Full scale operators – with their own fully developed infrastructure. 2) Operators – with their own not yet fully developed infrastructure 3) Virtual operators – with an infrastructure mainly based on leased lines. 4) Service providers – without own network facilities, reselling services to end users. 5) Infrastructure providers – providing basic infrastructure to other telecom companies, but not to end users

The first category includes at present in reality only the incumbent operators, as they are the only ones with a fully developed fixed access network. This implies that all other operators in varying degrees depend on access to telecom facilities owned by the incumbent. Some mobile operators may possess extensive network facilities of their own, but will often depend on the incumbent when connecting the different sites. The companies in categories 2 and 3 may supplement their own infrastructure by leasing raw infrastructure facilities from others. In this way, they are able to maintain full control of all network functions. They may, however, also outsource part of their transmission through the use of switched interconnection to other networks. Mobile virtual network operators (MVNOs) belong to this category. Another possibility is to act only as a service provider and let other telecom companies be responsible for the entire network operations. In this way some of the most important barriers of entry found on the telecom market are avoided. The last category includes companies from other public utility sectors which have their own network facilities, such as railway and electricity network companies. Such companies may wish to profit from their access to infrastructure without engaging in all segments of the value chain in an entirely new business area. This category also includes municipalities and neighborhood associations which want to provide cost effective access to citizens in a particular area. These organizations invest in their own local loop facilities, but will usually rely on other operators with regard to provision of service provision and access to core network facilities. Internet service providers will often belong to categories 2)-4). In particular, smaller service providers have only made limited investments in their own network facilities. An important aspect of vertical separation is that it becomes possible to provide services without local presence. This implies that services like VoIP may be provided by companies outside the jurisdiction of the national regulator. Skype is an example of a VoIP provider which has based its business model on a global reach of its services provided in a way that does not require investments in local network facilities. Although vertical separation becomes technically possible, there might still be good reasons for some companies to maintain a vertically integrated structure. Incumbent operators want to leverage from their ownership of extensive network structures by means of value creation on other markets. However, regulators may wish to enforce a structural separation of vertically integrated telecom operators in order to ensure a level playing field for new entrants (see practice note on structural separation of British Telecom). Another possibility is that service providers invest in their own networks, consequently saving money in the transmission of their content and services to their customers. Although this is not a general trend, there are examples of companies considering this opportunity e.g. Google.

INTERNATIONALISATION

When the liberalization of the telecom markets took off in the mid-1980s, it was foreseen that the former national markets would be replaced by an international market dominated by transnational operators, and that the sharp distinction between national and international communication would vanish. While the markets for telecom equipment soon became truly international, the markets for telecom services are still national. Many of the incumbent operators have engaged in operations abroad, but these operations in different countries are not really integrated with each other. The telecom operators have become multinational, but are not yet transnational. This is not surprising, as network operations are tied to the physical telecom facilities in each country. It is very costly to invest in such activities, and in particular after the telecom crisis, even the largest operators have been reluctant to spread their activities to more than a limited number of countries. In the markets for mobile communication and for international high speed networks a few operators have achieved a wider international presence, but the retail markets for fixed network services remain nationally oriented. A vertical separation of network provision and service provision enables service companies to achieve a global presence without major investments in local facilities. Skype is an example of such a company. Vertical separation thus implies that national operators will get more competition from abroad. It will be much more difficult to enforce national regulation, and it may be necessary to allow local companies to offer their services on the same conditions as service providers from abroad. Vertical separation will therefore lead to more competition and more innovation at the service market. Service providers such as Skype and EasyMobile are mainly addressing the major first world markets, but also in developing countries for instance in Africa, service providers begin to offer their services. Here the impact may be more visible, as service providers here provide their services in an environment where competition and access to telecom services have been more limited.

KEY POINTS AND RECOMMENDATIONS
s

The concept of end to end services, where a monopoly operator is responsible for all parts in the value chain, is being replaced by a more disintegrated market structure. In addition to the traditional telecom operators, a number of specialized actors have emerged. These include first of all service providers providing their services through the use of leased infrastructures. If the provision of services is separated from the infrastructure, an important barrier to entry on the telecom service market is removed, and a more competitive market for telecom services can be created. An important challenge here is to ensure fair competition between vertical integrated companies, like the incumbent operators and service providers. Possible remedies to be considered include the regulation of terms for interconnection, accounting separation, and in some cases a complete structural separation between service and network operations. Vertical separation has made it possible for new local infrastructure providers to emerge. These may be founded by local citizens or municipalities, or by providers of other types of infrastructures, such as cable television operators or electricity companies. Such actors can make important contributions to improving local connectivity. It is crucial for the viability of such

s

s

initiatives that access to the core network and to services on fair terms is ensured.

2.2.4 Horizontal integration
Horizontal integration involves convergence between two or more of the four different branches of the infocom sector: IT, telecom, broadcast and other media (including film, music, newspapers, and other media not being a part of the IT, telecom and broadcasting sectors). The moving boundaries between telecom and related sectors are best illustrated through the use of the framework depicted below. The table distinguishes among the aforementioned four different branches of the ICT sector. Each of these sub-sectors is divided into three different layers representing different parts of the value-chain: content/services, transport/software and equipment/hardware. The importance of the three layers differs from sector to sector. In the telecom sector the transport layer is the most important, while the content layer is the most important in ‘Other media’. The equipment/hardware layer is most important in the ‘IT sector’, but is also important in the ‘Telecom’ and ‘Broadcasting’ sectors.

Convergence / integration and divergence / disintegration
IT Telecom Broadcasting Broadcast programs Other media Film, music, newspapers, etc.

Content / services

Software based content

Telecom based services and content Network services

Transport / software Equipment / hardware

Software

Transmission

Cinemas, video rentals, etc. Reproduction of films, printing, etc.

IT hardware

Telecom equipment

Broadcast equipment

Source: A. Henten, R. Samarajiva & W.H. Melody: Designing Next Generation Telecom Regulation: ICT Convergence or Multisector Utility? Regulate online.org 2002.
Convergence can take place in one or more of these three levels. Although they are interrelated, convergence at the content/service level does necessitate convergence at the transport level (network convergence), and convergence at the transport level does not necessitate convergence of services. The Internet is the most prominent example of a service combining elements from all of the four sub-sectors. The Internet provides a common platform on which IT services, telecom services, broadcasting and other media services can be provided. Moreover, digitalization of content has made it easier to provide the same content on different platforms. Mobile services are not convergent services as such, but mobile has created a new sub-sector, which in some respects can be seen as a part of the traditional telecom sub-sector, but in other respects represents an entirely new industry. Moreover, some mobile services, such as SMS and mobile broadcasting services, combine elements from ‘Telecom’ with elements from IT and Broadcasting.

Examples of convergent services
s s s s s s

Cable telephony Internet via Cable Internet TV and radio VoIP Triple play 3G Broadcast

Digitalization of voice and other communication services implies that it is becoming possible to separate networks and services, and many different services can be handled on the same networks (network convergence). One of the first examples of this was the provision of telephony sharing infrastructure facilities with cable broadcasting networks. Later on, Internet access and voice telephony via cable modems were also developed. This has led to the introduction of a new business concept, where all fixed residential communication services are provided in the same cable (triple play). Several different kinds of players populate this market: Cable companies upgrade the broadcast networks to include interactive services like voice telephony and Internet, and telecom operators offer broadcast of TV and radio via xDSL broadband connections. In addition to this, optical fiber connections are being offered by electricity companies and other new entrants. This development can be observed in the US, in Europe and in Japan. Later on, it can be expected that triple play services are provided though the use of Wi-Fi or other wireless connections. The concept of triple play services demands heavy investments in network infrastructures, but other types of convergence are less demanding in this respect. Convergence between radio and the Internet enables an extended reach of both services, even though

broadband services are not available. These trends do not imply an immediate unification of the markets for four different sub-sectors. Different services will be transmitted on a number of competing networks using different technology platforms (e.g. wired and wireless). However, as seen in the section on network innovations, each type of network will have its own comparative advantages in providing particular services in a particular environment. Although a unified pure optical network providing all sorts of communication services may be the optimum solution in the long run, this will not materialize in the immediate future, particularly not in low and middle income countries. The competition between network types will be shaped by the availability of existing network structures as well as demographic factors such, as density of customers and the demand for particular services. It is a regulatory challenge to ensure a fair competition without favoring particular technologies. If not properly designed, regulation may skew the competition between different networks. Markets will converge in the sense that different networks will compete in provision of particular services such as Internet provision. In the long term, however, it is possible that at least all fixed services will be transmitted via one unified network.

KEY POINTS AND RECOMMENDATIONS
s

s

The boundaries between IT, telecom, broadcasting and other media are constantly moving. New services combining elements from two or more of these subsectors are being created. Many of these services for instance IPTV and VoIP, are using the Internet as a common platform, but the Internet by itself is also a convergent service combining elements from the IT and telecom sectors. Convergence of services complicates regulatory practices where different services are regulated differently. For instance, it becomes still more difficult to distinguish between different types of Internet-based voice services, such as VoIP and voice mail and telephone services. Video services provided via the Internet and mobile phones are also difficult to distinguish from broadcasting services. The boundaries between different types of networks are also moving. Network infrastructures originally designated for the provision of particular services such as telephony or TV broadcast are increasingly used to provide a broad range of communication services. The most prominent example is the provision of triple play, providing Internet access, telephony and TV broadcast via the same connection. This makes it more difficult to maintain different regulation of different types of network infrastructures. For instance, obligations will with regard to open access or a universal service tax imposed on particular types of networks will create unfair competition between different types of network technologies.

2.3 Price regulation
This section will concentrate on price regulation of end user products, i.e. the retail market, while regulation of intermediate products, i.e. the wholesale market, is included in the section on Interconnection. The section will discuss how existing models for price regulation are challenged by market and technology trends, and discuss future needs for price regulation. This section includes following subsections:
s s s s

The Objective of Price regulation Current approaches to price regulation Technology implications for price setting Key points and Recommendations

RELATED INFORMATION Interconnection Regulating Prices (Module 2)

2.3.1 The objective of price regulation
There are two major objectives of price regulation: s to ensure efficiency of telecom operations s to ensure access, equity and the protection of consumers from excessive prices. Pricing is important for ensuring the efficiency of telecom operations, as high monopoly prices may reduce incentives to reduce costs. However, public intervention in price setting may in the long term also have an impact. For instance, limiting the return on investments or setting prices below the actual costs may reduce incentives for investment in network development and new services. In a market with real competition, there will under normal circumstances not be any reason to intervene in the price setting, as market prices will develop so to reflect the costs of the most efficient operator, and all operators will need to improve efficiency in order to survive. The second objective of price regulation is a bit more complex, as considerations on equity may lead to a need for subsidizing provision of certain services in particular areas, especially if universal access is desired. This will be discussed in more detail in the section on

universal access. Vertical separation of network operations and service provisions enables a more competitive market for end user services. While the cost profile for network operations has been characterized by having a large share of sunk costs, pure service provision is less demanding with regard to investments and the entry barriers are lower. Regulation of prices for end user services may therefore become less important if there is sufficient competition. In some countries, competition in certain areas has already reached a level where tariff regulation of end user services is deemed unnecessary. For instance, call and subscriber charges for mobile services are unregulated in many countries where users have a choice of competitive providers. For example, even if mobile service markets are competitive with respect to subscriber and call charges, there is monopoly power over the termination of calls on specific phone numbers. Thus, a number of regulatory authorities have taken action to reduce high mobile termination charges in many markets. However, regulation of prices may still be an issue in markets where competition is less developed, and there are also areas where regulation may become relevant even if the overall telecom market is competitive. RELATED INFORMATION Universal access Why regulate Prices (Module 2)

2.3.2 Current approaches to price regulation
There are basically four different types of price regulation involving different degrees of regulatory interference:
s s s s

Discretionary Price setting Rate of return Price Cap Cost-based price setting

In addition to these four approaches for setting prices, it should be mentioned that remedies other than direct price regulation may be used to help achieve the same objectives. These include measures such as demand for transparency, non-discrimination and accounting separation. All of these approaches may include considerations on costs of production in their price setting, and may therefore follow to some degree the cost-based pricing principle. However, costs are not necessarily the only consideration. For instance, demand is always important, and affordability may be a factor as well. Prices may also include the financing of a universal service fund, for example. These approaches have, in the telecom area in particular, been applied towards regulation of fixed telephone services, while prices for other services tend to be less regulated, depending upon the degree of competition.

DISCRETIONARY PRICE SETTING

Discretionary price setting is a “heavy hand” type of price regulation, and usually implies that tariffs are determined on the basis of political factors by the telecom authorities or even the minister. Discretionary price setting has in particular been used in countries where telecom operators have been partly or fully publicly owned. In these cases the price setting has in reality been left to the monopoly operator, with subsequent approval by the telecom authorities. Discretionary prices have often been set below costs in order to serve social objectives or political interests. However, such a type of price setting conflicts with the goal of efficiency, as this will hamper financing of further investments in the sector. The result tends to be a very long waiting list for services that cannot be provided. Discretionary pricing by private operators with significant monopoly power will also result in inefficiency and under-investment, as they will have an incentive to charge high monopoly prices and restrict the market they serve. Discretionary pricing by the operators is appropriate if the market is truly competitive and the particular operators do not have significant market power. Thus even in liberalized markets, there is generally price regulation for monopoly services of incumbent operators, but discretionary pricing is permitted for new operators that have no significant market power. The purpose is to regulate monopoly power over price setting, wherever it is found, and to allow discretionary pricing wherever competition is effective.

RATE OF RETURN

Rate of return regulation (ROR) limits the return an operator can earn on its investment in providing services. The principle is that operators should be allowed to earn revenue covering the total operating costs, plus a return on their investment. ROR can be applied either for each service or for the entire operations of the operator. If a rate of return is to be determined for each service, it requires a detailed breakdown of its revenues and costs. ROR both ensures that the operator is allowed to secure sufficient revenue to cover its costs and a reasonable return, and prevents monopoly pricing to gain excessive profits. The weaknesses of ROR are that the operator lacks an incentive to innovate or to minimize costs. If the rate of return is calculated on a cost-plus basis, with its allowed return as a percentage of its investments, the operator can realize an advantage in over-investment and inflation of costs. This incentive can be mitigated somewhat by building time lags into the process of price regulation, but this can only have a limited effect. RELATED INFORMATION

Rate of Return Regulation (Module 2) Rate of Return Regulation Versus Price Caps (Module 2)

PRICE CAP

Price cap is today perhaps the most common type of regulation. Price cap sets a maximum for how prices may develop over the next few years, usually 3 or 4 years. Operators are allowed to increase prices with inflation, but at the same time they are required to decrease prices with a certain percentage every year, as productivity is expected to increase. The allowed maximum price is thus: Pt = Pt-1*(I –X) where Pt-1 is the price for the previous year, I is the rate of inflation and X is the productivity factor. Price caps can be set either for individual services or for a combination of services. The latter is sometimes done to enable some rate rebalancing within service groups. The advantages of price caps are that this method can be simple and the operator maintains a stronger incentive to improve efficiency. However, one should note that there is a certain level of discretion in the setting of the X-factor, which is not subject to scientific calculation, and is generally negotiated between the regulator and the operator. Thus a strong and well informed regulator is needed to make price caps work effectively. If excessive profits are obtained after price caps have been implemented, it is a signal that the X-factor should increase in the following period. RELATED INFORMATION Rate of Return Regulation Versus Price Caps (Module 2) Implementing Price Caps (Module 2)

C O S T -B A S E D P R I C E S E T T I N G

A number of approaches for cost-based price setting have been developed, particularly as it has been necessary for regulators to approve or establish reasonable interconnection prices for competitors to obtain access to network capacity. These are often based on detailed modeling of network costs, and are mainly used for determination of interconnection charges. They are discussed further in comparison of existing regimes for interconnection. The shift in telecom markets from a limited number of uniform and homogeneous services to a market with a wide range of different services, tailored towards serving particular service and user needs, complicates the regulation of prices. In this new dynamic market environment, regulators must often seek new benchmark standards for price regulation, rather than attempting to determine specific prices with limited information.

2.3.3 Technology implications for price setting
Technology implications on price setting can be categorized according to the implications of the above mentioned four overall technology trends, and the implications mediated via the technology impact on the overall market structure. These implications are summarized below. Technology implications on Price regulation Mobile communication s Regulation of new pricing schemes such as prepaid and RPP
s s

Rules for handset subsidies and binding periods Pricing of mobile termination

Development of next generation network infrastructures

s

The same or very similar services are being provided via infrastructures having different cost characteristics. This complicates cost-based pricing. Enables a separation of pricing of infrastructure and pricing of services. Many different products offering similar facilities are become available. These products may be using different pricing schemes and be subject to different types of regulation. More competition reduces the need for price regulation

Increasing use of IP and other packet switching infrastructures Convergence

s

s

Implications caused by changes in market structure

s

Before all the complications that usage of new technologies implies for price regulation are discussed, it should be noted that new technologies in general lead to a downward pressure on costs, and if there is effective competition, a downward pressure on prices. This suggests that although it may be impossible to maintain the current standards for regulation of prices, the current developments should not

lead to an increase in prices for the same services unless those services have been priced below cost and are in need of rebalancing. The development of new types of infrastructures indicates that the same services can be delivered via different types of infrastructures. This improves accessibility and competition and reduces the need for price regulation, particularly if the individual user is able to choose between different infrastructures. Where price regulation is retained, it will however be necessary to take the cost structures of the different types of network infrastructures into account. It is not possible to enforce a price regulation which is both technology neutral and cost-based, as the same services are provided through the use of a range of different technology solutions implying different cost profiles. Here it must be considered whether tariffs should be designed in such a way that only the most cost efficient network operator can survive, or in a way that ensures that facility-based competition can persist. If the latter model is chosen, this may be seen as inappropriate support to an outdated technology; for instance, if competition is between a new wireless network and a (more costly) copper-based fixed network owned by the incumbent. Convergence implies that a host of different services are delivered via the telecom network. Among those services, fixed voice telephony is the one most often subject to price regulation, as it is provided under monopoly conditions. Pricing schemes are constructed in order to reflect both the underlying cost structure and considerations on consumer preferences and market behavior. True cost-based prices may imply a substantial barrier to entry for new customers, as high entry fees and subscription charges may drive away customers with a limited use. Pricing schemes must provide the right incentives towards potential customers, and should not be so complicated that it is impossible to foresee the total user costs. Finally, it must be possible for users to be able to determine pricing comparisons without great effort and cost. In some countries, it has been almost impossible for most users to determine what the real prices of various mobile service offerings are. This has led some regulatory authorities to publish comparisons on their websites so users will be able to make rational choices. It will be noted from the table below that very different pricing schemes are applied for different telecom services. Pricing schemes for selected telecom services Pricing schemes Fixed telephony Subscription charge Distance and service dependent usage charge Caller pays Mobile telephony With or without subscription charge Distance dependent usage charge Caller or receiver pays Internet access Capacity dependent usage charge Flat rate or traffic dependant usage charge Caller and receiver pays. Broadcast Subscription depending on the number of channels (content access) Pay per view possible Receiver pays Convergence will imply that it will be much more difficult to maintain different pricing models for different products. As long as the telecom services depicted in the table are clearly distinguishable and serve different purposes, these different pricing schemes can easily co-exist. The problems arise when new service applications such as VoIP and IPTV become more widespread. Regulation of prices for traditional telecom products like POTS will be complicated by the fact that substitutes for this service are using very different pricing schemes, pricing schemes which so far in most instances have escaped any type of price regulation. This threatens the whole underlying business model for the provision of fixed voice telephony. Operators may therefore need to redesign their pricing schemes through the introduction of flat rate or substantial reductions in usage charges. This may lead to substantial losses in revenues coming from POTS. These losses can only be compensated by higher subscription charges if this will be permitted by current price regulation. It should however be noted that losses in POTS revenues are likely to be compensated by increasing revenues on provision of Internet access. However, some of these revenues may go to other infrastructure providers e.g. cable operators, electricity companies or providers of wireless access. Pricing of mobile services poses a particular regulatory challenge. Being a market in many countries with active price competition for customers, and with less regulation than fixed service markets, the mobile market has been very imaginative with regard to the creation of new pricing schemes. These pricing schemes could in theory have been offered for fixed services as well, but they are especially well suited to mobile communication. The upfront investments needed for adding a particular customer to the mobile network are very limited. Therefore, it is not necessary to demand a minimum payment in the form of a subscription, although mobile operators will try to lock-in customers to long term contracts with high charges for early termination. However, even customers paying for a short phone call or a few SMS messages every month become profitable. In fact, customers who only receive calls can be profitable in many cases. Furthermore, prepaid pricing schemes minimize the financial risk of connecting new customers. The service can therefore be offered to customers without a prior assessment of their financial creditworthiness. In addition, customers can keep better track of their consumption, and do not risk receiving unforeseen bills. Both factors are of particular importance for low income customers, and have helped stimulate market expansion in developing countries. Handset subsidies are another innovation within mobile pricing schemes. Mobile phones are sold at subsidized prices if bundled with a subscription. This innovation is rooted in the fact that the handset terminal constitutes the major part of the investment needed for adding a new customer. Operators and manufacturers have a common interest in paying terminal subsidies. Manufacturers can sell more and more expensive terminals. The operators can get more customers, and more advanced phones tend to generate more traffic (and hence revenue) than ordinary phone terminals. Both of these innovations in pricing raise concern on consumer rights. With regard to prepaid services, customers must be guaranteed that

the service they have bought actually is available, and must be informed if there is any expiry date. Terminal subsidies are usually connected with a binding period with mandatory subscription, without change of operator. This type of restriction may harm the competition, and many regulators have defined a time limit for such restrictions of between six months and two years. Termination of calls in mobile networks poses a particular challenge. Fixed calls terminated in a mobile network are often charged differently than ordinary fixed calls. This premium can either be paid by the receiving party, or more commonly, by the calling party. Both models have their advantages and disadvantages. The receiving party does not initiate the call, and may have no interest in a particular call. Subscribers may therefore turn off their phone most of the time if they have to pay for incoming calls. This impedes the value of the total service market and may restrict penetration. If the calling party pays, it must be ensured that it is transparent for the calling party, if a call terminates in a mobile network, and this may prevent implementation of full number portability (see the section numbering). Cable TV services may also compete with IPTV, but here the implications for price regulation are less far reaching, as usage charges relate to payment of content and not to the transmission. A final technology implication is related to Internet telephony. This service can be provided as a content service, which is independent of the underlying infrastructure. The telecom operator is in this way reduced to a pure infrastructure provider. Provision of the content service Internet telephony does not require local presence and can be provided from any place in the world with Internet access. This market will therefore be more competitive than the market for the provision of infrastructure services, and it becomes difficult to argue for the necessity of price regulation. On the other hand, the competition in providing the infrastructure is still limited, and there may therefore be good reasons for regulating prices for the provision of infrastructure access.

2.3.4 Key points and recommendations
s

s

s

Telecom markets are developing from single service monopoly markets to become multi-service competitive markets. This makes the regulation of pricing both more complicated and less important. However, it may still be necessary to ensure transparency of pricing schemes and maintain other consumer protective measures, even in competitive markets. Regulation of price charges for local loop services may still be necessary, as competition in provisioning of the local loop infrastructure will also be limited in the future. The introduction of mobile services in particular has led to the introduction of new pricing schemes, such as receiving party pays and prepaid services. Variations in pricing schemes can be an advantage to consumers and may stimulate growth in penetration. However, they also make markets less transparent. Regulators can play an active role in providing comparable pricing information on different pricing schemes. Regulation based on cost-based pricing is complicated by the fact that the same services can be offered through the use of different types of infrastructures having different cost profiles.

2.4 Interconnection
This section deals with technology implications for regulation of interconnection.
RELATED MATERIALS

Module 2, section 3, "Regulating for Interconnection"

2.4.1 The objective of interconnection regulation
Interconnection is important both as a consumer issue and for securing fair competition. Interconnection of networks is necessary in order to ensure that all users can communicate with each other or connect to all the services they demand. Interconnection is ranked by many countries as the single most important regulatory issue with regard to development of real competition on the market for telecom services [1]. Interconnection is not a new issue, as it was necessary to interconnect the various national networks each operated by national monopolies for enabling international communication. However, the liberalization of national telecom markets has added a new dimension, as both supplementary and competitive networks now have to be interconnected. Even though new technologies enable the creation of new alternative communication networks, the regulation of interconnection will still be an important tool for the facilitation of competition in both services and facilities. Interconnection is one of the most complicated areas to regulate. First of all, interconnection is technically complicated by itself. There are a wide range of technical possibilities for interconnection, and new types of interconnection products are being developed constantly. Second, even small changes in the terms for interconnections may have huge financial consequences for the operators, and it is therefore important that the terms for interconnection are specified in detail. The terms for interconnection are of particular importance for small operators, which depend on access to incumbent operators’ network facilities. For these operators, costs for interconnection may constitute as much as 30-80% of their total costs. For these reasons most regulators maintain that interconnection agreements are commercial agreements between the operators involved. It is therefore not the task of the regulator to draft the agreement, but only to ensure that agreements made are following the guidelines prepared by the regulator. The regulation of interconnection involves the definition of procedures for establishing interconnection

agreements. These are necessary to avoid unnecessary delays in effecting the agreement. The dominant operator will often have an interest in delaying the implementation in order to postpone more competition with their own end user services. There is also an international dimension, as payments also relate to international communication. Under the present system for international revenue settlements, most developing countries are net recipients of revenue, as the volume of incoming calls greatly exceeds outgoing calls, often because outgoing calls are much more expensive. The US is the major contributor of these international payments and has taken steps to restrict what it will pay in future years. Also, with the new technologies available, most international traffic now bypasses the international revenue settlements payment system, which is unlikely to last many more years, as it was designed on a national monopoly structure using older technologies. In this section we will, however, limit the analysis to include the major issues where technological development will play a major role in the design of future regulation. These are:
s s

Definition of interconnection products/services offerings; Financial terms of interconnection, in particular the structure of interconnection charges.

Endnotes [1] Intven 2000 ch. 3
RELATED MATERIALS

Module 2, section 3, "Regulating for Interconnection"

2.4.2 Interconnection products/service offerings
Convergence and the increasing importance of packet switched services will affect the kind of interconnection products and service offerings which are important for competition. These products include network products provided by telecom copper-based networks, telecom optical networks, telecom wireless networks, broadcasting cable networks, broadcasting terrestrial networks etc. This section provides an overview of the major interconnection products used for ensuring connectivity between networks and between operators and end users. The major interconnection products are listed below. The list is far from exhaustive and includes regulated as well as unregulated products. List of major interconnection products Switched interconnection products Fixed network products
s s

Whole sale of end user products (subscription, traffic) Origination (local, single transit, double transit)Termination (local, single transit, double transit) Transit Exchange of international traffic Whole sale of end user products (subscription, traffic)Origination Termination Roaming (national, international) Raw copper Shared access Bit stream access Fiber Coax Mobile Fiber (dark fiber, lit fiber with or without regeneration, direct access (lambda- connection) Coax Radio links Satellite Sub-marine cables Other infrastructures Local switching

s s

Mobile network products
s s s

Unbundled network components

Network access lines
s s s s s s

Transit lines (national, international)
s

s s s s s

Switching functions
s

s

Tandem switching Subscriber listings Operator services Directory assistance Others Bilateral peering Public peering Transit (hierarchical bilateral) ATM SDH Ethernet

Network management, directory service functions etc.
s s s s

Interconnection of packet switched networks

Interconnection of IP networks
s s s

Interconnection of other packet switched networks
s s s

Co-location and sharing of common facilities

(Constant, variable or available bit rate) Installation of telecom facilities
s s

At local exchanges At concentration point

Sharing of ducts Sharing of mast Interconnection of applications Web browsing Instant messaging services VoIP Other applications It follows that the list of interconnection offerings is strongly affected by technological developments. The list of the traditional interconnection products related to exchange of voice communication is extended by a number of new interconnection products. These products are related to: 1. 2. 3. Separation of infrastructure and services: This creates a demand for interconnection of either services or facilities. Development of new types of infrastructures: This creates a demand for interconnection products related to these infrastructures. Development of new services and applications: This creates a demand for interconnection between them.

Interconnection products include interconnection of both services and facilities. In addition, interconnection of various applications is included in the list. It can be argued that the term 'interconnection' is misleading with regard to applications, as different applications can be communicated effectively if the underlying services are interconnected and if they are using interoperable standards. On the other hand, it is often difficult to distinguish between applications and services, which is clearly indicated by the difficulties related to the classification of VoIP as either a network service or just another information service made possible by the application of IP. The development of instant messages has also raised concern with regard to interconnection. In this context we have chosen to distinguish between interconnection of circuit switched services (primarily fixed and mobile telephone services) and interconnection of packet switched services (primarily Internet interconnection). Finally, a separate category includes colocation and sharing of common facilities such as ducts, masts etc.
RELATED MATERIALS

Module 2, section 3, "Regulating for Interconnection"

CIRCUIT SWITCHED INTERCONNECTION PRODUCTS

Switched interconnection products offer connectivity in particular for voice telephony services. They also include other services, such as dial-up Internet services and SMS, services which are all provided through fixed or mobile circuit switched PSTN networks. Switched interconnection products (interconnection of circuit switching services) include whole sale of traffic minutes. This product is used by fixed or mobile service providers. These service providers take care of all sorts of customer relations such as marketing, billing etc., but leave all network operations to the network operator. Origination is used if a customer subscribes to a different operator than the one handling the call (this is for instance the case if pre-coded carrier selection is used). The handling operator must then pay a fee for using the access line. Termination is used if the subscriber with whom the call is terminated subscribes to another operator. A fee must also be paid in this case. For both origination and termination the charge may depend on at what point in the network the call is handed over to the other operator. Here a distinction between interconnection at three different levels in the network is made:
s

local: interconnection at the same local exchange to which the subscriber is connected

s s

Single transit: Interconnection to the same transit exchange to which the subscriber is connected. Double transit: Interconnection at any other point in the network.

The closer the interconnection point is to the subscriber, the lower the fee may be. Therefore, the handling operator will usually choose to interconnect as close to the subscriber as he can, taking the coverage of his own network facilities into account. Transit is used by operators not having their own network facilities connecting all local exchanges in order to provide full coverage for their services. Exchange of international traffic may include origination, transit and termination of fixed telephone as well as mobile calls, as they may be channeled through the same core network. Roaming is used to connect mobile subscribers in areas not covered by the network of the operator they subscribe to. National roaming is important for new entrants or small operators not providing full coverage.

UNBUNDLED NETWORK COMPONENTS

Unbundled network components may include physical facilities such as transmission and switching facilities, as well as software facilities such as directory databases. Provision of unbundled network components allows new entrants access to telecom facilities owned by the incumbent operators (or in some cases to any other operator with a strong market position). This enables them to enter the market without investing in a fully developed network. They can choose to invest only in some key facilities and complement these with usage of network elements from the incumbent operator. In this way unbundling can be used to create facility-based competition for certain network elements without duplicating investments in all parts of the network. Access to unbundled network elements will thus lower the barriers of entry as the investment requirements are lowered. The most important unbundled network element has been the local loop. The local loop constitutes a large share of the total investments in fixed networks (see section III.B.2). New entrants will therefore be hesitant to invest in this part of the network infrastructure if adequate facilities are in place already. Unbundled local loop facilities have in particular been important for provision of broadband services, as the revenue from voice telephony alone is deemed not to be sufficient to cover the full costs of leasing the access line. In this case, use of switched interconnect will most often be a more attractive solution. Providers of broadband by use of xDSL will usually have three different options: 1. 2. 3. Raw copper (full unbundling) Shared access Bit stream access

The first option implies that the new entrant has the full control with the access line. The new entrant must install their own equipment at the local exchange in order to provide the communication services the subscriber needs. Shared access is used if the subscriber wants to maintain his PSTN subscription with another operator. In this case the new entrant may choose to get access to that part of the frequency spectrum which is not used for PSTN. This solution is to be preferred by ISPs not providing voice telephony services, but Internet access only. Bit stream access is provided through the use of the incumbent operator’s DSLAM. This implies that the new entrant receives a bit stream directly from the incumbent, and does not need to install any equipment at the local exchange. It can be argued that this interconnection product in reality is a packet switched service rather than an unbundled network component. The new entrant buys a service and does not need to invest in his own equipment. Unbundled network components are first of all provided from network facilities owned by the incumbent fixed telecom operators. However, the emergence of other communication infrastructures creates a long list of new opportunities for provision of unbundled network elements. These include access to raw fibers or to raw coax links (wire only), as well as basic infrastructure facilities in wireless networks. Also in these networks, unbundled network elements can be provided in different ways, implying different levels of service included in the infrastructure provision. For instance, fibers can be provided as:
s s s s s

Dark fibers Fibers with or without regeneration of signals under way Fibers with or without amplification of signals under way Fibers with or without end equipment Lambda connections providing a limited spectrum range

While the dark fiber option includes the raw infrastructure, a lambda connection can be compared with bit stream access. In the same way various service options defining different types of unbundling network components can be defined for wireless services, as the interconnection product can include passive components only, or can also include various active components. As noted above, unbundled network elements may include parts of the physical network infrastructure as well as software facilities and databases. Vertical separation and development of new services and applications implies that a need for service related unbundled components may arise in order to ensure interconnection between services. Such components could be directory information on users of instant messaging or other IP-based communication services.

PACKET SWITCHED INTERCONNECTION PRODUCTS

The most important types of interconnection in packet switched networks relate to Internet interconnection. Internet interconnections are interconnections between different public IP networks operated by Internet Service providers (ISPs). Internet interconnection serves a purpose similar to that of switched interconnection, as it enables subscribers to the Internet to connect to subscribers (ordinary users or content providers) served by other ISPs. Internet interconnection agreements are commercial agreements. In contrast to many other types of interconnection agreements, they are usually not subject to regulation and most often their content is confidential. Agreements can therefore take many different forms. In order to understand the difference between these products, it is necessary to look at the different types of actors on the market for Internet services:
s

End users: End users include both business and residential customers. These may use the Internet for very different purposes:
r r r

Use of communication services such as e-mail, VoIP and instant messaging. Information retrieval Provision of information (may be financed through commercials or user charges)

s s

Internet Service Providers (ISPs) providing end user access to the Internet. Internet Backbone Providers (IBPs) providing connectivity between ISPs and to other IBPs.

Some companies act both as an ISP and an IBP. IBPs can be categorized into different layers according to their network coverage. Only the largest ones provide world-wide connectivity, while others provide national or regional connectivity only. In principle there are two different types of agreements, peering and transit:

Interconnection of Internet providers Source: Daniel Roseman: The digital divide Peering Peering may be defined as: 'An interconnection of two public networks that provide connectivity to hosts whose routes are on the global Internet on a settlement-free basis that allows customers of one network to exchange traffic to customers directly on the second ISP’s network.'[1] Peering was the kind of settlement used in the early days of the Internet, when interconnection mainly involved interconnection of research and education networks. The most important aspect of this definition is that no settlements are made between the two networks. Peering agreements will usually be made between two ISPs of similar size with regard to network infrastructure and customer base. In this case both ISPs will have a mutual interest in establishing interconnection of their networks, and neither of them is able to impose a fee on the other. Another characteristic of peering is that both ISPs will accept the receipt of traffic terminating within their own network only. A peering ISP will not allow traffic from another ISP to transit traffic to a third party through his network. Peering arrangements can either be done on a bilateral basis (private peering) or by using a common Network Access Point (NAP), also called an Internet exchange point (an IX), where a number of service providers exchange their traffic on a peering basis (public peering). Such a network access point may be established by national or regional ISPs in co-operation or by the public, who in this way wants to improve connectivity.

Transit Transit agreements are made between an ISP and an Internet Backbone Provider (IBP). Internet-backbone providers are characterized by having extensive network facilities interconnecting with ISP as well as other IBPs. A transit agreement allows ISPs to extend their reach into regions not covered by their own infrastructure or by their peering partners. In contrast to peering agreements, transit agreements involve some kind of payment from the minor to the major provider. On the other hand, the major provider guarantees delivery of packets not only to its own network, but also to all other parts of the global Internet. Such agreements usually include a service level agreement (SLA) specifying capacity and other service parameters. A wide range of settlement schemes can be defined for such agreements, but the amount of settlements will depend on the level of service specified in the SLA, amongst others. In addition to Internet interconnection, there is a need for interconnection of other types of packet switched networks. These networks are often used by Internet providers as bearer services supporting transmission in IP networks, but such connections are also used by other large-scale users of data transmission. ENDNOTES [1] Quoted from Antelope Consulting: Internet cost study App. G, 2001. p. 15.

2.4.3 Comparison of existing regimes for interconnection
Price control is only one out of a number of options for regulation of interconnection services. A number of more light-handed remedies may be used in order to facilitate the development of a competitive market (see Box). The choice of remedies depends on the level of competition on the markets for each interconnection product. Remedies for regulating wholesale markets for telecom services
s s

Transparency Non-discrimination
r

Provide equivalent conditions to 3rd party

s s

Accounting separation Access obligation
r

SMP must meet reasonable requests for access to, and use of, specific network elements and associated facilities

s

Price control and cost accounting

Source: European Regulators Group: ERG Common Position on the approach to appropriate remedies in the new regulatory framework ERG (03) 30rev1 It follows from the previous section that current regimes for interconnection of Internet services and for interconnection of telephony services are very different. Switched interconnection products are usually priced on a per minute basis. Prices are in most countries regulated and cost-based. Costbased prices can be determined in many different ways, and regulation can be either proactive or reactive. Reactive price regulation implies that the regulator assesses prices available on the market and intervenes only if prices are deemed to be above cost-based prices. This model is used for instance in Norway. Proactive price setting implies that the regulator announces price ceilings at regular time intervals. Different cost models may be applied. Both the US and the EU use forward-looking cost models to determine the charges for major switched interconnection products, but historic cost accounting (fully distributed costs) or price minus (end user prices reduced by a certain percentage) are used as well. Forward-looking cost models are considered to be the most relevant method for price setting, but this method is very resource demanding and is based on cost estimates for hypothetical networks, and so is subject to uncertainty. In the United States ISPs are not classified as common carriers under the Telecommunication Act 1996. Internet services are classified as information services and interconnection between ISPs is therefore unregulated. The EU framework is in principle technology neutral and obligations for interconnection apply to telephone operators as well as ISPs if they are infrastructure based (virtual ISPs – VISPs are not included). However, there are important differences in how national regulators intervene in interconnection agreements. Interconnection agreements are in principle commercial agreements. Only if disputes arise, the regulator may intervene. While most regulators have published standard agreements and calculated cost-based prices to be used for settlements regarding switched interconnection and ULL products, regulators have been hesitant to intervene in negotiations of peering and transit agreements. European ISPs are in principle entitled to interconnect at cost-based prices, if negotiating with a provider with a strong market position, but regulatory intervention has in practice been very limited when it comes to interconnection of packet switched networks. The EU regulatory framework defines 18 different sub-markets for end user and interconnection products. On each of these markets, the level of competition is to be assessed and actors with a strong market position to be identified. Thereafter regulatory actions are to be considered if they are deemed necessary. But neither markets for provision of Internet peering nor the markets for Internet transit are among these 18 submarkets. There are important differences between interconnection of telephony services and Internet services, which affect how settlements can be determined, and how they can be regulated:
s s

Single-Service vs. Multi-Service networks Network Transaction Unit – Calls and packets

s s

Reliable and Unreliable Network Services Symmetric and Asymmetric Network Paths

While the telephone system supports one well defined service, the Internet supports many different applications. The basic transaction is an unreliable packet delivery with no direct correspondence to a single service. In a circuit switched network a session can be clearly identified and settlements can be related to the direction of the call. On the Internet, it is not possible to see which party has initiated a session. Neither is it obvious that the direction in which packets are sent can be used to determine the direction of settlements to be paid. For mail services, users send packets, while they mainly receive packets when using information services such as web browsing. The Internet does not offer a firm guarantee that all packets actually are delivered, and it is complicated to implement an accounting system only counting packets successfully delivered. Finally, different network paths may be used for different directions and for different parts of the session. For these reasons, it has been difficult to design a simple and transparent settlement scheme corresponding to the ones used for switched interconnection. Peering arrangements are based on the Sender Keeps All principle, while transit agreements are based on negotiated financial settlement charges. These charges are based on the capacity provided by the IBP and the volume of traffic exchanged. However these are not the only parameters. ISPs with a wide coverage serving a large population or hosting important content providers may obtain better terms than small local ISPs. Interconnection with an IBP does not provide the ISP with access to other parts of the Internet only. It also provides the IBP with access to those parts of the Internet which are operated by the ISP. While peering used to be the dominating form of Internet interconnection, transit arrangements are becoming still more widespread, in particular bilateral transit agreements.
RELATED MATERIALS

Module 2, section 3, "Regulating for Interconnection"

2.4.4 Regulatory implications
Technology implications on interconnection regulation can be categorized according to the implications of the abovementioned four overall technology trends, and the implications mediated via the technology impact on the overall market structure. These implications are summarized below. Technology implications on Interconnection regulation Mobile communication s IC between mobile operators and mobile and fixed operators
s s

National and International roaming IC of SMS and other mobile data services

Development of next generation access network infrastructures

s

The trend from single platform to multi-platform service provision implies a need for new types of interconnection products. These include interconnection between different types of networks and new types of unbundled network components. The trend from voice centric to multi service centric networks challenges interconnection of both voice and data services. Interconnection of IP networks becomes a crucial issue for network development and competition. A need for Interconnection between new services such as VoIP and instant messaging. Dominance by a few international carriers. New imbalances in payments of international settlements.

Increasing use of IP and other packet switching infrastructures

s

Convergence Implications caused by changes in market structure

s

s

RELATED MATERIALS

Module 2, section 3, "Regulating for Interconnection"

MOBILE COMMUNICATION

While the incumbent operator usually enjoys a monopoly on fixed telecom network facilities in the local loop, most countries have two or more operators with their own facilities in other network segments e.g. in the core network or in the mobile network. The various types of interconnection will therefore be subject to various degrees of competition. This has implications for the kind of remedies to be used on the various markets. While price regulation may be necessary in some markets without real competition, demands for access obligation and non-discrimination may be sufficient on other markets. On some markets competition may be so well developed that regulation may become

unnecessary. It is however important not to treat interconnection with mobile networks as one market. The markets for fixed-to-mobile, mobile-to-mobile and mobile-to-fixed are different markets, the levels of competition on these markets will often be very different, and each market must be assessed independently. In particular the market for termination of fixed call in mobile networks has been of concern, as the charges for this type of termination sometimes are three times as high as the charge for termination of mobile calls in fixed networks. Such differences cannot be explained by cost differences but are due to the fact that the markets for these two products are very different. Every mobile operator enjoys a monopoly on termination of calls in his own network, while there is competition in offering the lowest charges on outgoing calls. National and international roaming creates a particular challenge to regulators. National roaming is of particular importance, if some operators lack full national coverage of their own network. National roaming can in this situation be used to extend geographical coverage of their services. This will in particular help new entrants and thereby facilitate more competition. On the other hand it may delay expansion of network facilities. This impact may however be compensated through inclusion of specific demands to network coverage in licensing conditions. Rates for international mobile communication are in general substantially higher than national rates, and the price gap goes far beyond what can be justified by the underlying costs. A Regulation of international roaming demands international cooperation.

DEVELOPMENT OF NEXT GENERATION ACCESS NETWORK INFRASTRUCTURES

Development of new types of infrastructures has created a wide range of new possible interconnection products. It must be asked whether the demand for unbundling posed upon operators of copper-based telecom networks should also include these new network infrastructures. Is it fair that operators of broadcast networks are allowed to refuse other operators access to their facilities, if they want to provide e.g. Internet access by use of cable modem? Why should optical fibers not be subject to the same type of regulation as copper lines? If the philosophy of technology neutrality is applied, the same rules should be applied for all network technologies. On the other hand, these networks are often established on very different market conditions. Long term investments in network facilities may be hampered if heavy regulatory measures are used in order to promote service development in the short term. In particular in the least developed countries it is important to ensure that investors can recover their costs in order to encourage foreign investments. It may therefore be necessary to apply generous access rules for new entrants. Another problem relates to pricing of interconnection products. Technology neutral regulation will imply that the same charges should be used for the same services provided by different types of networks, but this does not always correspond to the principle of cost-based prices. It is therefore possible that operators may be requested to provide interconnect services at prices below the cost of production, if interconnection charges are not carefully designed. Use of forward looking cost accounting models will be even more resource demanding than it is today. Use of new types of infrastructures demands that more cost models must be constructed, and these models must be revised every time it becomes possible to implement new technologies affecting costs. As the cost models are forward looking and based on the most efficient operator principle, new technologies must be taken into account even they if not yet are implemented in the domestic network infrastructure.

INCREASING USE OF IP AND OTHER PACKET SWITCHING INFRASTRUCTURES

Switched interconnection of voice telephony and the related charging mechanisms will become less important. As phone calls only will be one out of a number of different types of communication services in a packet switched network, the core issue will be interconnection between unbundled network elements and exchange of data in an IP network rather than mediation of voice calls between different networks. In a fully implemented next generation access area, circuit switched interconnection is no longer a relevant product. This does not mean that interconnection of telephone services becomes unnecessary, but this interconnection product will affect the service layer only. This implies that interconnection of network facilities and interconnection of services are clearly becoming separated, and the needs for regulation in these two areas may become very different. Regulation of interconnection of network facilities may be necessary due to lack of facility based competition in certain areas. Competition at the service layer is much more likely to occur as services can be provided without local physical presence. However there may still be a need for regulation of interconnection of services such as VoIP. ‘If anyone possesses a termination monopoly, it is the VoIP service provider, not the provider of the broadband pipe’. A need for regulation of services relates to the fact that communication services inhibit substantial network economies, which only can be realized if various service providers are interconnected. Therefore access to essential facilities such as directory databases must be ensured. With regard to Internet interconnection establishment of one or more public National Access Points (NAPs or IXs) can help to provide better access to local content and reduce interconnection costs for local ISPs.

CONVERGENCE

Convergence of services involves a clash between two very different interconnection regimes: Interconnection of telephony services and interconnection of Internet services. While the latter is largely unregulated the former is often subject to complicated regulation based on cost-based prices.

VoIP enables provision of voice telephony in a setting which in some cases escapes the current regulation made for voice telephony. Therefore VoIP operators may have difficulties in demanding the right to interconnect with PSTN networks. It is here important that development of new innovative business models for providing VoIP or similar services are not hampered unintentionally by regulation. Convergence has created a wide range of new communication products. These products may flourish without any public intervention, but there may also be cases, where public intervention may prove to be necessary. For instance by demanding open access to facilities such as name and numbering registers in order to enable interconnection between different platforms for instant messaging. Technology also has an impact on the kind of remedies it will be relevant to consider. For instance use of ex ante price regulation will demand a clear definition of all kinds of service products.

IMPLICATIONS CAUSED BY IN MARKET STRUCTURE

Many developing countries have been net recipients of payments from international settlements from international calls, and for some operators these payments have been a major source of income. This source of income is being undermined both by drastic reductions in tariffs for international calls and by increasing use of VoIP. On top of this, developing countries may face substantial expenditures with regard to settlements paid for Internet interconnection transit services. These expenditures can be substantial. In Colombia, Internet interconnection charges constitute 33% of the total costs paid by users, and in Mexico the international rates alone constitute 23%[1]. Such costs lead to a vicious circle, where high internet costs delay penetration, and low penetration increases Internet interconnection costs. The market for such services is not very competitive, especially in developing countries. It has been estimated that there are in total about 6000 ISPs and 60%20 IBPs on the global market[2]. However out of these only six can be termed as tier-one IBPs. In particular in developing countries the number of IBPs offering connectivity between local ISPs and the global Internet is very limited. In Burkina Faso, Teleglobe was in 2002 the only provider of a single dedicated communication link to the Internet backbone.[3] As noted above, establishment of National access Points may be a way to reduce interconnection costs, but a trend among IBPs to refuse interconnection at NAPs instead of bilateral transit agreement has been observed[4] This is however only one out of a number of examples of anticompetitive behavior observed at the market for Internet Interconnection. Anti-competitive behavior by large Internet backbone providers “Complainants about Internet interconnection services (IIS) – developing countries, Australia, Mexico and Singapore – allege inequitable and anti-competitive behavior by large Internet backbone providers (Tier-1 IBPs) at the expense of smaller IBPs and Internet service providers.” Source: Daniel Roseman: The digital divide and the competitive behavior of Internet backbone providers. It is difficult to document how widespread anti-competitive behavior is, as conditions for Internet interconnection often are kept secret. Antelope suggests therefore in their study that IBP should be demanded to allow regulators to see the agreements. In addition to this, they also suggest a requirement for cost accounting and cost based prices. This implies that they suggest a regulation similar to the type enforced on circuit switched interconnection. According to Antelope the current EU framework allows a requirement for cost-based prices, and enables developing countries to raise a case on price lifting, if this occurs[5]. China has in collaboration with other developing countries in ITU made an even more far reaching proposal in a response to the ITU recommendation on international Internet connection, claiming that it is unfair that other countries bear the full costs for interconnection to the American part of the Internet. It is important to define a framework interconnection of IP networks that benefits all parties. If some operators lose revenue by migration from switched interconnection towards interconnection between IP-networks, and this is at present the situation for many operators in developing countries, it is likely that they will seek to delay the process. This will delay growth in use of NGN in developing countries and hamper the development of the Internet. ITU-T Recommendation D.50 International Internet Connection: General considerations for charging criteria and options for international Internet connectivity I.1 Connection criteria Administrations* may agree to interconnect their networks based on charging criteria including, but not restricted to, the extent of network connectivity and degree of reachability to Internet end users and websites. The agreed level of traffic exchanged may also be taken into consideration, provided that Administrations* may use suitable safeguard agreement to address any concerns that international traffic flows are not fraudulently manipulated. Service performance is another factor that may be considered. Administrations* may agree to consider network performance, availability of contact points, trouble reporting, among other considerations. I.2 Charging options Administrations* may find these charging criteria helpful in establishing the method of charging. Interconnection methods, and therefore charging methods, include peering, transit, hybrid forms of peering or transit, and any arrangement as mutually agreed between them, including indirect interconnection. I.3 International link capacity Where one or more international links are required, arrangements for the international link capacities required and the apportionment of cost for the international link recognize that Administrations* bring value to the connectivity agreement. In determining the apportionment

of cost, multiple methods of apportionment are acceptable, as long as mutually agreed to by Administrations*, including making alternative arrangements. China and others expressed reservation on this appendix. Source: ITU-T Response to ITU-T Recommendation D.50 WORLD TELECOMMUNICATION DEVELOPMENT CONFERENCE (WTDC-02) The World Telecommunication Development Conference 2002 (Istanbul 2002) Asia-Pacific Telecommunity CONTRIBUTION TO THE WORK OF THE CONFERENCE DRAFT NEW RESOLUTION FOR INTERNET ACCESS AND AVAILABILITY FOR DEVELOPING COUNTRIES AND INTERNATIONAL INTERNET CONNECTION CHARGING PRINCIPLES The Asia-Pacific Telecommunity (APT) submits this document to WTDC-02 for its consideration. The proposal contained in this document is commonly endorsed by the following APT members, who are members of ITU: Proposal from Australia, Bangladesh, Bhutan, Brunei Darussalam, China, Fiji, India, Japan, Malaysia, Philippines, Singapore, Sri Lanka, Thailand, Vietnam, TRAI (India). noting a) that ITU Recommendation D.50 on International Internet Connection recommends that administrations* involved in the provision of international Internet connections negotiate and agree to bilateral commercial arrangements enabling direct international Internet connections that take into account the possible need for compensation between them for the value of elements such as traffic flow, number of routes, geographical coverage and cost of international transmission amongst others; b) the rapid growth of the Internet and IP-based international services; c) that international Internet connections remain subject to commercial agreements between the parties concerned; and d) that continuing technical and economic development require ongoing studies in this area, recognizing initiatives by service providers have the potential to deliver cost savings for Internet access, for example by the development of more local content and the optimization of Internet traffic routing patterns in a manner that provides for a greater proportion of traffic to be routed locally, resolves to invite Member States 1 to support the work of ITU-T in monitoring the implementation of ITU Recommendation D.50, bearing in mind the importance of this issue for developing and least developed countries' international Internet connectivity; 2 to create policy conditions for effective competition in the international Internet backbone network access market as well as in the domestic Internet access service market as an important aspect of lowering the cost of Internet access for users and service providers, urge regulators 1 to promote, within the context of national policy, competition among all service providers, including small and medium-sized Internet service providers and incumbent network access service providers, urge service providers 1 to negotiate and agree to bilateral commercial arrangements enabling direct international Internet connections that take into account the possible need for compensation between them for the value of elements such as traffic flow, number of routes, geographical coverage and the cost of international transmission, amongst others, instruct the Director of the BDT 1 to organize, coordinate and facilitate activities that promote information-sharing among regulators on the relationship between International Internet Connection Charging Arrangements and affordability of international Internet infrastructure development in developing and least developed countries. Source: ITU – D ENDNOTES [1] Daniel Roseman: The digital divide and the competitive behaviour of Internet backbone providers part 1. Info 5.5.2003. [2] Antelope Consulting: DFID Cost Study Appendix G: Regulating Internet Interconnection, 2001. [3] Daniel Roseman: The digital divide and the competitive behaviour of Internet backbone providers part 1. Info 5.5.2003. [4] ACCC: Internet Interconnect: Factors affecting commercial arrangements between network operators in Australia. Discussion paper. 2000. [5] Antelope op. cit. p. 30.

2.4.5 Key points and recommendations
s

New network product and service offerings have created a need for a wide range of new interconnection products and services. These differ from a technical as well as a market point of view. Therefore each of these products must be reviewed carefully before regulatory decisions are made. Creation of next generation access networks and IP as a common platform implies that circuit switched interconnection will become less important in the future. Interconnection with mobile networks raises a number of regulatory questions. In general mobile markets are more competitive than fixed telecom markets, and the need for regulation is not as great, but markets for mobile termination have created particular problems. Each mobile operator enjoys a monopoly on termination in its own network, and prices are substantially

s

s

higher than can be justified by the underlying costs.
s

Interconnection of packet switched networks and services will become still more important. The setup for interconnection of packet switched networks is very different from that of circuit interconnection. While many developing countries have been net recipients of international revenue settlements for international calls, they must pay substantial amounts in settlements for Internet interconnection. It is possible to reduce these costs through facilitation of interconnection between local ISPs, for instance through establishment of national Internet exchange points. Services like VoIP can become monopolized by the largest providers by interconnection restrictions applied to smaller providers. Regulation of the terms for interconnection between different providers may therefore be necessary in order to ensure competition in the long term.

s

RELATED MATERIALS

Module 2, section 3, "Regulating for Interconnection"

2.5 (Re) Licensing
This section deals with technology implications for (re) licensing.

2.5.1 Objectives and current regulation
Licensing is a restriction created in order to control market access. In ICT regulation, both individual licences and class licences are used. In contrast to individual licences, a class licence does not require a specific application. All companies complying with the specified conditions are granted a licence. There can be several reasons for demanding licences in a particular industry. The most obvious is that it is necessary to control scarce resources such as radio frequencies, numbering or rights of way. This aspect will be discussed in more detail in the section on spectrum management. Another reason is that licences enable authorities to connect certain specific conditions on how services are provided. These conditions can, for instance, be related to specific requirements with regard to offering universal access or protection of consumers. Such requirements can also be included in telecommunication/ICT laws, although implementation of licensing conditions can be used as a substitute until a legal regulatory framework is in place. In addition licences can be used to regulate market structure. For instance some operators may ask for exclusive rights in a certain period of time, before they engage in major infrastructure projects. A guarantee on the number of licences issued may be used to stimulate investments by existing operators. Finally licensees can be used to generate revenues for the government. This has been a consideration in introducing auctions for allocation of spectrum resources for instance for 3G mobile services. By all means licences represent a barrier to entry and the general trend is that individual licences increasingly are used for allocation of scarce resources.
RELATED MATERIALS

For a detailed examination of licensing of telecommunciation/ICT services and trends in authorizations, please see Module 3, "Authorization of Telecommunication/ICT Services".

2.5.2 Implications of technology trends
The major implications of the four technology trends and implications mediated via changes the market structure is depicted below: Technology implications on licensing Mobile communication Specification of licensing conditions for mobile operators Development of next generation network infrastructures Fair competition between different network infrastructures demand a technology neutral licensing regime. Unified licensing will stimulate optimal use of technology options by operators. Increasing use of IP and other packet switching infrastructures A separation of networks and services enables licenses to be specified to particular layers Convergence s It becomes more difficult to control provision of a particular service through licensing. s Service specific licenses will become a barrier towards development of new services. Convergence leads to a demand for service neutral licensing. s Coordination between regulation of content services and

broadcasting licensing Implications caused by changes in market structure Providers may easily circumvent licensing barriers especially in the content and application markets.

RELATED MATERIALS

For a detailed examination of licensing of telecommunciation/ICT services and trends in authorizations, please see Module 3, "Authorization of Telecommunication/ICT Services".

MOBILE COMMUNICATION

The development of the market for mobile communication depends on how issuing of licenses is managed. Licenses will by all means be necessary as mobile operators need allocation of radio spectrum, which is a scarce resource. It must be considered how licensees for mobile communication should be formulated. Licenses may include demands to geographical coverage, capacity, level of service, pricing etc. Rules for right of way and infrastructure sharing must also be specified. It is here important to carefully assess the market potential before demands in these areas are detailed. Demands for extensive geographical coverage may be an important contribution to ensuring universal access. On the other hand, overly ambitious requirements will be so costly for the operators that economic viability of providing the service is threatened. This will hamper network development as well as universal access in the long run. Methods for assignment of mobile licenses are discussed further in the section III.G on frequency management.

DEVELOPMENT OF NEXT GENERATION NETWORK INFRASTRUCTURES

Development of new types of infrastructures enables more facility based competition, where different network structures compete for provision of the same or similar services. It is necessary to ensure that regulation in different areas is technology neutral, if favoring of a particular infrastructure is to be avoided. In this respect licensing is one of the regulation areas that may impact technology development. If different licensees are needed for provision of different types of network structures regulation can easily unintentionally favor a particular technology. It is, however, not always possible to treat all technologies in the same way. Provision of wireless technologies may need special regulation in order to avoid interference, and the laying of cables may also need some regulation in order to avoid unnecessary digging etc. The task is here to prepare a licensing system avoiding subsidizing or taxation of particular technologies, and on the other hand to ensure that the external costs of possible externalities are included in the licensing fees. Apart from this, it should be noted that requirements for technology specific licensees will impose a barrier of entry. A network operator may refrain from use of an alternative network solution, if this needs an additional license. Technology specific licensees will therefore hamper development of new types of alternative infrastructures.

INCREASING USE OF IP AND OTHER PACKET SWITCHING INFRASTRUCTURES

Separation of the various layers in the network enables use of separate licensees for each layer. Such a model is used for instance in Malaysia (see box III 18). This may at first glance be seen as a further complication for telecom operators, as they will now need to obtain more licenses than in a situation where the same license applies for all layers. On the other hand such a system enables service provision to be liberalized without loosing control with infrastructure development. Therefore it becomes possible to reduce barriers of entry for actors not engaged in all layers, as requirements for licensees in some areas can be reduced or lifted entirely.

CONVERGENCE

Convergence implies that the same services are delivered by different types of networks, and that new convergent services arise. It will therefore be necessary to review service coverage of licenses and consider introduction of unified licenses in order to enable facility based competition and remove regulatory barriers towards development of new converging services. A unified licensing system is being considered in India. The aim here is to enable provisioning of one unified license, which allows provision of all types of telecommunication services. This is done for instance in India. VoIP is an important example of a new service which has challenged the current framework for licensing. The demand for licensing will often depend on the technical solution applied for provision of the service. The current practice varies between countries. For instance are different rules applied for licensing of VoIP. Licensing can help provide regulatory certainty, as in Peru, but the demand for a license can also be a barrier of entry for new operators. Convergence does also affect licensing in broadcasting. Broadcasters are able to provide their services on multiple platforms such as

Web-TV and mobile phones, and telecom operators are able to develop content services, which are difficult to distinguish from broadcasting services. There is therefore a need to co-ordinate licensing for telecom content providers and broadcasters.

IMPLICATIONS CAUSED BY CHANGES IN MARKET STRUCTURE

Telecom markets are becoming still more international and this complicates national regulation in all areas. Licensing is one of the areas most affected as it becomes more difficult to restrict market access. Apart from satellite, all other network infrastructures demand some kind of physical presence at the national market. Service and content provision can be provided from anywhere. Trends in costs structures, and separation of network functions and service functions facilitate development of internal markets for provision of services and applications. On these markets national licensing requirements are easily circumvented. It becomes impossible to restrict market access at the national level and general requirements to providers can only be imposed through development of a common international framework.

2.5.3 Key points and recommendations
s

Fair competition between different network infrastructures requires a technology neutral licensing regime. Unified licensing will stimulate a more efficient use of technology options by operators. Separation of networks and services enables licenses to be restricted to a single layer in the technical network hierarchy. Licensing rules for content services provided on different platforms, e.g. 3G services and broadcasting services must be harmonized or at least coordinated to encourage efficient development.

s s

RELATED MATERIALS

For a detailed examination of licensing of telecommunciation/ICT services and trends in authorizations, please see Module 3, "Authorization of Telecommunication/ICT Services".

2.6 Universal Access
This section deals with technology implications for Universal Access.

2.6.1 Objectives and current regulation
Provision of universal access is one of the most important policy objectives of telecom regulation. The methods for attempting to achieve it sometimes go beyond those included in traditional telecom regulation, e.g., industrial policy remedies such as support from regional development programs, tax incentives etc. In this section we will however, limit the discussion to the technology impact on the primary remedies used in telecom regulation; namely Universal Service Obligations and the establishment of a universal service fund as a method for providing it. The use of internal cross-subsidies by a national telecom monopoly is no longer a viable method in the face of the new technologies and competitive telecom markets.

UNIVERSAL SERVICE OBLIGATION

The Universal Service Obligation is an obligation which can be imposed upon the dominant telecom operator (usually the incumbent). This obligation includes a demand to meet any request for provision of a particular telecom service to anybody within the country. The purpose of having such an obligation is to ensure national coverage of a particular telecom service also in remote rural areas, where provision of telecom service may become less profitable. Most often USO includes provision of PSTN services only, but some countries include more advanced services as well. In Denmark ISDN and leased lines with a capacity up to 2 Mb are included in the USO, and within the EU it has been considered to include broadband services. In many developing countries, it is too ambitious to demand a universal provision of PSTN in all parts of the country and the USO may be replaced with a less demanding obligation of universal access. Universal access can be defined for instance as provision of at least 10 working phones in a certain area, as it is done in Bhutan. A universal service obligation may include some price regulation in order to ensure affordability or a non-discrimination clause demanding that rural areas are provided telecom services under conditions similar to those applied in other areas.

UNIVERSAL SERVICE FUND

The operator on whom the universal service obligation is imposed may demand some kind of compensation to fulfill this obligation. For this purpose, a universal service fund may be set up in order to fund service provision in high cost areas. Such a fund is necessary in particular, if high cost areas are covered partly by small local operators, as is the case in the US. Here local carriers are eligible for financial support, if their costs are more than 15% above the national average. A universal service fund may also be used to fund development of other ICT services to rural health care and educational institutions. Universal service funds can be financed via the public budgets or by a service charge imposed on telecom services. A common solution is to impose a minute charge on certain telephony services (e.g. international and long distance calls), or on interconnection rates. Such charges must however be designed carefully, in order to avoid market distortions such as favoring one technology compared to another.

2.6.2 Implications of technology trends
The major implications of the four technology trends and implications mediated via changes the market structure is depicted below: Technology implications on Universal service Mobile communication
s

Defining the role of mobile services in obtaining universal service. Demands for geographical coverage to mobile operators Universal service can be obtained by use of a combination of many different network technologies. Demand for a US definition, which is independent on network technology. It must be clearly defined how different types of networks should contribute to funding of USO. It becomes more relevant to define US as network access instead of service access. USO should in this case be imposed on network providers rather than service providers. A definition of USO which only includes voice becomes less relevant as voice is delivered in combination with a host of other services. Provision in rural areas becomes cheaper and more economically viable.

s s

Development of next generation network infrastructures

s

s

Increasing use of IP and other packet switching infrastructures

s

Convergence

s

Implications caused by changes in market structure

s

MOBILE COMMUNICATION

There has been an impressive growth in the demand for mobile communication, and the penetration of mobile phones is higher than the penetration of fixed lines. Innovations in mobile business models such as prepaid services have considerably lowered access costs for low income – low usage customers. It is therefore relevant to consider how to include mobile communication in universal service schemes. Should mobile operators be eligible to receive financial support in order to extend their own coverage? Is it relevant to fund expansion of fixed network facilities to areas already covered by mobile services? Should mobile operators contribute to the financing of universal service schemes? All these questions must be assessed in a strategy for achieving universal access. As mentioned in the section on licensing, licensing conditions may be instrumental in ensuring a positive contribution to universal access by mobile services. Mobile licenses may include demands for covering less populated and hence less profitable areas. Too strict demands of this kind will however threaten the economic viability of mobile services and therefore have a negative impact on network expansion. RELATED INFORMATION (Re) Licensing

DEVELOPMENT OF NEXT GENERATION NETWORK INFRASTRUCTURES

The traditional copper-based telecom infrastructure is not well suited to low density regions in developing countries. In particular wireless solutions may be more cost efficient. It is therefore important that universal service obligations are defined in a way not favoring particular network infrastructures. In the most remote areas, satellite may be the most efficient way to provide universal services. In other areas, wireless local loop or use of

mobile network facilities may be the optimal solution. For instance it may be possible in rural areas along major traffic routes to get access to a GSM network, while the fixed network is not yet accessible. In a 5-10 years time horizon, there will be a significant potential for WiMAX in rural and remote areas. In many countries, different network structures may be operated by different companies. It is therefore necessary to allow other operators than the incumbents to participate in the task of providing universal access. It can be done by appointing different universal access providers in different regions, but this solution will hamper use of different network technologies within the same region. Therefore a more flexible solution will be to let a possible subsidy follow the subscribers. In this way, operators receive subsidies according to the number of subscribers they connect in each area.

INCREASING USE OF IP AND OTHER PACKET SWITCHING INFRASTRUCTURES

Today universal access mainly deals with provision of voice telephony although provision of Internet services e.g. from tele-centers may be included as well. By introducing packet switching network structures, it becomes more reasonable to focus regulation on network access rather than access to services. Provision of network infrastructures will maintain those techno-economic characteristics that make real competition in low density areas unlikely, while the markets for service provision will become more competitive.

A definition of universal access, based on access to network infrastructures rather than on access to a particular service, will contribute to an unbundling of networks and services, and it will become more difficult for a universal service provider to achieve a monopoly position in both markets. An argument against this position is that consumers are not interested in pure access, but in access to services. An assessment of affordability must therefore include the full service costs (including all types of devices needed to ensure service provision of a required standard) and not just the costs of access. According to a background paper from ITU it is too early to ascertain that the scope of universal access can be met by an IP connection. The possible separation between service providers and network operators challenges however enforcement of a service based definition of universal access obligation. It is more difficult to impose a universal service charge on service providers. Use of IP enables nomadic use of services. This implies that subscribers can use the same VoIP service from many different locations. This makes it harder to enforce contributions to a universal fund, as the VoIP operator may be located in a different country. Without contributions from VoIP, it may become increasingly difficult to finance a universal service fund (see practice note on VoIP and the US Universal Service Fund).

CONVERGENCE

Convergence challenges the traditional USO in two ways: first, funding of universal services is usually obtained through extra charges imposed on certain telecom services e.g. access charges or interconnection charges. These charges are only demanded from use of telecom networks. USO funding may therefore create a market failure favoring transmission in alternative networks. Second, it becomes questionable to fund one particular service (POTS) when it is produced in conjunction with a host of other services. Regulators must therefore consider a reformulation of the definition of the content of the USO and find new ways of funding which are technology neutral. An important question with regard to convergence, is whether broadband should be included in the definition of universal services. The following considerations are relevant for considering inclusion of a specific service in the universal service obligation:
s s s

Is the service available and subscribe to by a majority of the population? Is the service essential for education, public health or safety or does non use lead to social exclusion? Does widespread use contribute to the benefit of current subscribers (network effect)?

These or similar parameters are applied for review of USO in US and the EU. Even if broadband is deemed to be eligible for being a part of USO, it must be considered whether an obligation will be the best way to stimulate penetration. If not properly designed a universal service obligation may distort the market and delay network investments. A more light handed approach is therefore to be preferred.

IMPLICATIONS CAUSED BY CHANGES IN MARKET STRUCTURE

In particular development within wireless network technologies has reduced the costs of providing universal access. This implies that network provision becomes more viable also in low income low density areas. Regulatory intervention may therefore become less important, as universal access can be achieved on market basis.

Another implication is that developments in usage and demand for telecom services may imply a demand for an upgrade of the definition of universal service. It has been argued above, that from a technological point of view it does not make sense to treat voice telephony as a special service. As the Internet become still more essential for access to all type of community services and to public life in general, it can be argued that Internet access at a certain level of quality of service must be guaranteed to all citizens. If such an upgrade is made, universal service obligation may play a role in future telecom regulation, even when voice telephony has become widely accessible. The concept of universal service obligation has been developed in a non-competitive environment and its application in a liberalized market can be very problematic. It is difficult to subsidize provision of particular services in particular areas without distorting the market. Usually the USO is imposed on a particular operator (most often the incumbent), and this may discourage other operators from investing in competing infrastructures. If subsidies are paid, development of more cost-effective infrastructures becomes less attractive and the outcome may be poorer services and higher costs in the long run. Therefore USO should be used only if reliance on the markets is deemed insufficient to ensure availability also in the long term.

2.6.3 Key points and recommendations
s

s

s

Voice communication can be delivered via many different platforms and network infrastructures including mobile, WLL and IP networks. Some of these will be far more cost effective in certain areas than PSTN over a fixed network by an incumbent operator. Regulation of universal service should not give preference to any particular technical solution or operator, but treat all potential solutions on their merits. Voice communication is only one out of a host of different telecom services. It must be considered to include other services such as Internet access in the concept of universal service. There can be significant economies in providing voice and Internet access to unserved areas together. Use of VoIP alternative network infrastructures may undermine financing of universal service funds. This creates a need for developing of new technology neutral financing schemes.

2.7 Spectrum Management
Following the liberalization of the telecom sector, a more market oriented approach to spectrum management has been introduced. Trade with spectrum resources is being discussed in many countries and restrictions in terms of use have been relaxed. Developments in wireless technologies imply more demand for use of radio frequencies. On the other hand technical solutions towards more efficient utilization of spectrum resources such as software defined radio are being developed. It will become easier from a technical point of view to introduce a more dynamic frequency planning preventing scarcity within specific applications as long as unused frequencies are available in other parts of the spectrum.

2.7.1 Objectives for spectrum management
The focal point of regulation of scarce resources has, due to liberalization of the telecommunications sector, changed from a matter of pure co-ordination and planning to be an important tool in creation of a competitive environment for various telecommunication services. Regulation of the spectrum has to a certain extent been subject to the same development. Spectrum interference and scarcity have traditionally been used as arguments for governmental intervention. With the deregulation of the telecom sector another argument has been that spectrum management could be used to promote competition in the sector. Hence, spectrum management should serve at least three different purposes:
s s s

Granting of exclusive rights Ensuring efficient use of a limited resource Promoting competition

Ensuring effective use of the radio spectrum can be seen both from a technical and an economic viewpoint. While a technical approach to frequency management mainly focuses on maximizing of the supply of radio frequencies by making the use more efficient, economic approaches are dealing with the demand side. Economic measures are used to ensure that the existing supply of radio frequencies is used in a way that maximizes the economic value of spectrum use. From a technical viewpoint the objective will be to optimize the physical use in terms of number of users and the amount of radio signals. From an economic viewpoint, the objective will be to give preference to the most valuable applications. Allocation and assignment of the radio spectrum resources have traditionally been seen mainly as a technical/administrative and not an economic issue. The purpose of regulation has been to avoid interference between users and to optimize the allocation of spectrum resources in order to provide access to everybody, or at least as many users as possible. However, the ongoing developments on the telecommunication market have led to more focus on market oriented procedures for managing spectrum resources. Co-operation between the former telecommunication monopolies has been replaced by competition among a large

number of rival operators demanding fair treatment on equal terms. In addition an increasing number of wireless applications are competing for the same frequencies. These applications often serve quite different purposes, and prioritizing among them demands comparison of different types of economic, social and cultural needs.

2.7.2 Regulatory framework for spectrum management
The framework for spectrum management has been rather detailed regulation, where Government agencies have been responsible for both defining the applications and assigning resources to particular users. The framework is however not based on technical and economic factors only. In addition to the above mentioned factors, the current institutional framework for spectrum management is a product of historical and political factors. The framework for spectrum management is developed over more than a century and the suggestions to a new spectrum policy needs to take this framework into account. The first application of the radio spectrum as a transport medium for communication was wireless telegraphy. This service was an extension of fixed telegraphy and was therefore organized in the same way namely in state monopolies. Only domestic telegraphy was organized in state monopolies, while the market for international telegraphy was more liberal. This historical starting point had tremendous impact on how new wireless services were organized and regulated later on, and how the radio spectrum was regulated nationally and internationally. Another important factor was the close relation between spectrum management and regulation of broadcasting services. Here spectrum management was not only used as a tool for regulation of the transport infrastructure, but also as a tool for content regulation. Issuing of frequency licenses for broadcasting services was in many countries either given to public broadcasting companies or conditioned by fulfillment of certain public service obligations. The current set-up for spectrum management includes co-operation between national and international bodies established to cope with ‘allocation policies’. This process consists of three separate but not always distinct processes:
s s s

Allocation - The division of the spectrum in blocks of frequencies to be used for specific services. Allotment - The distribution of the spectrum rights within allocated bands to uses in various geographical areas. Assignment - The choice among potential users of allocated and allotted channels or frequency bands.

ALLOCATION

The model has been to allocate frequencies between different types of applications. This allocation has mainly been conducted according to technical considerations concerning the suitability of frequencies for certain applications, and political priorities between different types of applications. Different frequencies have different propagation characteristics and different capacities for transport of information. When frequencies are low (long wavelengths), the electromagnetic waves bend around corners, buildings or mountains. When they are high, they travel only in straight lines, requiring line of sight between transmitter and receiver. This implies that it is easier to transmit to a wide area using low frequencies, but that high frequencies are easier to reuse. The amount of information that can be carried increases with increasing frequencies. For these reasons not all frequencies are equally attractive to users. Low frequencies are particularly suitable for communication with long distances and low information loads, while high frequencies are useful for carrying high loads of information where line of sight is obtained through the use of satellites. The allocation of frequencies to various applications takes these differences in technical characteristics into account in order to optimize technical performance. However, market considerations play a role, just as scarcity within a certain application may lead to allocation of additional resources. Allocations are first made at the international level, although the national regulators have some freedom in the detailed planning of frequency resources.

ALLOTMENT

Allotment includes allocation of frequencies between countries and, in large countries, allocation between states or provinces.

ASSIGNMENT

Licenses are given to a specific application (e.g. broadcasting of a radio channel) and it is prohibited to use the frequency for other purposes. As licenses are issued to a well-defined application, specifics of this application in terms of location, power of signals etc. can be taken into consideration in the design of the licenses, and thereby technical performance can be optimized. Assignment within this model is mainly carried out either by issuing of licenses to selected public/semi-public institutions or according to the ‘first come, first served ’principle, and it is not possible to transfer a license to another user. This principle works well if scarcity is limited and the users mainly are public utilities or other public companies, but in a liberalized market where access to radio frequencies is an important strategic resource, it is necessary to establish rules that can ensure equal access among competitors. This has led to the development of a series of alternative assignment procedures, such as beauty contests, auctions

and, in some cases, in the US lotteries. In a beauty contest, companies are selected on the basis of offerings covering a wide spectrum of qualitative and quantitative measures such as quality of service offerings, coverage and number of customers and license holders are mainly public utilities and other providers of public infrastructures. Another alternative is auctions, where licenses are given to the companies with the highest bids. Auctions have been used in the US since the mid-1990s, but were first introduced in New Zealand[1]. In Europe auctions have been used mainly in the recent assignments of UMTS licenses. The argument for auctions is that licenses will go to those who are able to profit most from use of the spectrum by creating valuable services, as they will come with the highest bids. Auctions create income for public budgets and thereby draw money out of the telecom sector. This may be beneficial for public budgets but has been a concern for the telecom sector. The problem became most evident during the UMTS auctions in the UK and Germany, but has been experienced in the US as well[2]. ENDNOTES [1] A. Grünwald, Riding the US wave: Spectrum auctions in the digital age. Telecommunications Policy 25 10/11), Nov./Dec., 719-728. 2001. [2] Eli Noam, Taking the next step beyond spectrum auctions: Open spectrum access. Working Paper, 1995.

2.7.3 Implications of technology trends
The major implications of the four technology trends and implications mediated via changes the market structure are depicted below. Technology implications on frequency management Mobile communication s More pressure on spectrum resources
s

Harmonization of spectrum use in order to facilitate international roaming New wireless networks increase demand for network resources. Need for refarming of spectrum resources. Modulation and compression techniques, smart antennas and widening of spectrum range increase capacity Intelligent radio terminals enable more flexible spectrum allocation Radio technical considerations become less important for allocation of frequencies. Restrictions on use become less relevant when different applications uses the same protocols. Restrictions on use become more complex and hamper development of new applications Many different types of actors providing different types of applications compete for the same limited resources

Development of next generation network infrastructures

s

s s

s

Increasing use of IP and other packet switching infrastructures

s

s

Convergence Implications caused by changes in market structure

s

s

DEVELOPMENT OF NEXT GENERATION OF NETWORK INFRASTRUCTURES

The question of scarcity is important for the choice of regulatory approaches to frequency management. The radio spectrum represents a very valuable resource and the 3G auctions proved that scarcity could lead to a situation where operators were willing to pay substantial amounts to ensure the right resources. Technological development has led to an ever increasing number of applications for the spectrum and it is foreseen that demand for spectrum resources will increase in the future. However there are also trends going in the opposite direction:
s s s

More effective use through better modulation, compression techniques and smart antennas More flexible allocation mechanisms Usage of high frequency radio waves.

In addition to this, it should be noted that scarcity will not be the same in different parts of the spectrum and in different regions. Scarcity will be a much more important issue in the most favorable frequencies for supplying the most profitable products and services in densely populated regions. If these developments will remove scarcity in most frequency bands, this will imply that the traditional ‘first come first served’ principle can

be used. However, this principle may be used in quite a different way than today, as it will become possible to apply a much more dynamic approach to spectrum management, where several users share a common frequency band, but have assigned parts of this for exclusive use on a temporarily basis. Many new applications e.g. Wi-Fi are using unlicensed spectrum bands. It is therefore important to make such bands available in order to facilitate development of new applications, but about half of the countries have already allocated parts of the 2.5 GHz spectrum band to other purposes – e.g. military purposes. Mexico provides a prominent example on how use of Wi-Fi can be facilitated. Here, the regulatory authority COFETEL has allowed use of several new frequency bands for developing WiMAX Internet access for remote regions. Development of new wireless service applications may create a need for freeing new spectrum resources in a particular band in order to be able to comply with international market standards and enable use of mass produced wireless equipment. Regulators may therefore need to refarm current use of spectrum resources.

MOBILE COMMUNICATION

Development of mobile communication networks has increased the demand for spectrum resources. This increases chances for spectrum scarcity, and assignment by use of auctions or public tenders becomes more relevant. Auctions and public tenders can be used if the demand exceeds the number of licenses available. The number of licenses may be restricted for various reasons, but if restrictions are caused by spectrum scarcity, regulators should consider the possibility of removing scarcity by use of frequency administrative measures such as frequency refarming, before a public tender or a spectrum auction is initiated. International roaming is an important mobile service, which demands some harmonization of allocation of frequency bands for mobile services. This must be taken into account when deciding upon which frequencies to allocate for mobile services.

CONVERGENCE

INCREASING USE OF IP AND OTHER PACKET SWITCHED INFRASTRUCTURES

Use of similar standards for transmission of different applications implies that reservation of particular frequency bands for particular applications will become less important for technical optimization of spectrum use. Many of the current restrictions in use of spectrum can therefore be relaxed without affecting technical efficiency. This will simplify spectrum management and it will become more flexible. It is no longer important for the regulator to know how license holders are using the spectrum resources, and it will become easier to introduce new applications as this can be done without issuing separate licenses.

CONVERGENCE

Restrictions on use become more difficult to maintain in an environment, where services are developing rapidly and borders between different applications are blurring. For instance 3G and in particular 4G will support broadcasting services. Convergence will therefore support the aforementioned relaxation of license conditions.

IMPLICATIONS MEDIATED VIA CHANGES IN MARKET STRUCTURE

Development of many different new wireless applications has in combination with the liberalization of telecom markets created an environment, where a wide range of different types of actors are applying for frequency resources. This challenges spectrum management which has developed from being a primarily technical discipline to become an area, where economic and policy considerations play an increasing role for both allocation and assignment of frequency resources. Implications of technology trends must be seen in the context of the ongoing liberalization and market orientation of frequency management. It can be argued that this trend by itself is affected by technology trends, but economic as well as political factors also play a role in this context. Spectrum auctions are one way to introduce more market-based prices for spectrum resources, but auctions do not solve the allocation problem, as licenses are still connected to a specific application. A response to the allocation problem will be to issue licenses without restrictions on the type of applications the spectrum is used for, and to establish a free market where spectrum licenses can be resold. Such an approach was suggested in a study commissioned by Oftel[1]. In such a system no prior allocation of the spectrum between different types of applications is necessary. It is up to the licensees to decide how to use their own spectrum resource.

Within the EU there is a clear trend towards a more market-oriented regime for spectrum management, but there is no general consensus on the use of auctions. Several countries have preferred to use the beauty contest approach for assignment of UMTS licensees, and a free market for spectrum resources as proposed by the Oftel study has not been on the EU agenda at all. The EU is more concerned about harmonization of both allocation and procedures for spectrum management (CEC, 2000). There is, however, a trade-off between harmonization and flexibility of allocation. A free market based on issuing of tradable licenses without restrictions on usage cannot ensure a harmonized use of the spectrum. The free market approach may be the best way to ensure that frequencies are assigned to those users who can create the most value using the spectrum within a certain frequency band. It is however more doubtful, whether the market approach can be used to obtain an appropriate allocation of frequencies between different services. At least in the near future, several specific problems mainly of a technical nature complicate a free market approach towards spectrum management. For instance interference between various users and applications depends on the power of transmissions and the types of applications. For instance, analogue TV broadcasting will not only prevent other broadcasters from using the same frequencies in a considerable geographical area, it will also prevent use of certain other frequencies for broadcasting purposes. Therefore the number of licenses that can be issued depends on what they are going to be used for. This implies that restrictions on usage will enable more licenses to be issued and hence a more efficient use of the spectrum. A free market approach may be suitable for assigning frequency bands between operators within the same application. Regulators can sell licenses on the market and users may be allowed to buy and sell licenses from each other. A relaxation of current restrictions may in certain frequency bands lead to a use that is closer to optimal from a technical as well as an economic point of view. However, as long as restrictions persist on how the various bits of the spectrum are used, it will be necessary for a regulator to decide on allocation of resources between different applications, and regulatory intervention will also be necessary to ensure efficient and harmonized patterns of usage across countries. In the long term, however, convergence will make it more difficult and somewhat meaningless to maintain a strict regulation of what applications a certain spectrum license is used for. Convergence may therefore pave the way towards a free market for spectrum licenses also for allocation, but until then a prioritization among applications must be made.

[1] Valetti, T. M.: Spectrum trading. Telecommunications Policy, 25(10/11), Nov./Dec. 2001, 655–670.

2.7.4 Key points and recommendations
s s

With development of new wireless applications, the radio spectrum is becoming a still more important economic resource. It is not clear whether scarcity of spectrum resources will increase, and if so, for what frequencies and services. Demand for spectrum use is increasing, but so is the efficiency of spectrum use. Better modulation and compression techniques, smart antennas etc. will lead to increased capacity of the spectrum. In addition intelligent radio terminals will enable a more flexible allocation. However, development of new applications will often necessitate refarming of spectrum resources. Convergence implies that allocation of spectrum for specific purposes is becoming both more complicated and less necessary to maintain. Restrictions of use imposed on license holder should therefore be liberalized where they may create artificial scarcity.

s

2.8 Numbering
This section examines objectives of numbering, the regulatory framework for allocation of numbers, implications of technology trends as well as the respective key points and recommendations.

2.8.1 Objectives of numbering
The objective of a numbering plan is to establish a unique identification of subscribers and providers of services. The numbering plan for the telephone network is only one out of several numbering plans each facilitating communication between different users connected through a common infrastructure. Other numbering plans include numbering plans for data networks, such as the IP-numbers used for identification of users of the Internet. E-mail addresses and domain names may also be seen, as a sort of numbering systems as they also are used to provide users with a unique identification code.

2.8.2 Regulatory framework for allocation of numbers
Telephone numbers are designed by the national telecom regulators according to the E.164 numbering plan (see the practice note on the E.164 numbering plan). For technical reasons numbers have been designed according to their geographical location in such a way that subscribers connected to the same local exchange share the same area code. A numbering plan can be either open or closed. In an open numbering plan special arrangements are made for local calls, so the dialer only needs to include the area code in long distance or international calls. In a closed numbering plan the same number of digits must be dialed in all national calls. Most numbering plans reserve special series for different kinds of calls or services. For instance special prefixes are used for mobile numbers in India and in Brazil. Each operator may also have its own prefixes, as in Colombia. Many countries have reserved 800 and 900 numbers for freephone and value added services. Thus numbers do not only serve as a tool for unique identification, they also provide information on location, type of service etc. This information is often necessary in order to know the tariff for the call. The international regulatory framework for administration of IP-addresses is headed by ICANN and completely different from those in the E.164 numbering plan. For more information about the international regulatory framework for the administration of IP-addresses, please see the Practice Note on ICANN. At the national level similar models of self regulation have often been established in order administrate the national allocation of IP-addresses and domain names.

2.8.3 Implications of technology trends
The major implications on numbering of the four technology trends and implications mediated via changes in the market structure are depicted below: Technology implications on numbering Mobile communication s Wireless infrastructures providing opportunities for roaming challenges geographical numbering plans.
s

High growth in the number of mobile phones has increased the demand for numbers. Special number series for mobile services. Number portability between fixed and mobile services Need for transparency in infrastructure use if this affects tariffs. Wireless infrastructures providing opportunities for roaming challenges geographical numbering plans. High growth in the number of mobile phones has increased the demand for numbers. Need for transparency in infrastructure use if this affects tariffs. Nomadic use and separation of networks and services challenges geographic numbering plans. Tracing of geographical origination e.g. emergency calls becomes more complicated. Needs for coherence between different numbering plans facilitating communication across platforms. More competition implies need for more focus on fair allocation of numbers. More competition implies more need for portability of numbers

s s s

Development of next generation network infrastructures

s

s

s

Increasing use of IP and other packet switching infrastructures

s

s

Convergence Implications caused by via changes in market structure

s

s

s

Geographical numbering plans were once a necessity, but today a growing number of countries have stopped demanding a relation between numbers and geographical location. This enables a more efficient use of an increasingly scarce resource, and facilitates number portability for subscribers moving to a new location. Moreover, convergence between local and long distance calls makes it less necessary to inform about geographical location of subscribers. The growing penetration of mobile subscribers has encouraged this development. First, growth in the number of mobile phone has increased the total demand for numbers, which potentially can lead to number scarcity in some regions. Second mobile phones are used from many different locations and numbering and tariff schemes for mobile phones are most often national and not regional. Geographic numbering plans and even national numbering plans are also challenged by VoIP. VoIP services can be provided at any location where the Internet can be accessed. The VoIP operator does not need to be present in a particular country. The operator can

acquire numbers from the UK numbering plan, for example, and offer them to subscribers in any other country. A particular problem related to VoIP is tracing of emergency calls. At present, it is not possible to trace an emergency call from a VoIP phone. The call may therefore be routed to a wrong location (even in a wrong country). It is possible to assign a physical address to a VoIP number, but the same VoIP number may be used from many different locations just as mobile phones are (nomadic use). Another issue related to mobile communication is cross portability of numbers. Number portability is essential for new market entrants, as subscribers are much more reluctant to change operator if they are unable to keep their number. Number portability is important for both new market entrants offering their services by use of existing technologies, and for operators building their service provision on new alternative infrastructures. Number portability between fixed and mobile services could potentially lead to increased competition between fixed and mobile services, as it then becomes more attractive to give up the fixed-line subscription and substitute with a mobile. Cross portability will lead to less transparency in tariffs, as it will become impossible for a consumer to distinguish between calls to fixed and mobile phones. A similar problem arises with respect to national and in particular international settlements between operators. As long as termination charges in fixed and mobile networks differ, it is necessary for operators to distinguish between calls to fixed and mobile subscribers. Furthermore it is doubtful if cross portability is necessary to facilitate competition between fixed and mobile services. There are now more mobile than fixed subscribers in many countries. This indicates that mobile phone services have been able to compete without offering number portability. The situation may be different with regard to VoIP. The penetration of VoIP subscribers is still much lower than that of mobile subscribers, and the service offered by VoIP is more similar to fixed telephony. In this case lower prices are the main advantage. Therefore cross portability between POTS and VoIP will be an important measure in order to facilitate development of VoIP. Convergence between Internet and telephony implies a need for a coordination of numbering plans. VoIP can be offered without allocation of numbers from the E.164 numbering plan. Domain names, IP addresses or service specific identification addresses may be used instead. However, it is not possible to call a VoIP subscriber from an ordinary phone without allocation of a number which can be recognized by the traditional fixed telecom network. ENUM – tElephone Number Mapping addresses this problem by offering a method for conversion of telephone numbers to IP-addresses and visa versa. ENUM is supported by a number of VoIP operators and is being implemented in a number of countries. So far three countries (Austria, Poland and Romania) have reached the stage of production, while trials are implemented in Finland, France, Germany and Japan (Ripe ENUM working group http://enumdata.org/). ENUM is supported by international organizations such as ITU and the EU and is expected to be the future standard for connection between IP networks and PSTN networks. However, current experiences have revealed a number of critical issues related to use of ENUM. All of these are related to the fact that ENUM is developed outside the formal regulatory framework. ENUM may be adopted through agreement among major Internet service providers without any Governmental involvement. On the other hand the E.164 numbering plan is administered by the national regulatory agencies. Therefore the allocation of responsibilities regarding the administration of E.164 numbers and the administration of the E.164subdomain within .arpa must be clearly defined[1]. Endnotes [1] Annette Hillebrand a.o.: Business Models and ENUM – Opportunities and Challenges EuroCPR 2005.

2.8.4 Key points and recommendations
s

It is no longer a technical necessity to use location dependent numbering schemes. Moreover use of mobile phone services and VoIP implies that the same number can be used from different geographical locations. Some countries have therefore introduced national numbering plans allowing subscribers to maintain their number if moving to another region (geographical number portability) Numbers are used as a unique identification of PSTN subscribers as well as subscribers of mobile services or VoIP. If different charges are applied for calling different types of subscribers, consumers may demand clearly distinguishable numbers for PSTN, mobile and VoIP subscribers, but such a numbering plan will prevent cross portability and may hamper competition between services. Translation between IP numbers and the E.164 numbering plan used for telephone services is provided by ENUM. ENUM enables Internet based users to communicate with telephone subscribers and vice versa. ENUM is being implemented in a number of countries, but it also is subject to being replaced in the future as a result of technological improvements.

s

s

2.9 Summary of regulatory implications
Regulatory implications caused by changes in market structure as well as by technology trend are summarized.

2.9.1 Regulatory implications from mobile communication
Regulation area Pricing Implications
s

Regulation of new pricing schemes such as prepaid and

RPP
s s

Rules for handset subsidies and binding periods Pricing of mobile termination

Interconnection

s

IC between mobile operators and mobile and fixed operators National and International roaming IC of SMS and other mobile data services

s s

Licensing Universal Service

s

Specification of licensing conditions for mobile operators Defining the role of mobile services in obtaining universal service. Demands for geographical coverage to mobile operators

s

s

Spectrum Management

s

Harmonization of spectrum use in order to facilitate international roaming More pressure on spectrum resources Wireless infrastructures providing opportunities for roaming challenges geographical numbering plans. High growth in the number of mobile phones has increased the demand for numbers. Special number series for mobile services. Number portability between fixed and mobile services Need for transparency in infrastructure use if this affects tariffs.

s

Numbering

s

s

s s s

2.9.2 Regulatory implications from NGN

Regulation area Pricing

Implications
s

The same or very similar services are being provided via infrastructures having different cost characteristics. This complicates cost-based pricing. The trend from single platform to multi-platform service provision. A need for new types of interconnection products arises. These include interconnection between different types of networks and new types of unbundled network components. Fair competition between different network infrastructures demand a technology neutral licensing regime. Unified licensing will stimulate optimal use of technology options by operators. Universal service can be obtained by use of a combination of many different network technologies. Demand for a US definition, which is independent of network technology. It must be clearly defined how different types of networks should

Interconnection

s

Licensing

s

Universal Service

s

contribute to funding USO. Spectrum Management
s

New wireless networks increase demand for network resources. Need for refarming of spectrum resources. Modulation and compression techniques, smart antennas and widening of spectrum range increase capacity Intelligent radio terminals enable more flexible spectrum allocation Wireless infrastructures providing opportunities for roaming challenges geographical numbering plans. High growth in the number of mobile phones has increased the demand for numbers. Need for transparency in infrastructure use if this affects tariffs.

s s

s

Numbering

s

s

s

2.9.3 Regulatory implications from IP

Regulation area Pricing Interconnection

Implications
s

A separation of networks and services enables licenses to be specified to particular layers It becomes more difficult to control provision of a particular service through licensing. Service specific licenses will become a barrier towards development of new services. Convergence leads to a demand for service neutral licensing. A separation of networks and services enables licenses to be specified to particular layers It becomes more relevant to define US as network access instead of service access. USO should in this case be imposed on network providers rather than service providers. Radio technical considerations become less important for allocation of frequencies. Restrictions on use become less relevant when different applications uses the same protocols. Nomadic use and separation of networks and services challenges geographic numbering plans. Tracing of geographical origination e.g. emergency calls becomes more complicated.

s

s

Licensing Universal Service

s

s

Spectrum Management

s

s

Numbering

s

s

2.9.4 Regulatory implications from convergence
Regulation area Pricing Implications
s

Many different products offering similar facilities become available. These products may be using different pricing schemes and be subject to different types of regulation. A need for Interconnection between new services such as VoIP and instant messaging.

Interconnection

s

Licensing

s

s

s

It becomes more difficult to control provision of a particular service through licensing. Service specific licenses will become a barrier towards development of new services. Coordination between regulation of content services and broadcasting licensing. A definition of USO which only includes voice becomes less relevant as voice is delivered in combination with a host of other services. Restrictions on use become more complex and hamper development of new applications Needs for coherence between different numbering plans facilitating communication across platforms.

Universal Service

s

Spectrum Management Numbering

s

s

2.9.5 Regulatory implications caused by changes in market structure
Regulation area Pricing Interconnection Licensing Universal Service Spectrum Management Numbering Implications
s s

More competition reduces the need for price regulation Dominance by a few international carriers. New imbalances in payments of international settlements. Providers may easily circumvent licensing barriers in particular on the content and application markets. Provision in rural areas becomes cheaper and more economically viable. Many different types of actors providing different types of applications compete for the same limited resources More competition implies need for more focus on fair allocation of numbers. More competition implies more need for portability of numbers

s

s

s

s

s

2.10 Key points and recommendations
Market impact: Service innovations Network innovations Vertical Separation – the Scope for new Business Models Horizontal integration Regulation impact: Price regulation Interconnection (Re) Licensing Universal Access Spectrum management Numbering

3 New Regulatory Paradigm
Based on the technology trends and regulatory implications examined in the two previous sections on technology implications, the present section puts forward and discusses a new regulatory paradigm and its constituent elements. While the section on ‘Market and Regulation’ analyzes the regulatory implications regarding traditional regulatory fields, the present section takes a more transversal view. The overall Executive summary purpose is to present a challenging new regulatory paradigm going beyond known regulatory models and aiming at facilitating the deployment of different technology solutions based on the establishment of an open and level playing field for all commercial companies as well as non-commercial, community-based and end-user-organized network initiatives. A new regulatory paradigm should not only reflect existing best practice regulation but should also build on an understanding of upcoming technology and market developments, which however have immediate implications for present regulatory practices. Introduction Based on the technology trends and regulatory implications examined in the two previous sections of this study, the present section discusses a new regulatory paradigm and its constituent elements. The basic point of departure is that regulatory activities and organizations must reflect the changing technology and market developments. The reason is that regulation is an important factor in technology and market developments. Good regulation can be a vital factor in supporting growth and increasing spread and use of technologies, while failing regulation can be a major barrier. This is why regulation continuously has to adapt to the changing technology and market environments in which it operates. The study focuses on the relationships between technology changes and regulatory changes. These relationships are mediated by the actual and foreseen changes in the markets induced by technology developments, as it is these actual and foreseen market changes that necessitate changes in regulations. In a broader societal perspective, the relationships between technology, market and regulatory changes are co-evolutionary. However, to the individual policy making organization or regulatory agency, the issue presents itself as the regulatory changes made necessary by the introduction of new technologies and the challenge to develop a regulatory framework that enables the best possible use of new technological potentials. The basic aims of regulation, to a large extent, remain the same as they have generally been since the beginning of the telecom reform process starting in the 1980s and 1990s, i.e. to facilitate competition in the markets primarily by means of access and interconnection regulation, to enhance the access to limited resources such as radio frequencies, and to implement the social aims instituted by policy decisions regarding, for example, universal access and service. In spite of the same basic aims, the specific content of these aims, however, changes considerably with technology and market developments. Interconnection in a packet switched environment is different from interconnection with regard to circuit switching; the vastly increasing number of new radio based access technologies makes it necessary to change frequency allocations and assignment procedures; and, universal access policies can take advantage of new access technologies such as mobile communications. One of the basic aims, which has been important in all countries and especially important in developing countries with under-developed telecommunication infrastructures, has been to promote investments in order to extend and expand backbone and access networks. A crucial element in promoting investments, national and international, is to establish a stable regulatory framework so investors know the regulatory terms on which they make their investments. This includes predictability and due process. This, however, may also create problems in an environment of increasingly fast changing technology developments. If, for instance, radio frequencies have been assigned to an operator for 20 years, the operator will expect to have the right to exploit these frequencies for the period of time agreed using the agreed technology solution even though new and more efficient technologies are developed. Similarly, an operator which has been appointed as universal access provider using a specific technology, e.g. PSTN, with some exclusivity rights, will also expect to retain these rights for the period agreed even though access may be obtained more efficiently using mobile or other access technologies. This is a general problem that regulators have to address and it becomes more acute with the increasing speed of technology development. It requires the development of a regulatory framework which is, at the same time, stable and flexible. Another important issue concerns the relationships between policy and regulation. A fundamental part of the regulatory reform process has been to separate operation, regulation and policy. However, the borderlines between policy and regulation may differ from country to country and may also change over time. There is no single way of defining these borderlines although the general aim of limiting political interference in day-to-day regulatory deliberations must be maintained. Furthermore, there is a general development in direction of diminishing direct policy interventions in the markets and instead relying more on the establishment of a general regulatory framework. This is a continuous process. In the monopoly period, policy intervention in network operation was very direct. With liberalization, most public administrations have maintained a direct influence on market players via licenses, but there is increasingly a movement towards regulating the telecom environment by means of framework regulation instead of regulating the players directly. There is, for instance, a movement away from strict and narrow licenses in direction of broader licenses, class licenses and even no licenses. An important question in this connection is the extent to which regulation is necessary. The liberalization of telecom markets has in all countries been accompanied by regulation. However, during the past few years, the aim of diminishing regulatory market intervention has become increasingly explicit. The growth of operators competing with the incumbents and new technologies offering new market entry potentials for market players make it possible to rely more on general rules for market competition and less on sector specific telecom regulation. The necessity and degree of regulation, therefore, has to be evaluated in each specific case. In the present section of the study, regulation is discussed in light of such specific market and technology developments. Furthermore, telecom regulation cannot be seen in isolation from general economic and social developments. Information and communication technologies (ICT) play an increasing role in the economic and general social life and conditions in societies with the result that policies for

their implementation and use become increasingly important. Universal access, for instance, is a social policy as well as a precondition for improving the business development of countries. Regulations of communication, therefore, have to be seen in a broader ICT related policy framework. A comprehensive and coherent approach, consequently, has to be developed. It is in this spirit regarding the necessity of a regulatory framework, which is at once stable, flexible and part of a larger policy framework, that the development of a new regulatory paradigm is put forward in this section. The overall purpose of the section is to present a challenging new regulatory paradigm going beyond known regulatory models and aiming at facilitating the deployment of different technology solutions based on the establishment of an open and level playing field for all commercial companies, as well as noncommercial, community-based and end-user-organized network initiatives. A new regulatory paradigm should not only reflect existing best practice regulation but should also build on an understanding of upcoming technology and market developments, which however have immediate implications for present regulatory practices. Main pillars of the new paradigm The main pillars on which such a new regulatory paradigm rests are the following: s First, to open the communication area for as many different initiatives as possible and to diversify participation in order to establish a basis for the development of network access and communication services in the highest possible quality at the lowest possible price to as many people as possible, with the implication that an open market environment should be supported, as this is the best mechanism to facilitate the growth of markets and technology use. It also means an increased emphasis on universal access or service policies in developing countries, where the other regulatory fields should support the universal access or service aims, for instance in the field of frequency regulation where new and more open frequency management policies should be implemented. s Second, it is important to develop a comprehensive and, at the same time, coherent national ICT policy of which communication regulation constitutes an essential part. Communication regulation encompasses the regulation of communication infrastructures and services, which play an essential role for all technology mediated communications and communication dependent processes in society, residential as well as business. Communication regulation must, consequently, be seen in the light of this broader societal context. Furthermore, the regulatory approach must reflect a greater degree of pro-activeness and facilitation including a broader view on systems failures instead of simple market or policy failures. A coherent analysis of the totality and intersection of market and policy developments must be the basis for the appropriate regulatory measures to be implemented. s Third, an appropriate regulatory paradigm must take technology and market convergence into consideration. Where telecom formerly was a separate industry in relation to IT, broadcasting and other media industries, there is a growing degree of integration between these industries based, to a large extent, on technology developments with similar technology foundations in the whole ICT and media area. This leads to new competitive possibilities, as different infrastructures can be used for conveying the same kinds of services. It also leads to new services being developed and to new questions regarding the interrelationships between infrastructure and content regulation. s Fourth, it is crucial to acknowledge that the national developments in ICTs take place in a wider international context with an increasing importance of global developments in the communication field. The influence on basic technology developments is rather limited in the case of most countries and it can be important to attract international investments. However, local and national initiatives also play a significant role, and national and regional policies are the foundations for developing these international, regional, national and local contributions to network and service expansion. s Fifth, the organizational aspects of regulation must be adapted to the changing technology and market developments in terms of scope and regulatory practices. Even though the general technology trends have many similarities all over the world, it should be acknowledged that there is a wide range of regulatory and organizational responses and that specific technology developments may diverge. Convergence will lead to new technology solutions and the development of new market opportunities, and developing countries will have the possibility to leapfrog some technology solutions implemented in economically more developed countries. They will also have the opportunity to leapfrog organizational forms used in regulation in developed countries. Furthermore, the lack of regulatory resources can make it necessary to implement organizationally less demanding methods of market regulation.

3.1 The development so far and the new paradigm
The developments of liberalization of the telecom area so far are presented. The purpose is to establish the present status from which a new regulatory paradigm will develop. Furthermore, the reasons for public interest and regulation are discussed with an emphasis on the changing environment and, therefore, new reasons for and ways of regulating. The overall trend in telecom policies internationally for the past 20-25 years has been to liberalize the area. This has included introducing competition, privatization - partly or fully - of the former monopoly operators, and a new regulatory set-up, often termed deregulation and now and then re-regulation[1]. The present system is thus based on liberalization, privatization and deregulation and a separation of operation, regulation and policy making. It should be emphasized that there are many variations of this model depending on the national contexts. However, the overall ‘system design’ in the telecom reform process is, to a large extent, the same. ENDNOTES [1] The term deregulation alludes to the decreasing influence of the state on the sector, while re-regulation is used to illustrate that the amount of rules and regulations actually increases when rules have to be made explicit with liberalization and the externalization of regulation in relation to policy-making and operation.

3.1.1 The telecom reform process
The process started in the US, followed by the UK and the rest of Europe and Japan, and spread to the rest of the world, facilitated partly by initiatives of international organizations, e.g. the WTO agreements on the basis of the Uruguay Round[1]. The background for the process was a combination of technology changes, market demands from, in the first instance, business users advocating lower long distance tariffs, and general policy changes in favor of more liberal market conditions. Each of these trends have had a relative autonomy, but in combination led to a fundamental change of the telecom area from monopoly towards competition. The points of departure of economically developed and developing countries, when starting on the liberalization process, have been quite different. Indeed, far from all economically developed countries had a real universal service in telephony but had a much wider diffusion than developing countries. Diffusion levels in developing countries were often less than 1% of the populations. A widely used argument for liberalizing the telecom sector was that, not only was a reform necessary, but it was also made possible by the achievements in the monopoly period in terms of diffusion. The argument was based on the understanding that monopoly had been a good system for extending diffusion. Such an understanding was part of the foundation for the universal service policy implemented in, for instance, Europe, in the sense that post-liberalization universal service policies were seen partly as a defense of what had been achieved hitherto. This type of thinking could, conversely, lead to the conclusion that developing countries with low diffusion levels were not sufficiently mature for embarking on a process of liberalization. However, telecom reform was as necessary in developing countries as in the economically more developed countries, as there was no prospect that a broader diffusion would see the light of day in developing countries based on a monopoly model. The telecom reform process in developing countries was, indeed, inspired by the initiatives taken in the economically developed countries, but in spite of some fears and claims that liberalization would lead to an even worse situation, the overall outcome of the process, so far, has been positive. Diffusion levels have expanded, maybe not so much in PSTN, but mobile has developed dramatically in many countries, and Internet access is increasing. New access technologies and services have been introduced and prices have dropped. ENDNOTES [1] See, e.g., C.A.P. Braga: ‘Liberalizing Telecommunications and the Role of the World Trade Organization. Public Policy for the Private Sector’, Note 120, July 1997, Washington D.C., The World Bank Group. The main elements of the WTO agreements on telecom from the mid and second half of the 1990s are the GATS agreement itself and the special ‘Annex on Telecommunications’, the ‘Agreement on Basic Telecommunication Services’ and, specifically with respect to regulation, the ‘WTO Regulation Reference Paper’. These agreements and the latest WTO development on telecom services can be found on: http://www.wto.org/english/tratop_e/serv_e/telecom_e/telecom_e.htm. Information on the WTO agreements and their implications can also be found in the infoDev ‘Handbook for Telecommunication Regulators’ at: http://www.infodev.org/en/Publication.22.html

3.1.2 The natural monopoly thesis
Telecom was formerly considered as a public utility, which was also the case with water, gas and electricity supply. The traditional definition of the American term public utility is that it is an infrastructural necessity for the general public where the supply conditions are such that the public may not be provided with a reasonable service at reasonable prices because of monopoly in the area[1]. This is the reason for regulatory intervention, namely to secure the interests of the public. At the same time, the public utilities can also be seen as natural monopolies, meaning that from a societal point of view the highest degree of efficiency exists with just one operator in the market area. The reason for a higher efficiency with just one operator is based on large economies of scale and scope, meaning that there are increasing returns to scale and scope. Such a situation is in contrast to the general assumption in neo-classical economics claiming that there, at a point of time, will be decreasing returns to scale and scope and that competition, therefore, is the most efficient mode of organizing production in society. A natural monopoly situation constitutes a market failure and requires regulatory intervention of some kind. With respect to telecommunication, the reasons put forward for characterizing this as a natural monopoly case have been the large necessary investments for setting up national telecommunication systems and the strong network effects associated with point-to-point communications. The implications are, that there are increasing returns to scale created on the production as well as consumption side. In the monopoly period, there have thus been two partly contradicting reasons for regulating public utilities: on the one hand, to protect users from the abuse of public monopoly utilities; on the other hand, to protect the natural monopolies against competition. The rationale behind this second purpose is that there can be cross-subsidies inside the individual utility areas between geographic areas and different services. Utilities with monopoly rights will, therefore, have to be protected against competition at the fringes (cream-skimming). This can be seen as a social contract between public interests and monopoly operators in utility areas, and in retrospect has even been seen as a contract giving monopoly rights for universal service provisions. In most countries, this is a gross exaggeration, as monopoly operators only in few cases have developed a real national universal service. The obligations that monopoly operators mostly have had to live up to concern tariffs and quality of services, while broader diffusion has played a more limited role. With the transition from the former monopoly regime to the present system with competition, the focus has changed regarding monopoly and competition. Though the natural monopoly argument can still, in some instances, be made in different market segments, the singleinfrastructure argument does not take into account the inefficiencies caused by monopoly because of the lack of competitive pressure. The developments since the start of the liberalization of the telecom markets have fully shown the efficiency gains in introducing competition. Furthermore, technology changes have undermined the natural monopoly argument. Where formerly, at the high point of the monopoly period, there would be an implicit understanding among most economists that telecommunication was a case of natural monopoly, technology developments have contributed to changing this, as different networks can carry similar services and as vertical disintegration becomes easier, allowing operators to offer services on other operators’ networks.

ENDNOTES [1] This definition is taken from William H. Melody: ‘Policy Objectives and Models of Regulation’, in William H. Melody (ed.): ‘Telecom Reform: Principles, Policies and Regulatory Practice’, Lyngby, DTU, 1997, p. 14. Melody in this piece elaborates on the basic reasons for and models of regulation. In the ‘Handbook of Telecommunications Economics’, Amsterdam, North-Holland, 2002, edited by Martin Cave, Sumit Majumdar and Ingo Vogelsang, there is a historical overview on the regulatory developments in the USA by Gerald Brock, pp. 44-74. The volume as such includes a number of papers on the basics of regulation and economics in telecom.

3.1.3 Transition phases
In the first phases of liberalization most countries maintained certain protected areas for the former monopolist. In the economically developed countries, it was mainly the PSTN infrastructure and local/national telephony. In a number of poorer countries with less developed telecom infrastructures, the protected area has often been international connections, with the differences in protected areas being related to the differences in the main economic cash-cows in the different countries. In a second phase, more emphasis has been on helping competitors into the markets by means of asymmetrical regulation. The reason is that in markets dominated by former monopolists, it is important to help competitors in order to reach a stage with a sustainable level of competition. At present, the discussions in the economically more developed countries are mainly on when and to what extent such asymmetrical measures can be abandoned. i.e., when is it that market competition has reached a level where sector-specific regulation can be lifted? There have thus been different phases in the development of reasons for regulation, from the monopoly period with its protection of monopoly providers to the first transition phase with an opening of some of the hitherto protected business areas, further towards a phase with a stronger emphasis on helping competitors by way of asymmetrical regulation, and thereafter a situation with decreasing sectorspecific regulations. These phases relate to the competitive situation but have also implications for the universal service/access issue. At first, when monopoly began to be disbanded, the monopoly system was, to some extent, seen as a good environment for the development of universal service/access. Later on, it was witnessed that competition actually helps improving coverage. With respect to the third large area of telecom regulation, frequency management, the development has mostly been influenced by technology advances allowing for an increasing number of operators. However, it has also been important that the economic value the frequency spectrum has come more to the fore, with frequency auctioning and potentials for trading of frequency spectrum between operators.

3.1.4 Consolidation and further development
The achievements, so far, are far from sufficient. The telecom reform process still has to be deepened and consolidated. In a considerable number of countries, different elements of the basic paradigm, including liberalization, privatization and deregulation and a separation of operation, regulation and policy making still have to be implemented and carried through. However, based on the potentials in recent and forthcoming technology developments, the experiences with the present paradigm and its results in terms of market developments and technology spread, it is, nevertheless, timely to consider what the contours of a new regulatory paradigm look like. Such a new regulatory paradigm will build on and extend the present paradigm. It will have to take into consideration the positive experiences with the present paradigm as well as the problems encountered. The positive experiences are, by far, dominating in terms of increasing access, new services and decreasing prices. However, problems and challenges have also been met, for instance lack of real competition and a continued privileged position of incumbent operators, slow processes of innovation, and insufficient investments. A renewed paradigm has to establish a synthesis on the basis of the positive as well as negative experiences with the regulatory reforms hitherto performed.

3.1.5 Key points and recommendations
s

The telecom reform process in developing countries was inspired by the initiatives taken in the economically developed countries, but in spite of some fears and claims that liberalization would lead to an even worse situation, the overall outcome of the process, so far, has been positive. The achievements, however, are far from satisfactory. Based on the potentials in recent and forthcoming technology developments and the experiences with the present paradigm and its results in terms of market developments and technology spread, it is timely to consider what the contours of a new regulatory paradigm look like.

s

3.2 Technology and market trends
This section examines regulation of convergence and divergence, the change in the relations on national and international level and technology leapfrogging.

3.2.1 Regulation of convergence and divergence
Convergence is the coming together of hitherto separate elements/entities/spheres. As explained in the sections on ‘Technological Trends’ and ‘Market and Regulation’, the ICT area includes different communication and media area (telecom, IT, broadcasting, and other media) and ICT convergence deals with the coming together of these areas. As shown in the section on ‘Market and Regulation’, there is a trend towards convergence horizontally. At the same time, there is a trend towards divergence vertically. The technological possibilities for separating between the different layers in the IT and communication systems constitute the foundation for convergences horizontally. However, it should be noted that these trends are only the dominating ones. Market convergences also take place vertically, and market divergences horizontally.

CONVERGENCE

Issues on convergence in the ICT area can be divided into technology convergence, market convergence, convergence of regulatory provisions, and convergence of regulatory organizations. In the section on ‘Technological Trends’ of this study, the technology aspects regarding network and service convergence and, hence, end-user equipment are analyzed. In the section on ‘Market and Regulation’, there is focus on the market aspects of ICT convergence and divergence. In the present sub-section, the general features of the regulatory provisions regarding convergence are examined, and in the sub-section on ‘Organizational Aspects’, potential organizational implications are analyzed. Focus in the discussions on the implications of technology and market convergence on the regulatory provisions (laws) has partly differed in different parts of the world. In the US, emphasis has mainly been on the possibilities for enhancing competition, when different networks can deliver essentially similar services, and on the problems for competition created by the horizontal integration of operators, e.g. when the same operators own different kinds of infrastructures (see practice note on access competition in the USA). In Europe, focus has been more on the broader societal advantages that convergence potentially entails in the form of new services and new industries[1]. Conversely, the downside has been seen as the problems that this may create for content regulation related primarily to public service provisions in the broadcast area. In many developing countries, the main interest has been the new access possibilities and the discussions on unified licensing[2]. These differences in the main focus do not, however, mean that the other issues have been out of scope in the different countries. The issue of new access possibilities will mostly be strongly correlated with increased competition, but it means that the emphasis has differed. Technology convergence provides the possibility for new competitors to enter the markets. Telephony can, in addition to traditional PSTN operators, be offered by cable TV operators, for example, but these positive implications may be undermined by operators owning different infrastructures horizontally. Horizontal integration can also be seen in the content area, where content products, for instance computer games, can run on many different platforms. However, this does not constitute a problem equal to the infrastructure area, as content is not a bottleneck or an essential facility (see practice note on the essential facility doctrine). There may, however, be arguments relating to media pluralism and the problems in cross-media ownership affecting horizontal integration in the content areas. The ‘rule of thumb’ regarding horizontal and vertical integration of market players is that horizontal technology convergence potentially is beneficial to competition. However, the competition enhancing implications of technological horizontal convergence are diminished when individual companies integrate horizontally, increasing their market power across different technology platforms. With respect to vertical integration, this will, in most cases, not in itself hurt competition. However, when vertical integration is combined with horizontal market dominance in one of the market layers, vertical integration may impede competition[3]. An example of the combination of vertical integration and market dominance horizontally is the provision of DSL services in markets with one operator dominating the provision of local copper loops. In such a market, the owners of the local loops have a clear advantage in the provision of DSL services. This has been witnessed in the actual developments of market shares in the DSL area. The problem of vertical integration is much smaller if combining market dominance in the local loop market with content provisions, i.e. ‘higher’ in the value chain than conveyance services. Even though an access infrastructure provider can have an interest in offering content services to supplement its infrastructure provision, the interest in having traffic on the network caused by the provision of content services by other companies may supercede the interest in dominating the content market. An additional type of question relating to ICT convergence has centered on how to promote new convergent services and converging industries. Whereas the different communication and media industries formerly were more confined to their specific areas, technology convergence creates possibilities for companies to develop and deliver services across technology platforms, and for users to get access to new kinds of communication and media services. In connection with this, issues regarding cross media ownership problems have also played a role. For many decades, some countries have had legislation limiting cross media ownership in order to support pluralism in the media. Convergence tendencies in the media are thus not entirely new. The new issue is that with technology convergence, there are not only market reasons for integrating companies from different business areas; there is also a stronger technology basis. In developing countries, the highest interest has centered on the new possibilities for people to get access to communication facilities by means of different technological solutions. The implications are that the use of different technologies should be promoted and, that universal access policies should not focus specifically on PSTN access but more on the wider range of different technology access possibilities. Part of this may be to change the licensing regime in direction of unified licensing instead of technology specific licenses, if licensing is considered to be necessary at all. RELATED INFORMATION Technical aspects of convergence Horizontal integration

ENDNOTES [1] European Commission: ‘Green Paper [COM(1997)623] on the Convergence of the Telecommunications, Media and Information Technology Sectors, and the Implications for Regulation: Towards an Information Society Approach’, December 1997. [2] On access, see ITU: ‘Trends in Telecommunication Reform 2003: Promoting Universal Access to ICTs: Practical Tools for Regulators’, ITU, Geneva, 2003. On licensing, see ITU: ‘Trends in Telecommunication Reform 2004/05: Licensing in an Era of Convergence’, ITU, Geneva, 2004. [3] These issues are discussed in a report from the Nordic competition authorities, ’Telecompetition: Towards a Single Nordic Market for Telecommunications Services?’, no. 1, 2004, Nordic Council of Ministers, pp. 90-104. The ‘rules of thumb’ are taken from this report. A theory overview can be found in, for instance, Patrick Rey and Jean Tirole: ‘A Primer on Foreclosure’, IDEI Working Papers no. 203, 2003, Institut d'Économie Industrielle (IDEI), Toulouse.

DIVERGENCE

Formerly, infrastructure and service provision of telecom were integrated in individual companies. It was the same company that rolled out the infrastructure and provided the relatively few services provided on this infrastructure, primarily telephony. However, with digitalization the technological potentials for separating the different layers in the provision of communication services have increased, and these potentials have further increased with packet switching technology. Internet is the prime example. However, this does not mean that, formerly, there were only technology reasons and no market reasons for integrating vertically. However, vertical integration is, presently, driven by market imperatives rather than technology imperatives. An example is triple-play coming as a service package on top of a specific access technology. It is often claimed that the money is in applications, services and content and not in infrastructure provision, see for instance the claim that ‘content is king’. This claim is problematic, as infrastructure provision combined with interpersonal communication services in many instances have shown to be a profitable business[1], but it is true that customers demand interpersonal services and content and not just infrastructure. Companies will, therefore, often seek to combine the provision of infrastructure with applications, services and content, as in the case of triple-play combining the provision of broadband access with Internet access, VoIP and broadcast. The problem in such an arrangement is that the companies offering these service packages get a monopoly on the individual customers. There may be other companies in the market, but once a customer or group of customers have signed up with a provider of such a package, they are tied or locked-in to this specific company reducing competition in the market. This could call for regulatory intervention, but there can also be a problem in hindering such vertical integration, as an important incentive for a company to roll out the access infrastructure is the demand for the services in the package. This leads to one of the important discussions on the separation of infrastructure and services. With the technologically improved possibilities for separating services from infrastructure, the idea has often been advanced that a market-based separation would result in a more sustainable competitive situation. In the present situation, operators are most often allowed to operate in the infrastructure markets as well as the services markets. Furthermore, with the traditional dominance of incumbent operators in the infrastructure markets, the situation is often difficult for the competitors. With a split between infrastructure and services, there would be a more level playing field in the service markets. Furthermore, such a split between infrastructure and service provision is known from other utility areas, for instance rail transportation and electricity. The proposal to separate infrastructure and services has been put forward in developed as well as developing countries, and there is no discussion that it would contribute to solving the issue of the leverage of market dominance in the infrastructure area to the service area. The less clear questions, however, are who the infrastructure provider(s) will be and what the incentive structure for extending the infrastructure will be. Infrastructure providers could be different competing companies with licenses to operate at the infrastructure levels. It could be a cooperative between the infrastructure parts of telecom companies. It could, furthermore, be a private infrastructure monopoly or a public monopoly. The basic question in all cases is what the incentive structure for such companies would be. There is a risk of establishing a less dynamic market structure than in an environment with close ties between infrastructure and service provision. This is an important issue to discuss if separating infrastructure and services. RELATED INFORMATION Vertical separation ENDNOTES [1] See discussion in Andrew Odlyzko: ‘Content is not King’, First Monday, vol. 6, no. 2, February 5th 2001.

KEY POINTS AND RECOMMENDATIONS
s

Focus in the discussions on the implications of technology and market convergence on the regulatory provisions has been partly different in different parts of the world. In economically developed nations, primarily the US, emphasis has been on the competition enhancing aspects of technology convergence. Furthermore, the growth of new services and new business areas is considered to be important. In developing countries, emphasis is mostly on new access possibilities with new technologies. In relation to technology divergence vertically, the proposal to separate infrastructure and services has been put forward in developed as well as developing countries, and there is no discussion that it would contribute to solving the issue of the leverage of market dominance in the infrastructure area to the service area. The less clear questions, however, are who the infrastructure provider(s) will be and what the incentive structure for extending the infrastructure will be.

s

s

The recommendation is for developing countries to take advantage of the new access and competition potentials created by new technologies and to limit horizontal integration of companies to the extent that it hinders competition. Furthermore, vertical integration should in general not be constrained when not presenting a serious problem to competition.

3.2.2 Changed relations between national and international level
This section presents internationalization of telecom markets, international tariffs, public sector influence and respective key points and recommendations.

INTERNATIONAL TARIFFS

The development towards packet switching is a process that will take a number of years. For some time still to come, circuit switched services will be important in the fixed network area as well as the mobile area. For international communication, the consequence is that service provision will still be running on the basis of well-known structures - including debatable, to say the least, tariff schemes. In fixed line communication, international tariffs have for many years been priced high above costs, based on international accounting rates, and this system continues in the mobile area with high international roaming prices. When dealing with international communication, this is an issue where a discussion on regulatory intervention is obviously required[1]. International tariffs for circuit switched services have been on the agenda in international organizations, e.g. the International Telecommunication Union (ITU) and the World Trade Organization (WTO), for a long time. It has been difficult to make any progress, primarily because operators have an interest in high international tariffs, as they make money on this traffic and have a cartel-like mutual benefit in keeping international tariffs high. In connection with developing countries, it should also be mentioned that with regard to countries with a larger ‘export’ than ‘import’ of communication services in the shape of either incoming traffic or roaming of mobile users from other countries, there has also been a certain reluctance towards abandoning the international accounting rate system and now also the roaming system. The reason is that the income from these areas, in some countries, still constitutes an important share of total income of operators. However, with the liberalization and the processes of privatization, the interests of operators and public authorities will increasingly diverge, and public authorities will be more inclined to see to the interests of the users of communication services, which include reasonable rates. Moreover, the massive bypass occurring across the world also puts great pressure on the international settlement regime and will ultimately make it meaningless. ENDNOTES [1] In the mobile area, the International Telecommunication Users Group (INTUG) has been very active in working for lower international mobile tariffs.

INTERNATIONALIZATION OF TELECOM MARKETS

During the past 25-30 years, producers of telecom equipment have become increasingly international. In the network and service monopoly period, it was not unusual with close links between national monopoly network and service operators and national equipment manufacturers. This has changed considerably with the liberalization of the telecom area. Today, the large equipment manufacturers (network equipment and handsets) must have an international outlook. The largest producers sell their products all over the world and have production sites in a large number of countries. This applies, for instance, to Nokia, Ericsson, Motorola, Huawei, ZTE, LG and Nortel. There is also a great amount of smaller and medium-sized equipment manufacturers, focusing regionally or internationally and producing intermediate and final products. However, the structure of the telecom equipment industry has changed significantly during the past couple of decades. This was, in fact, the first and very visible implication of the liberalization in the telecom area. The telecom area has also witnessed an increasing internationalization of communication backbone networks. Formerly, the international parts of networks were mostly owned and operated by consortia of national monopoly operators as in the case of satellite connections and sea-cables. This was an expression of the dominating national structure of the telecom area, where international communication had the form of correspondent relationships between national operators[1]. However, during the 1990s, wholly new and truly global backbone operators entered the markets operating not only the international parts of networks but also different national parts. This market segment witnessed a sharp setback with the telecom crisis in the beginning of the new millennium[2], but backbone networks are, presently, very internationalized. This applies not only to operators with their origin in the economically developed countries but also in developing countries. The Indian company Reliance, for instance, in 2003 took over FLAG, and the Indian international incumbent VSNL in mid 2005 acquired Tyco as well as Teleglobe. The same kind of internationalization does not apply to the access parts of infrastructures (apart from satellite access). Most access infrastructures are locally bound, whether wired or wireless. They can be owned by foreign/international operators but are physically bound to the local areas. This is part of the background for the relative strength of incumbent operators in their national markets. They own the physical access infrastructures and, therefore, access to the end-users. With respect to the services provided via these physical infrastructures, they can very well be provided internationally. In the case of traditional circuit switched services, exchanges may be placed in neighboring countries, but this only happens in very few instances. However, with packet switched Internet-based communication, services and content can easily be provided internationally. This applies not only to content services but also to interpersonal communication services as in the case of Skype telephony, which just requires that a

program be downloaded on the computers of the users. Such a ‘disconnect’ between the physical access infrastructures and the services running on top of them is a consequence of the layered structure of communication networks described in the former sub-section on divergence of infrastructures and services. Regulation-wise, the result is that where different kinds of services delivered on dedicated networks traditionally have been regulated nationally, it will be increasingly difficult nationally to regulate services which are provided internationally, but it will remain possible to regulate the access infrastructures. The overall results of the processes of internationalization, so far, in the telecom area are thus that equipment production has increasingly become international; international backbone operations have gone from a regime totally dominated by correspondent relationships between national operators towards a system based on international backbone operators and international end-to-end communication; access infrastructures are predominantly national (except for satellite access) but access infrastructures are not unusually owned by foreign/international operators; finally, services and content are increasingly delivered internationally and are less locally bound. These developments already have and will, to a growing extent, affect national regulations and, therefore, the relationships between national and international regulations. RELATED INFORMATION Internationalization and vertical separation ENDNOTES [1] A good and short analysis of this development can be found in William Drake: ‘The Transformation of International Telecommunications Standardization: European and Global Dimensions’, in Charles Steinfield, Johannes Bauer and Laurence Caby (eds.): ‘Telecommunications in Transition: Policies, Services, and Technologies in the European Economic Community’, Newbury Park, Sage, 1994, pp. 71-96. [2] The ups and downs of this industry are analysed, e.g. in Martin Fransman: ‘Telecoms in the Internet Age: From Boom to Bust to …?’, Oxford University Press, Oxford, 2002.

KEY POINTS AND RECOMMENDATIONS
s

Equipment production has increasingly become international, as have international backbone operations; access infrastructures, however, are mostly locally bound, while services and content can increasingly be delivered internationally. These developments have affected, and will affect more and more, national regulations and the interrelationships between national and international regulations. Tariffs for international circuit switched services are still well above costs, in the fixed and especially in the mobile area. With the liberalization of communication and the establishment of independent regulators, the possibilities for lowering these tariffs have increased. New technologies and the bypass of international settlement systems also lead in that direction. The processes of liberalization and internationalization of communication has led to decreasing influence of public institutions on international organizations and forums. To the extent this is considered a problem, working for the general improvement of the influence of public institutions in the different kinds of international organizations is important, and one of the ways forward is southsouth cooperation between regulators.

s

s

PUBLIC SECTOR INFLUENCE

With the wide-ranging changes in the telecom area internationally, the international organizational structure has also changed considerably. In the former system of national monopoly operators, the international relationships were of a correspondent character with ITU as the international organization in which not only recommendations for standards were elaborated but also agreements were reached on the basis of which international tariffs were bi-nationally agreed (the accounting rate system). Currently, the organizational ‘picture’ is much more complex. In the standards area, a host of international industry consortia have been created[1]; Internet has its own set of organizations centered on the Internet Society and the Internet Engineering task Force (IETF); and, WTO has also played a role, especially in the 1990s in relation to the Uruguay Round of negotiations. In this ‘picture’, the influence of developing countries is relatively small. Under the former system, it was also the stronger nations and players that dominated international relations, but the present multi-faceted and complex system is even more difficult for developing countries to influence[2]. A case in point is ICANN, the Internet Corporation for Assigned Names and Numbers. ICANN is a California-based non-profit corporation, contracted by the US Government to assume responsibility for space allocation of IP addresses, protocol parameter assignment, domain name system management and root server system management[3]. User representation in ICANN is continuously being discussed and was actually reduced in 2002[4]. The discussion on user and public representation is not especially related to developing countries. There are, however, different developing countries that have protested against the US centric character of the management of the Internet, among them China and Iran. Furthermore, the European Union has also been critical of the manner in which the Internet is managed presently. The main criticism is that ICANN should be brought under the rule of some kind of international law, as ICANN is an international regulatory organization in the Internet area[5]. The counter argument is that this could risk making procedures more bureaucratic and that there is a danger that countries wanting to censure content on the Internet would gain influence. The present result of these discussions is a decision from November 2005 to establish an Intergovernmental Forum to discuss all Internet issues but to retain the arrangement with ICANN as an organization based on a contract with the US Government and under the Californian law. The ICANN construction and the root server system that it manages is a clear example of the difficulties for public organizations to gain influence on Internet developments. From a developing country point of view, the Internet is very US centric with 10 of all 13 root file servers on a global scale located in the US and Internet traffic centered very much on the US. There are good historical reasons for this, but it nevertheless illustrates the difficulties for developing countries and their regulatory organizations to gain influence on communication

developments, which are vital also for them. One of the implications of this is that, while interconnection principles for international PSTN traffic dictate each side involved in a transaction to pay for a half circuit, the arrangement in Internet traffic is that the country or region wanting to connect to another country or region must pay the full costs. This situation makes African Internet backbone operators, for instance, subsidize the connectivity costs for international backbone providers to the extent that the current burden of paying international bandwidth costs by African operators are estimated to cost the continent between $250 and $500 million a year. A further point to be mentioned is that when operators become international in the sense that they own communication infrastructures in different countries, their bargaining position in relation to national regulators will strengthen. This phenomenon applies to most countries but will apply specifically to economically weaker countries. All in all, the challenges for regulatory organizations to act in an increasingly international communications environment are significant, taking the shifting relations between national and international aspects of electronic communications into consideration. One way of dealing with this development is to improve the cooperation between regulatory agencies in different regional contexts and internationally. In the southern part of Africa, there is such cooperation in TRASA (Telecommunications Regulators of Southern Africa). The same applies to Latin America with Regulatel (Foro Latinoamericano de Entes Reguladores de Telecomunicaciones). This kind of south-south cooperation is one of the ways forward alongside working for the improvement of the influence on the different kinds of international organizations. ENDNOTES [1] See sub-section on standardisation in the present study. [2] This has, for instance been pointed out by Richard Hawkins, Robin Mansell and Edward Steinmueller: ‘Liberalization and the Process and Implications of Standardization’, in Robin Mansell and Edward Steinmueller: ‘Mobilizing the Information Society’, Oxford University Press, 2002. [3] For an elaborate introduction to ICANN, see Milton Mueller: ´Ruling the Root: Internet Governance and the Taming of Cyberspace’, The MIT Press, 2002. [4] At a public ICANN meting in Accra Ghana in 2002 it was decided to reduce direct user participation. See Hans Klein: ‘ICANN Reform: Establishing the Rule of Law’, a policy analysis prepared for The World Summit on the Information Society, Tunis, 16-18 November 2005, Internet & Public Policy Project, Georgia Institute of technology. [5] This case is well presented in the policy analysis by Hans Klein: Op.cit.

3.2.3 Technology leapfrogging
Technology leapfrogging is a term used to describe the bypassing of technological stages that others (other countries) have gone through. Technology leapfrogging is ‘bypassing some of the processes of accumulation of human capabilities and fixed investment in order to narrow down the gaps in productivity and output that separate industrialized and developing countries'[1]. Leapfrogging involves the technical aspects of implementing new technologies in the existing technological environments. It involves the economic, including financial, aspects, the power and broader social interests related to existing and new technology systems, and a wide range of other socioeconomic factors. However, the prospects of technology leapfrogging in the ICT area seem relatively good in relation to backbone infrastructures as well as access infrastructures and the services delivered, for instance the fast growth of mobile telephony in many developing nations in comparison with fixed-line telephony. Leapfrogging does not necessarily mean that countries which are technologically weak will bypass the other countries hitherto in the lead. An example is load sharing among prepaid mobiles, where one user can give small units to another user directly from one mobile phone to another. Some of the innovations that one will see in low-ARPU (Average Revenue Per User) countries are different from the innovations seen in high-ARPU countries. Often, technology leapfrogging means the catching up by skipping some of the intermediate technology stages. It can potentially lead to situations where the new technologies become dominant in developing countries while having a more complementary role in the economically more developed nations. The reason is that countries with large legacy systems can have inertia problems in switching to new technology systems. In countries with well-established infrastructures, new technologies are often implemented in manners complementary to the existing infrastructure elements. In countries with less developed infrastructures, new technologies may, to a larger extent, substitute older technologies. ENDNOTES [1] This is the definition used by Edward Steinmueller in a paper on the subject: ’ICTs and the Possibilities for Leapfrogging by Developing Countries’, International Labour Review, vol. 140, no. 2, 2001, p. 194. Another often cited paper on the issue is ‘Technology Leapfrogging in Developing Countries – An Inevitable Luxury?’ by Robert Davison, Doug Vogel, Roger Harris and Noel Jones, The Electronic Journal on Information Systems in Developing Countries (EJISDC), vol. 1, no. 5, 2000, pp. 1-10. A further publication is J.P. Singh: ‘Leapfrogging Development?: The Political Economy of Telecommunications Restructuring’, State University of New York Press, Albany, New York, USA, 1999.

FACTORS AFFECTING LEAPFROGGING

The first factor to consider in technology leapfrogging is the technology itself, i.e. how the new technology fits into the existing technology system. Communication technologies are system technologies, and if new technologies do not entirely substitute for existing technologies, there will be an adaptation process. The more modularized communication systems become, the easier it is to insert new elements. And, as Internet based technology solutions are modularized to a much higher extent than the earlier systems, processes of adaptation are getting

easier. Secondly, there are the economic aspects to take into account. On the investment side, financial resources are generally scarce in developing countries. This puts a limit on the deployment of new technologies and often makes it necessary to attract foreign investments. On the demand side, the market prospects can be uncertain in developing countries because of lack of resources among potential users. But experience has shown that demand often is far stronger than the existing diffusion of communication technologies seems to indicate. Furthermore, the possibilities for implementing new communication technologies are far better now, as the costs of reaching users with communication systems and services have decreased. Thirdly, the power relations surrounding technology systems must also be examined. New technology systems are nowhere – not either in developing countries – implemented in a total green field environment. There will always be different vested interests – even in hindering new technology solutions to be implemented. This applies when the new technology systems are implemented by others than the existing dominating companies and, especially, when the new solutions potentially substitute for the older systems. The policies of liberalization already implemented limit these kinds of problems. However, vested interests in existing technology systems play a role as barriers to the implementation of new technologies. A fourth point deals with the broad range of other socio-economic factors. Absorptive capacity, access to equipment and know-how, complementary technologies, and downstream requirements are important factors to be included in the analysis of the potentials for technology leapfrogging[1]. Absorptive capacity deals with the processes of learning and adaptation of new technologies. Access to equipment and know-how concerns the conditions under which technology can be transferred, including the limits created by intellectual property rights. Complementary technologies relate to the systemic character of technologies and the linkages with other industrial sectors. And, downstream requirements deal with the development of the relations with the users. ENDNOTES [1] The factors are mentioned by Edward Steinmueller (op.cit.). The following short description of these factors is also taken from this paper, p. 95-96.

KEY POINTS AND RECOMMENDATIONS
s

Leapfrogging involves the technical aspects of implementing new technologies in the existing technological environments, the economic, including financial, aspects, the power and broader social interests related to existing and new technology systems, and it involves a wide range of other socio-economic factors. The prospects of technology leapfrogging in the ICT area seem relatively good in relation to backbone infrastructures as well as access infrastructures and the services delivered, for instance the fast growth of mobile telephony in many developing nations in comparison with fixed-line telephony. The overall recommendation is to take advantage of such potentials for technology leapfrogging by developing a regulatory framework that favors an open technology environment and technology change.

s

TECHNOLOGY EXAMPLES

In spite of these factors limiting the possibilities for leapfrogging, technology leapfrogging actually does take place. Internet and wireless technologies are making leapfrogging easier than before. The layered and modularized structure of Internet communication makes it possible to implement different aspects of Internet based communication on the basis of existing communication systems. Furthermore, the fact that the intelligence in communication systems is moving from the centers to the peripheries also makes it easier to establish systems in a more decentralized manner with less upfront capital requirements. Moreover, the costs of equipment are decreasing, which also applies to different forms of wireless communication. For example, it is cheaper to reach customers by means of mobile telephony than fixed-line telephony, and, mobile networks are not the only alternatives for local loop access. Different wireless solutions, such as Wi-Fi, WiMAX, etc. can also be used. RELATED INFORMATION Next Generation Access Networks

3.3 Regulatory implications
In this sub-section, six general and transversal regulatory issues are discussed: s Sector specific and/or general competition regulation s Technology neutrality s Infrastructure vs. service competition s Cost based regulation s Alternative business models s Quality of service The aim is to present the latest discussions and developments in the field, setting the agenda for communication regulation in the coming years. However, in order to ‘set the scene’ for these discussions and developments, a brief introduction to the path from the ‘old regulatory regime’ to the ‘new regulatory regime’ is put forward.

With respect to sector specific and/or general competition regulation, the final aim since the beginning of the liberalization of the telecom area has been to reach a situation whereby the area would be regulated by general competition regulation. The final aim may not have been explicit all the while since intermediate aims have dominated the political and economic discourse, such as the creation of competition after a long period of monopoly. This has led to a situation whereby sector specific regulation has been implemented with the purpose of reaching a stage, where it could be lifted again as a result of the establishment of a competitive market. An open question is, however, the extent to which this goal will be reached. Nevertheless, the aim is to move continuously in the direction of creating fair competition on the telecom markets and to lift sector specific regulation when possible. The aim of technology neutrality is related to the aim of general competition regulation. Where telecom regulation, to a large extent, is network and service specific in the present context, the aim is to reach a situation where the choice of technology solutions is left to the companies in the markets. Technology convergence will facilitate more competition in the telecom markets, as new network and service providers can enter markets that previously were served by specific technology solutions, and will therefore increase the possibilities for going from sector specific to general competition regulation. It will also increase the possibilities for going from a more technology specific regulation towards a more technology neutral regulation. This applies to universal service/access where unified licensing procedures as opposed to technology specific licensing can be implemented to a higher degree. It apples to competition regulation where different technology solutions can be used to service the same communication needs, and to radio frequency regulation where the choice of technology solutions can be left to market players to a greater extent. The issue of infrastructure vs. service competition is strongly related to the two aforementioned issues and is also closely connected with technology convergence developments. Infrastructure competition has since the beginning of the liberalization of the telecom area been seen as a more sustainable form of competition than service competition, as service competition, in the present situation with one incumbent and dominant operator in most national contexts, implies a dependence of alternative operators on the incumbent. However, infrastructure competition has been difficult to obtain in fixed line areas, as the requirements on investments of alternative operators have been high, and service competition has therefore been used as a substitute or complement to infrastructure competition. Technology convergence enabling use of different infrastructures for a number of different services, however, renders infrastructure competition more realistic. A related issue concerns infrastructure sharing. In developing countries especially but also in other countries, there can be good arguments for allowing infrastructure sharing, which indeed may lead to less infrastructure competition, but also can foster infrastructure construction and increased access. Regarding cost-based regulation, two important issues are dealt with in the present sub-section. The first issue concerns the development towards basing interconnection prices and end user prices on costs. This has been an important aim since the beginning of liberalization and has been implemented in a large number of countries. However, many developing countries still need to apply more streamlined methods and procedures in their cost calculations. The second issue deals with the broader question of using cost calculations as a basis for technology choices. In the cases where new technologies are cheaper than existing technologies or other technological options, costs should be an important parameter in technology choice and substitution. Regarding the regulatory set-up and business models, there is presently a greater emphasis on exploring all kinds of organizational models in order to expand connectivity. In the first period of telecom liberalization, focus was mostly on establishing competition and on expanding telecom infrastructures by means of the activities of alternative but traditional telecom operators. However, it has turned out that there are limits to this strategy. Competition in important market segments is still not significant, and especially in the wake of the telecom crisis in the first years of the new century it has been difficult for developing countries to attract new operators to their markets. This is one of the reasons why there is increasing emphasis in the present situation on promoting all kinds of alternative, demand-led and cooperative telecom networks. Quality of Service (QoS) is an area which has been part of telecom regulation from the very beginning. The issue, nationally, was for many years the service quality that end users could get. Internationally, there was also the question of the quality of services when interconnecting with other national operators in international communication. With the liberalization of the area, the quality of wholesale services has also become an important issue nationally. The main issue approached in the present sub-section is whether the problems related to QoS have changed fundamentally with the transition from a circuit-switched to a packet-switched environment and whether or to what extent the regulatory issues consequently have been altered. RELATED INFORMATION Price regulation Interconnection Licensing Universal access Spectrum management Numbering

3.3.1 Sector specific and/or general competition regulation
General competition regulation is generally an ex post regulation. This means that public authorities will not intervene in the markets unless specific abuses of market power are registered, i.e. interventions are made in a reactive manner. In such situations, the competition

authorities will examine the specific cases of abuse and will make their rulings. Sector specific regulation, on the other hand, has traditionally an ex ante character which means that rules in addition to the general competition rules are laid down determining in advance how the market players are supposed to act, for instance the rules for asymmetric interconnection regulation with specific provisions on how market players with dominant positions should treat other market players.

KEY POINTS AND RECOMMENDATIONS
s

s

There are basically three market oriented reasons for public intervention in markets: market failures, social concerns and industrial policy. All three reasons have played a role in telecom and have contributed to the sector specific regulation in the area. However, the present discourse on sector specific and/or general competition regulation is centered on the level of competition reached in the different market segments. The issue of sector specific and/or general competition regulation is most prominent in economically more developed countries but applies equally to developing countries. If analyses document that a sufficient degree of competition has been attained in a market segment, specific regulations need not be applied and general competition rules, where they have been established, will apply in this segment. The issue is not necessarily sector specific regulation or general competition regulation at the general level. There will and should be a combination of general competition rules and sector specific regulations according to the different situations in the different market segments.

PRESENT DISCOURSE

In the present discourse on telecom regulation in the economically developed nations, the question of sector specific and/or general competition regulation is related to two intertwined issues: one on the market implications of technology convergence; the other one on the development of the telecom markets towards more ‘normal’ market conditions[1]. The reason that they are intertwined is that convergence, to the extent that it opens telecom markets to increasing competition, provides new modes of access, and eases the restrictions on radio frequency usage – will contribute to a normalization of the telecom area. It should be noted that there can be other reasons for having specific regulations of individual sectors besides their degree of normalization in terms of level of competition. In the case of the telecom area, one of the reasons has been the infrastructural character of telecom in society, i.e. its crucial role in the general social and economic communication processes. There are basically three market oriented reasons for public intervention in markets: market failures, social concerns and industrial policy. All three reasons have played a role in telecom and have contributed to the sector specific regulation in the area. However, the present discourse on sector specific and/or general competition regulation is centered on the level of competition reached in the different market segments. ENDNOTES [1] The reason that the word normal is in inverted commas is that it is difficult to define normal market conditions. Conditions on most markets are different and seldom live up to text book requirements for competitive markets. With the words normal and normalization are, therefore, meant conditions approximating genuinely competitive conditions on the markets.

REGULATORY AREAS

The present sector specific regulation in telecom was implemented in connection with the liberalization processes in the area. The main fields of sector specific regulation are universal service/access, interconnection and regulation of limited resources (radio frequencies, rights of way and numbers). Although this may not have been the general understanding from the beginning, a main goal of sector specific regulations has been to reach a stage in market developments in the area, where special regulation can be lifted. This applies to universal service/access, if market mechanisms are able to cater to the optimal spread of access technologies in the specific markets without any special universal service/access regulation. It also applies to interconnection regulation where, at least, the asymmetric parts can be disbanded, once it can be determined that there is sufficient competition between different providers. An open question is whether interconnection regulation of the relations between operators in a competitive market must continue, or whether network effects automatically will lead operators towards interconnection. It finally applies to the regulation of limited resources to the extent that new technological solutions will contribute to solving the scarcity issues. However, as there always will be some kind of exclusivity aspects in these scarcity issues, some kinds of special regulations (or self-regulations) will always be necessary. In the European Union, the present regulatory framework is based on the assumption that telecom markets are moving in the direction of normal competitive markets, and that convergence is part of the foundation for this development. In order to determine whether different telecom market segments are sufficiently competitive to lift sector specific regulations, a large number of market segment analyses are performed in each EU member state based on traditional competition analysis methods regarding market dominance[1]. If analyses document that a sufficient degree of competition has been attained in a market segment, specific regulations will be abandoned and general competition rules will apply in this segment. Conversely, if competition analyses show that there is (still) a situation with significant market power of one operator (or a couple of operators with collective dominance), sector specific regulation will be maintained/implemented. The issue is, therefore, not necessarily sector specific regulation or general competition regulation at the general level. It will be a combination of general competition rules and sector specific regulations according to the different situations in the different market segments. ENDNOTES[1] European Commission: ‘Commission guidelines on market analysis and the assessment of significant market power under the Community regulatory framework for electronic communications networks and services, 2002, http://ec.europa.eu/information_society/topics/telecoms/regulatory/maindocs/documents/c_16520020711en00060031.pdf.

3.3.2 Technology neutrality
In a telecoms regulatory context, the concept of technology neutrality means that different technologies offering essentially similar services should be regulated in similar manners. However, technologies offering similar services do not necessarily have similar features in all aspects, and exactly identical regulations may, therefore, result in the advantage of one technology over another in the market. Technology neutral regulation can, consequently, include slightly differing regulations for different technology solutions in the same market segments.

IMPLICATIONS

The technology neutrality concept has implications on most regulatory issues in telecom. With respect to universal service/access, the implication is that services designated for universal service/access can be provided on different platforms, and that policies should not promote a specific technological solution (mostly a PSTN solution). Technology neutrality will lead, for instance, to a policy of unified licensing, as different technology solutions can deliver similar services. And, it will also lead in direction of a policy of similar regulation of different technologies delivering similar services, taking into account the possible difficulties in implementing this principle entirely, as the services on different technology platforms are not entirely identical. The emergency systems, for instance, are mostly based on specific properties of the PSTN system, which VoIP, for example, does not have. This is why there should be a possibility for slightly different regulations of different services in the same market segment. Regarding radio frequency assignments, the traditional mode of operation in most countries has been to assign frequencies for specific technology solutions. However, this practice can be modified and the policy goals can be to implement frequency assignment systems with greater technology choice. This means that different radio technologies can be used to deliver radio based services in the same market areas. However, it should be noted that these decisions are partly guided by national or regional technology strategic interests. In one national or regional context, a special technology solution can be preferred for different reasons, among them the support for specific technology producers located in the countries and regions in question. This may not be a desirable situation but can influence technology choices and the degree of technology neutrality. Finally, technology neutrality affects the competitive situations on the markets. Technology neutrality in the choice of technology solutions for delivering access or services will lead to increased competition, as different solutions can compete on the same markets. The OECD, for instance, uses the concept of technology neutral regulation to promote inter-modal competition between different access technologies [1]. ENDNOTES [1] OECD: ‘Regulatory Reform as a Tool for Bridging the Digital Divide’, OECD, Paris, 2004, http://www.oecd.org/dataoecd/40/11/34487084.pdf

INTERPRETATIONS OF THE CONCEPT

The meaning of the concept of technology neutrality slightly differs internationally. The technology neutrality concept first gained ground in Europe with the regulatory developments in the EU but also plays a role in other regulatory contexts internationally. In the EU, the concept is closely related to the market analyses mentioned in the previous sub-section on ‘Sector specific and/or general competition regulation’. Instead of regulating services, a central intention of the New Regulatory Framework[1] in the EU is to regulate markets as in general competition regulation. This means that instead of regulating technologies delivering specific services, regulation - or the lifting of regulation - is directed towards markets, where different technology solutions can be used to deliver services which are similar and, therefore should be subject to similar regulation, with the abovementioned proviso that exactly identical regulation should be avoided if the result is that one technology is given a competitive edge in relation to others[2]. The manner in which markets are defined is on the basis of the market concept used in general competition regulation, i.e. based on substitutability analyses including demand substitution as well as supply substitution[3]. In the EU and elsewhere, the concept of technology neutrality is also used in a broader sense. It is becoming a basic guidepost for regulation around the world. The concept is used as a prescription for limiting public intervention in the choice of technology solutions in the markets. Technology neutrality is based on technology convergence, as similar services can be supplied on different technology platforms, and as regulations should seek to promote competition between different technology solutions instead of ‘picking a winner’. However, the implications of technology neutrality go beyond technology convergence, as the concept is based on a more profound policy of limiting public intervention in the directions of technology development[4]. The idea is that market mechanisms are better at making these choices, and that the risks of ‘wrong’ technology choices by the public sector are substantial. ENDNOTES [1] Information on the EU New Regulatory Framework can be found on: http://europa.eu.int/information_society/topics/telecoms/regulatory/new_rf/index_en.htm [2] A good introduction to the use of the concept in the EU is Peter Alexiadis and Miranda Cole: ’The Concept of Technology Neutrality …’, ECTA Review, 2004, p.76-80. [3] For a short introduction to competition law and substitutability analyses: Office of Fair Trading: ’Market Definition: Understanding Competition Law’, Office of Fair Trading, UK, 2004: http://www.oft.gov.uk/shared_oft/business_leaflets/ca98_guidelines/oft403.pdf [4] This applies, for instance, to a US-based publication advocating keeping Internet an open network environment and refraining from

implementing in the Internet environment the regulations to which the more traditional telecoms is subject: Mark Cooper (ed.): ‘Open Architecture as Communications Policy: Preserving Internet Freedom in the Broadband Era’, Center for Internet and Society, Stanford Law School, 2004.

KEY POINTS AND RECOMMENDATIONS
s

s

s

The concept of technology neutrality means that different technologies offering essentially similar services should be regulated in similar manners. However, exactly identical regulations may result in the advantage of one technology over another in the market. Technology neutral regulation can, consequently, include slightly differing regulations for different technology solutions in the same market segments. Technology neutrality is based on technology convergence, as similar services can be supplied on different technology platforms, and as regulations should seek to promote competition between different technology solutions instead of ‘picking a winner’. However, the implications of technology neutrality go beyond technology convergence, as the concept is based on a more profound policy of limiting public intervention in the directions of technology development. The technology neutrality concept has implications on most regulatory issues in telecoms, including universal service/access, frequency management and competition. In a number of developing countries where narrow technology licenses have traditionally been granted, the technology neutrality concept should be used to promote a greater degree of unified licensing.

3.3.3 Infrastructure vs. service competition
An important topic, these years, is the issue of infrastructure and service competition. The background is that infrastructure or facilitybased competition is seen as more sustainable than service-based competition, where alternative operators competing with the incumbents have to rely on the infrastructure elements provided by incumbents. The contention is also that if service competition is made too easy, so that the conditions for using the infrastructures of incumbents are too favorable in comparison to building infrastructure, the new operators in the markets will refrain from investing in infrastructure, and competition will never become sustainable enough to lift sector specific competition regulation.

ACCESS COMPETITION

In economically developed nations, focus with respect to infrastructure and service competition has been on access competition. In the backbone area, operators competing with the incumbents have often established infrastructures, and the ‘soft’ spot of creating a competitive market has been the access area, where the incumbents generally are dominant. In the beginning of the liberalization process, the economically developed nations have given priority to promoting competition by way of service-based competition formulas. Policies have not been identical in the different countries, and some countries have pushed harder than others for infrastructure-based competition, for instance the UK. Most countries, however, have been seeking to combine service and infrastructure competition, but with service based competition playing a major role[1]. With the current intention to arrive at a situation where sector specific regulation can be minimized, there is increasing emphasis on infrastructure-based competition. Broadband is at the center of attention in the economically developed countries, as this is the new infrastructure question in these countries. Furthermore, this is an area where new operators have a greater chance than in the traditional narrowband area of competing on a level playing field with the incumbents. However, it is also an issue affecting narrowband services, for instance telephony, as this service nowadays can be provided on many different technology platforms. ENDNOTES [1] Papers on local loop unbundling regulation and the issue of infrastructure and service competition (in respectively, USA, UK, Italy, Germany, Sweden, Netherlands and Denmark) were presented at the 16th European Regional Conference in September 2005 in Porto, Portugal: http://userpage.fu-berlin.de/~jmueller/its/conf/porto05/start.html

KEY POINTS AND RECOMMENDATIONS
s

s

Infrastructure and service competition has become an important issue. The background is that infrastructure or facility-based competition is seen as more sustainable than service-based competition, where alternative operators competing with the incumbents have to rely on the infrastructure elements provided by incumbents. In developing countries with little developed access infrastructures, new operators will often be inclined or forced to build their own access loops. This can be seen as an advantage compared with many developed countries but puts great emphasis on the necessity of network interconnection rules and also raises the discussion on network sharing in order to hasten infrastructure construction. The optimal policy is thus to support all kinds of competitive strategies and network expansion strategies. Instead of viewing service and infrastructure-based competition as alternative policy options, an optimal policy is to consider them as complements. The reason is that infrastructure and service markets partly are different market segments with different market players.

NEW ACCESS INFRASTRUCTURES

In developing nations, access infrastructures are generally underdeveloped. This means that, potentially, there is room for new operators to reach un-served end users with access infrastructure provision. In a number of developing countries, new operators have difficulties in obtaining reasonable interconnection agreements with incumbents. The difficulties of ensuring access to the incumbents’ networks have resulted in infrastructure-based competition preceding service competition in many developing countries. To the extent that new operators are successful in building such access infrastructures, there is an important issue of network interconnection with other operators and primarily the incumbents. However, there are also operators using the access networks of incumbents, first and foremost in the Internet area as Internet Service Providers. A regulatory framework including a combination of tools for advancing service as well as infrastructure-based competition is, therefore, appropriate. However, great emphasis will placed be on the establishment of new access infrastructures because of the under-development in this area. Different forms of cooperatives can be one of the ways to expand access. In some countries such arrangements have existed for many years (see practice note on Argentina).

T H E ‘LADDER’ THEORY

The discussions have centered on whether and to what extent infrastructure and service competition exclude or supplement each other, and whether there is a ‘ladder’ in the markets from service to infrastructure-based competition, so that operators will advance in the markets from service-based competition to infrastructure-based competition. The ‘ladder’ theory has hitherto been the dominating framework, but has increasingly been criticized for not putting enough emphasis on infrastructure competition and even, in reality, being a barrier to infrastructure competition. The individual operators will make their choices regarding different competitive strategies based partly on the immediate costs and benefits of different business models. However, strategies are also determined on basis of longer term approaches including decisions on building own infrastructures or relying on the infrastructures of other operators. Instead of looking solely at the individual operators and their ascent up the ‘ladder’, it is also necessary to examine the long-term strategies of the different operators. Some operators will build their own infrastructures under all circumstances, while other operators will concentrate on a service-based approach. The optimal policy is, therefore, to support all kinds of competitive strategies. Instead of viewing service and infrastructure based competition as alternative policy options, an optimal policy is to consider them as complements.

3.3.4 Cost-based regulation
In the monopoly period before the start of the telecom reform process in the 1980s, rate of return regulation was used in, for example, the US to regulate end user prices of the national operator. The aim was to limit excessive profitability of the monopoly company, but rate of return regulation suffered from the downside that it did not create any incentive to increase efficiency, as there was a fixed rate of return added to whatever costs the operator would have. If an operator was inefficient and therefore had high costs, it would still be able to claim the same rate of return and, consequently, a higher profit than if it increased its efficiency.

C O S T -BASED PRICING

In relation to the liberalization of the telecom area, cost-based end user and interconnection pricing was implemented. The principle in costbased pricing is that prices are regulated on the basis of the calculated costs plus an add-on profit percentage. Operators can claim this price even if their actual costs are lower because of increased efficiency. At first, most countries implemented historical costs as the basis for cost accounting, as these were the immediate cost figures available. However, historical costs bear some of the same disadvantages as rate of return regulation, as an inefficient operator with high historical costs will be able to charge higher prices than a more efficient operator with lower historical costs. The costing method which, consequently, often has been chosen as the preferred one and which has been implemented in several countries, especially the economically developed countries, is based on forward looking costs. Costing calculations are based on the marginal costs that an operator would incur if it were to build similar network elements in the current situation. The elements are assembled in larger increments and the costing method is, therefore, known as the Long Run Incremental Cost (LRIC) method. The primary advantage of this method is that it rewards the efficient operators. Furthermore, the price signals sent to the markets will create a more forward looking basis for the choices made by alternative operators with respect to using the network resources of the existing operator or investing in their own network resources. The prices set on the basis on the LRIC method will not always be lower compared with historical costs even though the present technological solutions are more cost efficient than the solutions used hitherto. When using historical costs, parts of the original costs are written off, and the costs of building a new similar network increment may, therefore, be higher. This may apply, for instance, to the costs of unbundled local copper loops used mostly for DSL (Digital Subscriber Lines), while the LRIC prices for switched interconnection products will be likely to be lower than prices based on historical costs. There are, in addition, other costing and pricing methods, e.g. Wholesale Line Rental (WLR), where operators can lease network elements for a price equal to the retail price minus a percentage rate. Furthermore, for the purpose of lowering prices, as new technologies make networks more efficient, price caps can be used. This applies to retail prices as well as interconnection prices and is well-known, for

instance, in the case of retail price regulation, where RPI-x (Retail Price Index minus a percentage rate) is a method often used. There are thus different costing and pricing methods available, and often regulators have a discretionary choice with respect to the methods to be used in different cases. Even though LRIC is the method sending the right forward looking price signals to the market and does not carry the weight of historical inefficiencies, there may be reasons for choosing another method. One such reason can be that calculating and negotiating the LRIC based prices can be a very lengthy and resource consuming exercise. This can be a very good reason in a developing country, where regulatory resources can be scarce. Instead of implementing a time consuming LRIC process, it can be reasonable to apply simpler costing and pricing methods such as cost and price comparisons with similar countries.

COSTS AS A BASIS FOR TECHNOLOGY CHOICE

No matter which costing and pricing methods are used, it is important to implement the basic principle of cost based regulation. This will facilitate the development of more fair retail as well as wholesale prices. Furthermore, cost-based regulation can also be used as a tool for promoting technology changes. Eventually, it will be the market players who determine which technologies to use. They are in a better position than regulatory authorities to choose between different technology options. However, knowledge on the costs of different technology solutions can be used to shape regulations so that market players have incentives to take up new technology solutions.

KEY POINTS AND RECOMMENDATIONS
s

s

In relation to the liberalization of the telecom area, cost-based end user and interconnection pricing was implemented. The costing method, which has often been chosen as the preferred one and, which has been implemented in several countries is based on forward looking costs. The advantages of this method is that it rewards the efficient operators, and that the price signals sent to the markets will create a forward looking basis for the choices made by alternative operators with respect to using the network resources of the existing operator or investing in their own network resources. There may, however, be reasons for choosing another method. One such reason can be that calculating and negotiating the LRIC based prices can be a very lengthy and resource consuming exercise. This can be a very good reason in a developing country, where regulatory resources can be scarce. Instead of implementing a time consuming LRIC process, it can be reasonable to apply simpler costing and pricing methods such as cost and price comparisons with similar countries. It is important to implement the basic principle of cost based regulation. This will facilitate the development of more fair retail as well as wholesale prices. Furthermore, cost based regulation can also be used as a tool for promoting technology changes. Knowledge on the costs of different technology solutions can be used to shape regulations so that market players have incentives to take up new technology solutions.

3.3.5 Alternative business models
The specific implications of technology and market developments on regulations are examined in the second section of the module on technology implications. In the present sub-section, the aim is to treat the issues of licensing, interconnection, universal service/access and resource management in a transversal manner, discussing how the regulatory set-up as a whole will develop, taking technology and market developments into consideration. Furthermore, new business models and alternative organizational forms and providers are discussed. The overall purpose of the regulatory paradigm must be to open as many paths to network development as possible and to refrain from restraining these developments. Technology developments offer a wide range of new network technologies in access as well as core network parts, and a fundamental goal of policies and regulations in the field must be to take advantage of these new technologies in order to increase access and service provision.

DIVERSIFYING PARTICIPATION

This applies not only in a technology sense, promoting different ‘access pipes’ to the customers. It also implies promoting the diverse range of different business models and organizational forms. Telecom has traditionally been dominated by large companies benefiting from economies of scale and scope. These players will continue to play important roles in the area, but regulations should also seek to promote other organizational forms, including alternative operators and more demand led initiatives, such as end-user organized networks. Telecentres, demand aggregation, alternative operators offering new solutions – all such initiatives must be promoted by means of regulations and not hindered in their development. Though it is possible that some of these types of initiatives, if successful, will eventually be taken over by larger telecom operators, they will contribute to the overall growth of telecom access. It is also a possibility that future telecom structures, to a larger extent, will be based on end-user organized network elements. An important challenge for regulation is, therefore, to enable the combination of such end-user organized initiatives with the more traditional network and service operators. From a technology point of view, this entails the interconnection and interoperability of a heterogeneous web of different network elements and services. More importantly, from a regulatory point of view, it requires a framework allowing for such interconnection and interoperability of network elements and services.

FOCUS OF REGULATION

Among the different areas of regulation and regulatory means that authorities in the field mostly work with, i.e. interconnection, resource management, universal service/access and licensing, universal service/access stands out as the central issue in present regulatory developments in developing countries. It also stands out as the centerpiece in the economically developed nations with respect to broadband access. The expansion of broadband access is a central policy goal in developed countries, though it may not be part of a universal service policy, which is most often limited to narrowband access. Nevertheless, the goal of increasing access applies to developing as well as developed nations. This emphasis does not mean that other regulatory issues or means are less important. Interconnection is, as mentioned, of crucial importance and so is the management of limited resources in the areas of radio frequencies, names and numbers, and rights of way. It, however, means that the different regulatory issues all should be guided by the aim of increasing access. RELATED INFORMATION Universal access

KEY POINTS AND RECOMMENDATIONS
s

s

s

The overall purpose of the new regulatory paradigm must be to open as many paths to network development as possible and refrain from controlling and restraining these developments. This implies promoting the diverse range of different business models and organizational forms. Telecoms have traditionally been dominated by large companies benefiting from economies of scale and scope. These kinds of players will continue to play important roles in the area, but regulations should also seek to promote other organizational forms, including alternative operators and more demand led initiatives such as end user organized networks. Among the different areas of regulation that regulatory authorities mostly work with, universal service/access stands out as the central issue in present regulatory developments. This absolutely applies in developing countries where access is still limited. This does, however, not mean that other regulatory issues are less important. Interconnection is of crucial importance and so is the management of limited resources. It, however, means that that the different regulatory issues all should be guided by the aim of increasing access.

3.3.6 Quality of Service (QoS)
Quality of Service (QoS) is an area where national public authorities traditionally have exerted an influence. It is also an area where international negotiations on standards have played an important role. In developing countries, the problems of QoS are among the biggest issues in telecom. Often quality is poor on all parameters, e.g. outage, repair time, voice quality etc. In the present subsection, being part of a presentation of the new regulatory paradigm, focus is on QoS in packet-switched networks. The first question in this connection is whether and to what QoS issues have changed with the transitions from circuit-switched telephony to the present and future multi-service packet-switched environment. The second question is whether there is a need for public regulation regarding QoS and, if so, which kind of regulation. In this connection, there are two different areas to consider: QoS in relations between the operators, e.g. in interconnection agreements, and QoS in relationships between operators and end users. RELATED INFORMATION IP QoS

KEY POINTS AND RECOMMENDATIONS
s

s

s

s

s

In developing countries, the problems of QoS are among the biggest issues in telecom. Often quality is poor on all parameters. This applies to the PSTN technology and also applies to Internet technology. Internet is presently a ‘best effort’ network, which is not a big problem for most data communications, as smaller delays are unimportant for the transmission of data files, and as packets can be re-transmitted. However, it does constitute a problem for realtime communication as telephony. The technical parameters for QoS are different when comparing circuit-switching and the Internet. However, the real differences in terms of regulatory implications are related to the multi-service and multi-operator environment of the Internet and to the fundamentally global character of the Internet. The areas where regulation can have a role to play are in connection with service level agreements (SLAs) in relation to interconnection agreements between dominant market players and competitors depending on the quality of the interconnection services delivered by the dominant operators. Furthermore, QoS for end-users can be secured by way of regulatory provisions. Users will often experience that the QoS delivered does not correspond to the promises made, for instance with respect to transmission speed or the quality of VoIP services. There are, however, good reasons to hesitate with respect to strict regulatory interventions in the field of QoS on the Internet. One reason is the very dynamic character of the Internet and the continuously changing technology solutions used. Another reason is that the Internet environment is mostly competitive, and regulatory intervention should, accordingly, be light-handed.

REGULATION

The protocols for such prioritization systems are negotiated and decided upon in international organizations, primarily IETF (Internet Engineering Task Force), and are implemented by operators acting in the markets. In that sense, there is nothing much different from the former circuit-switched environment with ITU as the predominant international standardization organization. The technical parameters for QoS are, indeed, different when comparing circuit-switching and the Internet. However, the real differences in terms of regulatory implications are related – not only to the multi-service environment – but also to the multi-operator and fundamentally global character of the Internet, where the multi-operator character is a horizontal as well as vertical phenomenon because of the many different horizontally competing operators delivering Internet access and the vertically layered structure of Internet communications. The implementation of packet-switching technology allows for a layering of communication systems and a market division of labor between operators on the different layers. The relevant levels of QoS, therefore, have to be secured on all layers. The layering of communication systems provides a greater degree of transparency of the different layers of communication processes but also creates a more complex environment with different operators working on different levels. Furthermore, the truly global character of the Internet makes it difficult to secure QoS, as communications go through routers located in different parts of the world depending on the availability of transmission capacity at the particular moment of communication or information transfer. The role of regulation must be adapted to these circumstances. Standards providing the basis for QoS communication are mostly developed in international standardization organizations and are implemented by the market players in their international and national operations. The role of regulators can be to promote these standards and even enforce them if necessary. But most often, market players will see it as being in their best interest to implement the standards developed and no regulatory interventions will, therefore, be necessary with respect to standards implementation. The areas where regulation can have a role to play are, on the one hand, in connection with service level agreements (SLAs) in relation to interconnection agreements between dominant market players and competitors depending on the quality of the interconnection services delivered by the dominant operators. If, because of market dominance of an operator, it is difficult for competitors to obtain agreements securing their customers a sufficient level of QoS, there may be a need for regulation of interconnection agreements obliging the dominant operators to live up to certain requirements. Furthermore, QoS for end-users can be secured by way of regulatory provisions. Users will often experience that the QoS delivered does not correspond to the promises made, for instance with respect to transmission speed or the quality of VoIP services. There are, however, good reasons to hesitate with respect to strict regulatory interventions in the field of QoS on the Internet. The reason is the very dynamic character of the Internet and the continuously changing technology solutions used. The Internet environment is mostly competitive, and regulatory intervention should, accordingly, be light-handed. If the markets for Internet services are not sufficiently competitive – which is the situation in many developing countries – there is obviously a need for regulation. The criterion for competition regulation must be the status on the markets and not the technology itself. Self-regulation is also a possible tool in the interconnection between market players as well as in the relations between companies and end-users. With respect to end-users, one of the possibilities is to facilitate a more competitive environment by publishing QoS service data for the services delivered by the different operators in the market.

T R A N S I T I O N T O P A C K E T -SWITCHING

In a circuit-switched telephony environment, QoS refers to the call completion rate, downtime, noise etc. and the underlying technical parameters securing a sufficient voice quality. In principle, QoS in a connection-less packet-switched Internet environment[1] is similar in the sense that the quality as perceived by the users can be measured and the technical parameters to secure a sufficient quality can be determined. However, circuit-switched PSTNs are optimized for voice service, while Internet is optimized for data transfer. When Internet is used for other kinds of services, new problems also arise because of the differences in quality requirements of different services (e.g. data files, voice or video). Internet is presently a ‘best effort’ network, which is not a big problem for most data communications, as smaller delays are unimportant for the transmission of data files, and as packets can be re-transmitted, but it does constitute a problem for real-time communication as telephony[2]. The problems encountered in a connection-less packet-switched network like Internet can be insufficient throughput, packet loss, latency and jitter (variable delays) which may cause different problems for different services. Insufficient throughput, latency and jitter cause problems for real time video; latency and jitter are problematic in relation to voice services; and packet loss may constitute a problem in connection with the transmission of data files. There are two basic solutions to these problems. One is to have sufficient (or too much, i.e. over-provisioning) capacity in the network, and the other one is to prioritize communications so that, for instance, real time communication is given priority over less time-dependent services. Such prioritization is, presently, implemented in two different levels of quality, integrated services (IntServ) and differentiated services (DiffServ), where IntServ is a finer grained prioritization while DiffServ is coarser grained. The new IPv6 also includes functionalities allowing for prioritization of different kinds of communication. ENDNOTES [1] There are two basic types of packet-switched networks, connection-oriented and connection-less. In a connection-oriented network a virtual circuit is established and the throughput of packet can be secured. Asynchronous Transfer Mode (ATM) is an example. In a connection-less network packets will be transmitted via different paths. The present Internet is an example. [2] QoS issues are, for instance, dealt with in Wikipedia: http://en.wikipedia.org/wiki/Quality_of_service

3.4 Policy Integration
This section on policy integration takes a broader view on telecom regulation. First, there is a presentation and examination of the different types of public policy market intervention going from the least to the more interventionist modes. The aim of this is to analyze the relationships between different types of public policy intervention emphasizing the importance of developing a coherent public policy. An aspect of this is the promotion of innovativeness in the telecom sector. Innovation has not been the main focus of telecom policy, but has come increasingly to the fore, as the necessity of continuous innovations has gained growing attention. Moreover, the issue of standardization is taken up. Standardization in the telecom area has increasingly become a supra-national activity, and private interests have gained growing influence. However, specific applications are often dealt with by national authorities, and the role of standardization in the development of a coherent new telecom regulatory paradigm is examined. Furthermore, public-private partnerships are also dealt with and, finally, security issues are examined. Security issues have always had a central position in communication regulations, but the specific aspects change with technology and market developments. RELATED INFORMATION Information society technologies

3.4.1 Integration of different categories of public policy intervention
This section presents the broader policy context, modes of intervention and respective key points and recommendations.

KEY POINTS AND RECOMMENDATIONS
s

s

Communication regulation must be seen in the broader context of policy measures related to the development of ICT infrastructures and services. However, the principle of independent regulation must be upheld in order to secure a reliable, stable and accountable regulation. Most regulatory provisions will be similar in the vast majority of countries. There will, however, be some differences, especially with respect to the emphasis on the different areas of regulation. However, the mix of policy initiatives and the more specific features of the different kinds of policy initiatives will vary.

MODES OF INTERVENTION

The most direct form of public market intervention or construction in the communication area is public ownership and, therefore, also management of communications systems. Of course, publicly owned operators may act on market conditions in a liberalized market, which is the situation in many countries. However, the strongest form of public market intervention is, nevertheless, when there are public funds involved in the construction of communication systems. This still occurs, most often, in limited market segments, when states use public funds to finance infrastructures in, for instance, underserved areas. Apart from this direct form of market intervention, there are a number of other kinds of policy initiatives. In the following, six different modes are listed, going from the lightest forms of intervention to the strongest at the end: s Strengthening and harmonization of the internal uses of ICT infrastructures and services in public institutions. s Construction of communication systems and structures in relation to the citizens and business enterprises, with influences on the take-up and forms of communication used in the society at large. s Facilitation of the development of communication systems. This may include creating more transparency in the markets by way of public information on qualities and prices of communication services, and it may include the setting up of forums for discussion of, for instance, interconnection and frequency issues among the competing operators and the public authorities. s Regulation proper setting the ‘rules of the game’ in the markets and the enforcement of these rules. s Support for the demand for communication systems and services, which may be based on either the direct demand from public institutions or support for the demand from private citizens and business enterprises. Included in this category can also be mentioned education initiatives teaching people in using ICT equipment and services. s Support for the supply of ICT equipment and services, which may involve public funding of ICT companies, but which may also be of a more indirect character involving public research and development and also public education of people whose labor power will be used in business enterprises. The list illustrates different modes of public policy intervention and market creation. Other kinds of sub-divisions of policy initiatives can be made, but the list enumerates the most commonly used modes of intervention. Not only are they up for consideration for possible policy initiatives, they are actually used in most countries. However, the mix of policy initiatives and the more specific features of the different kinds of initiatives vary. The challenge of the individual countries lies in the best mix of initiatives in their specific contexts.

THE BROADER POLICY CONTEXT

Regulation is one of several different modes of public policy intervention in and construction of markets. The present study focuses on communication regulation, but regulation should be seen in a broader context of policy measures related to the development of ICT infrastructures and services. The point is not that all different modes of policy initiatives should be made part of communication regulation. The politization of regulation should, furthermore, be avoided. It is important to uphold the principle of independent regulation in order to secure a reliable, stable and accountable regulation. However, it is also important to view communication regulation in the context of the broader array of different ICT policy areas in order to be able to lead a coherent public communication policy. This means that the specific regulatory goals and activities should, on the one hand, be guided by the general policy visions and decisions in the ICT field in order to be able to prioritize regulatory activities. On the other hand, regulatory independence should be defended. There is no contradiction in this, as the independent regulation is guided by the overall policy goals set at the policy level. Such a division of power is in accordance with the general principles of divisions of power in the sphere of public institutions. This is, in fact, the manner in which communication regulation is dealt with around the world. New regulatory provisions were, for instance, a vital part of the information society visions and plans from the first part of the 1990s, which a vast number of countries put forward[1]. Regulatory changes have also continuously been part of the more comprehensive ICT policies, as is also the case in relation to the processes in connection with the World Summit on the Information Society (WSIS)[2]. The obvious reason is that communication infrastructures and services constitute the foundations on which a much broader range of activities build, whether they are business activities or other social activities. The development of appropriate ‘tools’ (infrastructures and services) for communication is crucial for all other business and social activities to develop, for instance e-business and e-government. Most regulatory provisions will be similar in the vast majority of countries. There can, however, be some differences, especially with respect to the emphasis on the different areas of regulation. In developing countries, there will often be focus on universal access provisions, as it is considered to be important to extend the means of communication to all areas of the countries (see sub-section IV.3.5.b on ‘Focus of regulation’). In the economically more developed countries, universal service policies have become less important during the past few years[3]. However, the extension of broadband access has come high on the policy agendas in developed nations lately, but seldom under the label of universal service with all the accompanying provisions regarding availability, affordability and quality. The approach to communication regulation as part of a broader policy setting has also implications for the understanding of why regulatory interventions can be necessary and, therefore, also the kinds of interventions made. The economic reasons for regulations are often presented as based on failures in the markets, i.e. that markets do not live up to the requirements for fully competitive markets. On the other hand, arguments against regulatory interventions are based on arguments regarding policy failures, i.e. that policy intervention distorts the markets and leads to even less well-functioning markets. Such a conceptual framework is, however, too simplified and seldom reflects the complex reasons for market intervention. In reality, there will often be a number of different reasons for intervention, and the theoretical framework for conceptualizing these reasons is better understood in a broader context of system failures encompassing the complex mix of market and policy failures. This will also make it easier to situate the regulatory provisions in the broader context of ICT or information society policies. ENDNOTES [1] See, for instance, Anders Henten, Morten Falch and Knud Erik Skouby: ‘European Planning for an Information Society’, Telematics and Informatics, vol. 13, no. 2/3, 1996, pp. 177-190. [2] http://www.itu.int/wsis/ [3] In Europe, for instance, universal service policies in the telephony area were implemented in the 1990s very much as a reaction to the liberalisation of the telecoms area. Universal service in telephony is, however, reached in most European countries, via fixed and/or via mobile connections. Basic universal service policies are, therefore, not a matter of high priority in a number of European countries.

3.4.2 Focus on innovation
In connection with the liberalization of the telecoms area in the 1980s and 1990s, the primary emphasis in telecom policy and regulation in most economically developed nations has been on establishing competition and lowering prices. In developing nations, focus has, first and foremost, been on increasing take-up (penetration) on the basis of well-established technology solutions (PSTN) but also, and more importantly, by means of mobile communications. During the past few years, however, there has been an increasing awareness of the importance of promoting innovation in the sector[1]. This very much has to do with the emergence of new access technologies in the wireless area, for instance wireless LAN and MAN (e.g. Wi-Fi and WiMAX), as well as wire-based solutions such as cable modems, fibers and power lines. The increasing interest in promoting innovation is centered on the introduction of new access technology solutions. However, innovation in new applications, services and content is as important and plays a major role in user demand and the demand for new access facilities. ENDNOTES [1] A good example is a report prepared for the Swedish NRA, Post- och Telestyrelsen (PTS), by Erik Bohlin, Paola Garrone and Erik Andersson: ’Investment, Innovation and Telecommunication Regulation: What is the Role of the NRA?’, PTS, September 2004.

OPEN THE SECTOR FOR INNOVATIONS

Technology convergence is at the center of attention, as convergence developments constitute the basis for the development of new access technologies, i.e. the linear innovations related to new mobile technologies as well as the non-linear wireless LAN and MAN

technologies and ad-hoc technology solutions. With the purpose of facilitating the use of the best of such different technology solutions, it is important that regulation is as open as possible to all kinds of access, application and service contributions to the local, regional and national ICT and telecom developments. This implies maximizing participation via open access, development of all technology advantages, maximum opportunities for competing firms to achieve public interest goals, and stimulation to market development through private as well as public participation[1]. It is important that communication regulation allows for the widest possible range of innovative activities in the sector. Regulation should contribute to opening the sector for innovations, and regulation should also be seen as part of a broader national strategy for innovation of the communication area. It is specifically important that regulation promotes the introduction of new technological possibilities for providing access and also facilitates the entry of new operators and organizational innovations. New technology solutions as well as new kinds of organizations in the area such as community organizations setting up telecom facilities require flexibility in the interconnection of networks and interoperability of services. In contrast to a model with a limited number of technology solutions, a model with a wider range of technologies in use and a larger degree of differentiation in the kinds of operators and organizations active in the field requires much flexibility in regulations. The issue of innovation clearly illustrates that the relationship between technology change and regulation goes two ways. It is not solely a question of adapting regulation to technology changes, but regulatory changes and flexibility are necessary in the promotion of technology and organizational innovations in the area. ENDNOTES [1] For more information and discussion on this approach see the 3rd cycle of the research activities of the World Bank supported World Dialogue on Regulation for Network Economies on ’Diversifying Participation in Network Development’, http://www.regulateonline.org

COMPETITION AND INNOVATION

An important part of the discussions on innovation is the relationship between competition and innovation[1]. A competitive environment will in most cases promote innovations. In discussions on innovation, the argument that monopoly rents are important for innovation often surfaces. The argument is supposedly based on the writings of Joseph Schumpeter[2]. However, without going into a lengthy debate on the works of Schumpeter, the interpretation that monopoly is a prerequisite for innovation is surely not true, and in the telecom area the development of liberalization of the sector up until now has clearly demonstrated that competition has been beneficial to innovation. The opposite argument that incumbents do not innovate is, however, not true either. The monopoly period in telecom has actually witnessed many innovative developments. The point is that different kinds of operators innovate in different ways. An incumbent will most often seek to introduce innovations which complement its existing portfolio of services. This applies in a monopoly situation as well as in a situation where competitors have entered the market. New operators, on the other hand, will have a bigger incentive to introduce technologies potentially substituting for existing solutions. This is a theme lately popularized in the discussions on sustaining vs. disruptive innovations[3], where the new wireless technologies have been a case in point as presumably disruptive technologies. New wireless technologies will be introduced in the markets as means to compete with the incumbent operators. However, it also turns out that incumbents are often able to adopt new technologies even if they potentially substitute for some of the products in their existing portfolio. This happens, for instance, when under pressure from new operators offering new technology solutions. However, as a starting point there is no doubt that incumbents will have a tendency to implement new technology solutions which complement their product portfolios, while new operators will have a greater incentive to market new technologies that will substitute for the existing offers on the market. This also means that not only will competition enhance innovative market offerings; innovations will also increase competition. There is thus a two-way interrelationship between competition and innovation. ENDNOTES [1] H. Gruber: ‘Competition and Innovation: The Diffusion of Mobile Telecommunications in Central and Eastern Europe’, Information Economics and Policy 13, pp. 19-34, 2001. [2] A brief introduction to Joseph Schumpeter can be found on http://en.wikipedia.org/wiki/Joseph_Schumpeter. [3] Discussions often centre on the writings of Clayton Christensen and especially his first book on the theme: ‘The Innovator’s Dilemma: When New Technologies Cause Great Firms to Fail’, Harvard Business School Press, Boston, 1997.

NETWORKS AND SERVICES

Another important aspect of the issue of innovation in the communication area is the relationship between network innovation and application, service and content innovation. What comes first, acting as a driver for the development of the other network capacity expanding innovations, or applications, services and content? This has been a continuous debate in many countries. As illustrated in the figure, the answer is that there is a two-way relationship between network and application, service and content development. Network developments will lead to the implementation of new applications, services and content, and the demand for applications, services and content will act as a driver for network capacity expanding initiatives.

Relationships between demand for and supply of networks and applications

KEY POINTS AND RECOMMENDATIONS
s

s

s

s

Regulation should contribute to opening the sector for innovations and be seen as part of a broader national strategy for innovation of the communication area. The issue of innovation clearly illustrates that the relationship between technology change and regulation goes two ways. It is not solely a question of adapting regulation to technology changes, but regulatory changes and flexibility are necessary in the promotion of technology and organizational innovations in the area. There is also a two-way interrelationship between competition and innovation. This means that not only will competition enhance innovative market offerings; innovations will also increase competition. There is, furthermore, a two-way relationship between network and application, service and content development. Network developments will lead to the implementation of new applications, services and content, and the demand for applications, services and content will act as a driver for network capacity expanding initiatives.

3.4.3 Standardization
Standardization is essential in a network context, as compatibility standards are the technical prerequisites for interconnection and interoperability. Often a differentiation is made between de facto and de jure standardization, where de facto means that standards are established in the market by market players ‘as a matter of fact’, while de jure standardization takes place in official standardization organizations with some kind of public authority influence. The general development since liberalization started in the telecom area has been that de facto standardization has gained strength at the expense of de jure standardization[1]. This is related to the liberalization and privatization of the sector, but it is as much related to the internationalization of communications. Where communication systems formerly, to a large extent, were national systems and met at the borders where standardization of interfaces were necessary, communication systems are, at present, mostly international, and standards are negotiated between international manufacturers and operators, and less so between national public authorities[2]. ENDNOTES [1] See Richard Hawkins, Robin Mansell and Edward Steinmueller: ‘Liberalization and the Process and Implications of Standardization’, in Robin Mansell and Edward Steinmueller: ‘Mobilizing the Information Society’, Oxford University Press, 2002. [2] See, for instance, Philipp Genschell and Raymond Werle: ’From National Hierarchies to International Standardization: Modal Changes in the Governance of Telecommunications’, Journal of Public Policy 13(3), 1993, pp. 203-225.

APPLICATION AND SERVICE STANDARDS

With the increasing convergence of telecom and IT, there is an important job in securing interoperability between applications and services. This does not mean forcing private companies and citizens to use specific standards, but there is an important public task in promoting the use of standards and specifically open standards in and between public institutions, thus influencing the use of standards not only between public institutions and citizens and private enterprises but also among citizens and private enterprises. These kinds of standardization activities at national but also international level, specifically regarding applications and services, should not seek to restrain the use of new standards. The aim should be to facilitate the use of common standards, preferably open standards, and the interoperability between different applications and services. The standardization activities should thus be seen as a way of smoothing the internal workings and interrelationships between public institutions and as facilitation measures for market developments. In the practice note on IT standardization, work in Denmark on developing standards to be used by public institutions is presented. The necessity of developing such standards is generally accepted. However, an important discussion is whether such standards should be mandatory for public institutions or should remain as recommendations.

I N T R A - AND INTER-S T A N D A R D C O M P E T I T I O N

A long-standing discussion in standardization strategies concerns the choice between intra- and inter-standard-competition. With intrastandard competition, the different competing players on the market use the same standard and compete on other parameters than the ones related to differences in standards. With inter-standard competition, market players use different standards, which thus become part of the parameters on which competition is based. In recent years, the discussion on intra- and inter-standard competition has been heated in the field of mobile communications. With the development of the GSM standard, second generation (2G) mobile communications took off very quickly in European countries and successfully spread to the rest of the world. In the US, on the other hand, there has been a preference for inter-standard competition in mobile communications. Advantages and disadvantages of the two modes of competition have been debated. The main advantages of intra-standard competition are related to the immediate network effects of standardized communication systems and, for the users, the lowering of prices for terminals and services in a direct price competition between different providers. The primary advantage of interstandard competition is the greater push for innovativeness when different standards are competing and, consequently, less technology lock-in as well as the possibility, on a longer term basis, for the promotion of more cost efficient systems.

Apart from these more ideal reasons, there is of course also a great deal of company strategic interest and national interests involved. The choice of standards and the choice between intra- and inter-standard-competition is, therefore, very complex and there is no one right answer. It depends on the specific technology area and the specific situation. However, there is presently a preference for allowing/promoting more inter-standard competition, e.g. in relation to the uses of unlicensed radio frequencies and in relation to the trading of radio frequency licenses.

KEY POINTS AND RECOMMENDATIONS
s

s

s

s

s

Apart from the establishment of Internet specific organizations with responsibility for the development of the Internet including standards, the most outstanding feature of the new standardization environment is the establishment of a wide range of specialized forums for specific technologies with standardization as one of their major tasks. One of the results is that the influence of smaller and/or economically less strong nation states is weakened and that national public authorities in many countries only have very limited influence on standards for telecom networks. This, however, does not mean that standardization activities are no longer relevant for national regulatory authorities, but it does mean that the focus and direct influence of national authorities has shifted from the network aspects towards the application and service aspects. There is an important public task in promoting the use of standards and specifically open standards in and between public institutions, thus influencing the use of standards not only between public institutions and citizens and private enterprises but also among citizens and private enterprises. Another important area for public policy and regulation is the choice between intra- and inter-standard-competition. This choice is very complex and there is no one right answer. It depends on the specific technology area and the specific situation. However, there is presently a preference for allowing/promoting more inter-standard competition, e.g. in relation to the uses of unlicensed radio frequencies and in relation to the trading of radio frequency licenses. A further critical issue is the national character of type approval processes. This often creates very lengthy type approvals, which, in addition, may be subject to corruption. One of the ways to speed up type approval and to limit the problems of corruption is to implement a system of mutual recognition.

NETWORK STANDARDS

Standardization of a de jure character is one of the working areas of public authorities in the communication field and often involves regulators as the national representatives in international standardization organizations. Formerly, there were fewer international organizations in which standardization activities took place, with ITU (International Telecommunication Union) as the global organization for telecom standardization. It is important to notice that standardization activities have changed significantly since liberalization started in the telecom area. Different but parallel developments have affected the organized processes of standardization. In Europe, ETSI (European Telecommunications Standards Institute) was created in 1988 and contributed to a regionalization of de jure standardization. Beginning in the 1980s but speeding up in the 1990s, a large amount of so-called standardization forums or consortia have been created. At the same time, Internet developed, and around it a number of specific Internet organizations with powers to take decisions on Internet standards. In spite of the multifaceted character of the different standardization organization, differentiations can be made between, on the one hand, between de jure organizations (e.g. ITU), Internet organizations (first and foremost IETF, Internet Engineering Task Force), industry forums (for instance the WiMAX Forum), and professional organizations (with IEEE, Institute of Electrical and Electronics Engineers, as the most prominent one)[1]. On the other hand, a differentiation can be made between national (e.g. TTC, Telecommunication Technology Committee, in Japan), regional (for example ETSI in Europe), and international (e.g. the Internet organizations and most of the industry forums). Apart from the establishment of Internet specific organizations such as the Internet Society and IETF with responsibility for the development of the Internet including Internet standards, the most outstanding feature of the new standardization environment is the establishment of a wide range of specialized forums for specific technologies with standardization as one of their major tasks. As noted in the sub-section on internationalization, some of the implications of this development are that the relationships between the international and national aspects of standardization have changed and that public interests have experienced a decreasing influence. One of the results is that the influence of smaller and/or economically less strong nation states is weakened, and that national public authorities in many countries only have very limited influence on standards for telecom networks. This, however, does not mean that standardization activities are no longer relevant for national regulatory authorities. It is still important that national authorities support interconnection and interoperability and closely follow and seek to influence the standardization processes. Furthermore, national decisions regarding the implementation of network standards can be influenced via the licensing of operators using specific technologies should this be desirable. One aspect of this is the decision to promote either intra- or inter-standard competition. Furthermore, the focus and direct influence of national authorities has shifted from the network aspects towards the application and service aspects. Lastly, type approval deserves mentioning, as this has traditionally been and still is one of the important activities of many regulatory agencies. A critical issue here is the national character of approval processes. This often creates very lengthy type approvals, which in addition may be subject to corruption. One of the ways to speed up type approval and to limit the problems of corruption is to implement a system of mutual recognition. Such a system is implemented in the European context, so that the approval in one country automatically leads to approval in the other countries participating in the mutual arrangement. ENDNOTES [1] It should be noted that mixes of these organisation types also exist as in the case of, for instance, 3GPP (3G Partnership Project) in which an official standardisation organisation like ETSI is a participant.

3.4.4 Public-private partnership
Public-private partnership (PPP) involves the cooperation of two parties, the public sector and the private sector, and either refers to private sector entities carrying out assignments on behalf of public sector entities or to fulfill public policy goals or public sector activities helping private sector entities. Mostly, the term is used to denote that private sector entities, at different levels and scales, take care of activities traditionally performed by public sector entities. The Canadian Council for Public-Private Partnerships, for instance, states that PPPs are situated in the spectrum between direct provision by government departments and outright privatization[1]. The definition used by the Canadian Council for Public-Private Partnerships is that PPP is ‘a cooperative venture between the public and private sectors, built on the expertise of each partner, that best meets clearly defined public needs through the appropriate allocation of resources, risks and awards'[2]. ENDNOTES [1] See the website of the Canadian Council for Public-Private Partnerships (CCPPP): http://www.pppcouncil.ca/aboutPPP_definition.asp. [2] Ibid. Introductions to public-private partnership concepts can also be found on, e.g., http://www.ppp.gov.ie/ (the official Irish website for PPP) and http://ncppp.org/publications/books.shtml (a US website with publications on PPP).

KEY POINTS AND RECOMMENDATIONS
s

s

s

PPP is often part of a privatization discourse. However, it can easily be disentangled from this discourse and made part of a development of the potentials and problems in public-private cooperation in the build-up of infrastructures and the delivery of services. The issue of public-private partnership should be seen in light of the aim of developing a coherent policy for the establishment of national infrastructures and service provisions. The fact that communication services increasingly are subject to mechanisms of supply and demand renders it more relevant to examine the ways in which public sector initiatives can help build infrastructures, in the cases where private operators are unable to develop a market, for instance in geographically peripheral or poor areas. Instead of limiting the partnership arrangements to the public sector and private companies, there should be room for other individual and collective actors, including civil society organizations, non-governmental organizations resulting in Multi-Stakeholder Partnerships (MSPs).

PPP IN COMMUNICATIONS

The sectors mostly discussed in relation to PPP are, for instance, the infrastructural sectors of transportation and water supply, but health and other care sectors are also often involved. In the telecom area, privatization has been part of the broader development towards a liberalized market, and privatization continues to be an important tool in developing the market. In that sense, PPP is an important subject in the communication field, where many different combinations of mixed public-private ownership and different kinds of cooperation between public sector entities and private sector entities are used. However, communication services are decreasingly considered as public services in the sense where public authorities are seen as responsible for the availability of such services. More and more, communication services are considered as belonging to the large groups of goods and services which are subject to the supply and demand mechanisms of the market, even though there is still a large degree of public ownership around the world in the communication area, and even though universal access and service policies are expressions of public policy concerns in that area. Furthermore, traditional PPP arrangements where private businesses build, operate and later on transfer the facilities to publicly owned operators can also still be found in settings where communication operators are public enterprises. However, the fact that communication services increasingly are subject to mechanisms of supply and demand makes it more relevant to examine the ways in which public sector initiatives can help build infrastructures, where private operators are unable to develop a market, instead of solely discussing how private sector initiatives can assist in fulfilling public policy goals. This may apply in geographically peripheral or poor areas, i.e. in areas where private operators may not be able make a profitable business. It can also be that it is politically decided to build out infrastructures and service provisions at a faster rate than is considered commercially profitable. This actually happens in a number of economically developed countries in the deployment of broadband infrastructures. With the great emphasis that many economically developed countries put on broadband development, initiatives are taken to put public funding into the building up of broadband infrastructures. These infrastructures may eventually be taken over by private operators when they are able to run a profitable business. These cases, therefore, illustrate the opposite development of traditional PPP arrangements, where private businesses build infrastructures that later on are taken over by public sector entities. Furthermore, instead of limiting the partnership arrangements solely to the public sector and private companies, there should be room for other individual and collective actors, including civil society/non-governmental organizations. Consequently, it would be a multi-stakeholder arrangement, also called Multi-Stakeholder Partnership (MSP), which would not only center on the private/public dichotomy. This extension of the sphere of cooperation would be in line with the discussion in section IV.D.1 on the necessity of transcending the simplistic distinction between market failures and policy failures and to take a more holistic view on the system failures and the possible remedies including MSG arrangements.

THE PPP DISCOURSE

The Canadian Council for Public-Private Partnerships, furthermore, explains that while the terms PPP and privatization often are used interchangeably in the US, privatization in the Canadian context refers to but one of the two extremities in the PPP spectrum (see practice note for examples of different kinds of PPP arrangements). PPP is, however, often part of a privatization discourse. On the other hand, it can easily be taken out of this discourse and made part of a development of the potentials, but also problems, in public-private cooperation in the build-up of infrastructures and the delivery of services. Therefore, PPP should not only be seen in light of privatization developments but also in light of situations where private interests have difficulties in developing the markets on an entirely private basis. The issue of public-private partnership should thus be seen in light of the aforementioned aim of developing a coherent policy for the establishment of national infrastructures and service provisions. In such a context, policies of liberalization and privatization can also be viewed as instruments in reaching public policy goals and not only as negations of public policies as they often are considered to be.

3.4.5 Network and information security
Security in relation to technology mediated communication is a multidimensional term which encompasses a multitude of different issues. It reaches into the areas of privacy protection on the net and consumer protection in connection with e-commerce. It also reaches into the area of cyber crime, including protection against illegal and harmful content. Furthermore, it is connected with the protection of state institutions against information breaches, which conversely can be in conflict with privacy protection etc. In the present sub-section, however, only the issue of the network and information security, i.e. the technical aspects and their regulatory implications, is touched upon. Even this is an issue with many different dimensions.

IMPORTANCE OF SECURITY ISSUES

During the past few years, telecom regulators have increasingly been approached by market players expecting regulators to be involved in security issues. Indeed, network and information security has always been an important topic in telecom regulation, but the topic changes character with changing technologies. The security issues related to the present communication systems, including Internet, are more wide-reaching than security issues in the period where telephony was, if not the only, then by far the most dominant service. The reasons that present information and communication technologies raise a larger range of questions than formerly are, for example, that Internet is a more open environment than earlier communication systems; that wireless communication also raises new security problems; and that networks and services increasingly are international, which constitutes new dimensions of security. Societies today are far more dependent on technology mediated communication than before[1]. All processes of society depend on wellfunctioning communication infrastructures and services. This applies to business transactions where business-to-business interactions increasingly are conducted on networks and where business-to-consumer interactions also witness a growing number of network-based applications. It also applies to residential and non-commercial communication. In general, it applies to a growing part of all spheres of life, as societies and human activities are constructed in manners depending on mediated forms of communication. Seen in connection with the more open and thus more insecure communication environments, this is a significant challenge, where regulation has a role to play. ENDNOTES [1] Lately, a book by Robin Mansell and Brian Collins has been published on ’Trust and Crime in Information Societies’, Edward Elgar, Cheltenham, 2005, illustrating some of the security issues and dependencies f present day society on technology mediated communications. See Lawrence Lessig: ’Code and other Laws of Cyberspace’, Basic Books, 1999.

KEY POINTS AND RECOMMENDATIONS
s

Societies today are far more dependent on technology mediated communication than before. Furthermore, the security issues related to the present communication systems including Internet are more far-reaching than security issues in the period where telephony was by far the most dominant service. This combination poses a significant challenge, where regulation has a role to play. The important issue to be dealt with in connection with network and information security is to construct the appropriate combination of technology and legal measures taking the broader market and norms environment into consideration. Regulators are increasingly approached in security issues, but there are other organizations looking into these matters, for instance the national Computer Emergency Response Teams (CERTs). There can be many models for the organization of security work, partly or to a larger extent, involving communication regulators.

s

s

PROTECTION MEASURES

Regulations are not the only manner in which security is dealt with. Mostly, technology means are used, such as anti-virus software, smart card readers, encryption software, firewall software, and electronic signature software. These are some of the responses of individuals and individual organizations trying to protect their communication. However, they cannot stand on their own. Technical measures can only

partly deal with network and information security issues, as the functioning of technical means, to a large extent, relies on legal measures. The security situation is affected by a combination of the architectures of systems, the laws, the norms and the markets in which network and information security is in question. Most often, it is combinations of these different factors which determine the level of security, as in the case of procedures, laws and technologies, e.g. for authorization, authentication, integrity and non-repudiation, which are essential elements in a secure network and information system environment. With respect to the issue of the organizational responsibilities and competences in relation to network and information security, there are and will be differences among countries. Regulators are increasingly approached in security issues, as mentioned, but there are other organizations looking into these matters, for instance the national Computer Emergency Response Teams (CERTs). There can be many models for the organization of security work, partly or to a larger extent, involving communication regulators. However, with respect to the large number of other issues which are related to security, for instance privacy and consumer protection, they are in most cases best taken care of by other public authorities specialized in these fields. Many countries have agencies specialized in data and privacy protection, and consumer protection is best dealt with by specialized consumer agencies or organizations, even if commerce takes place on electronic platforms. The important issue to be dealt with in connection with network and information security is to construct the appropriate combination of technology and legal measures taking the broader environment of market and norms into consideration. This applies nationally as well as internationally, as communication is increasingly international, and security issues are therefore also international. This implies participating in or monitoring international negotiations and standardization initiatives on secure protocols and security systems.

3.5 Organizational aspects
The pages on organizational aspects focus on the scope and organization of telecom regulation in light of the analysis of the relations between different modes of public intervention. The implications of tendencies towards technology and market convergence are examined with respect to the organization of telecom regulation. Furthermore, the impacts of multi-sector market developments on the organization of regulation are discussed. Finally, the possibilities for developing countries to leapfrog regulatory developments implemented in economically richer countries are analyzed. There is a diverse range of organizational issues[1]. However, focus is on the issues relating to technology developments in ICTs with special emphasis on convergence developments. ENDNOTES [1] See the specific World Bank study and Toolkit Module 5,"Legal and Institutional Framework", where the broader range of organizational issues is dealt with.

3.5.1 Organizational dimensions
The different dimensions of organizational issues relating to technology developments in ICTs encompass the following: s Industry-specific vs. ICT convergence regulator: A telecom regulator is an example of an industry-specific regulator, while an ICT convergence regulator will cover a broader horizontal field, also including broadcasting infrastructures, for example. s Infrastructure vs. infrastructure and content regulator: An infrastructure regulator will regulate the infrastructural levels while leaving the content issues aside, whereas a combined infrastructure and content regulator in a vertical manner includes both. s Sector vs. multi-sector regulator: A sector regulator includes, for example, all ICT infrastructures, while a multi-sector regulator, for instance, covers different infrastructural sectors (utilities) such as telecoms, water and electricity. s Industry or sector-specific vs. general competition regulator: An industry or sector-specific regulator deals with industry and sectorspecific issues (provided there is an industry or sector-specific regulation), while a general competition regulator deals with general competition rules but may include more industry or sector-specific regulations under its domain. However, an industry or sectorspecific regulator and a general competition regulator do not necessarily exclude one another, but often co-exist. These are the issues examined in this sub-section. However, in order to complete the picture, the following organizational issues also have to be dealt with in a broader analysis of the potentially different organizational models to implement: s Local vs. national regulator: A local regulator can, for instance, be a state regulator as the public utility commissions in the US, while the national regulator in this case is the FCC (Federal Communications Commission). s Functional vs. industry or sector regulator: A functional regulator regulates specific functional issues, for instance universal service, as in the case of the Universal Service Agency of South African, while an industry or sector-specific regulator deals with the wider range of functional issues of an industry or sector. s Independent regulator vs. part of ministry: An independent regulator needs to be a separate organizational entity. However, the regulatory functions can also be performed by a section of a ministerial department as is, for instance, the case in Japan with telecom. The issue of independence is, however, a wide-ranging field with a broad spectrum of possible relations between the regulatory functions of regulators and the policy-making functions of ministries. Even though the kinds of regulatory agencies dealing with telecom are relatively diverse depending on national circumstances, the archetypal regulatory organization in the telecom area established as part of the regulatory reform process of the 1980s and 1990s is an industry-specific regulator. With the increasing convergence development technologically and on the markets, discussions have, however,

taken place in all national contexts as to whether an ICT convergence regulator, covering a broader array of ICT related issues including telecom, broadcasting and IT, would be more appropriate. The arguments in favor of a convergent regulator are strong, as different ICT infrastructures can be used for the conveyance of the same services (horizontal convergence), and as one common regulator for all ICT infrastructures, therefore, seems to be the obvious choice. The developments of convergence between telecom, IT, broadcasting and other media have also led to discussions on the convergence of infrastructure regulation and content regulation, as broadcasting and other media mainly are content industries. The argument could be that it would be unwise to separate the regulation of broadcast infrastructures and content, as these issues are interrelated to some extent. In countries with strict content regulation, licensing of network operators including content considerations is one of the ways to control content. The argument against is, however, strong, as infrastructure and content regulation are two very different topics. The horizontal convergence and the vertical divergence also speak against a common infrastructure and content regulator. With respect to multi-sector regulators, an argument in favor is that there can be commonalities in some of the objects of regulation. Utilities like electricity, gas, water or railway companies can, for instance, deploy optical fibers for telecom purposes in their ducts. However, the arguments relating to the issues of regulation are stronger. There are some commonalities in the economics of different infrastructures, which means that some of the regulatory issues are the same and that there can be economies of scope in the regulation of different infrastructures together. An important argument against this, on the other hand, is that these commonalities are too weak to justify a common regulator. The question of industry/sector-specific regulation and general competition regulation is also related to new technology and convergence developments in the ICT field, as horizontal convergence between different infrastructures entails the possibility of inter-infrastructure competition and, therefore, may increase competition allowing for a stronger reliance on general competition regulation and less industry or sector-specific regulation. The primary argument against abolishing an industry or sector-specific regulator and relying solely on a general competition regulator is the continuing existence of industry or sector-specific regulation, and the need to transpose the knowledge on this specific regulation into a competition regulator. This can be done by having a specific section with special ICT market competences in the competition authority. However, industry or sector specific regulation is, in a sense, alien to a competition regulator. However, to the extent that industry or sector specific regulators increasingly use regulatory tools from general competition regulation, it becomes more natural to transfer the regulation of ICT markets to a competition regulator.

3.5.2 Organizational issues
Before further discussing the different organizational options, it is important to realize that one should not go directly from the fact that there is technology and market convergence, and hence that regulations should take these developments into consideration to the conclusion that regulatory organizations also have to convergence. Technology and market convergence developments, indeed, constitute an important basis for organizational changes. However, organizational issues also have a relative autonomy and must also be dealt with on the basis of specific questions related to organizational development. The types of specific questions relating to organizational development examined in this sub-section are the following[1]: s Economies of regulation s Capture s Competition between agencies s Risk s One-stop-shopping s Transfer of regulatory practices With respect to all four dimensions discussed (industry vs. convergence regulator, etc.), the issue of economies of regulation has relevance, more specifically with regard to economies of scope and even economies of scale if the regulatory issues are identical across industries or sectors. Economies of scope and scale in regulation are relevant for regulatory organizations when the regulatory issues are similar as, for instance, in the case of access rules and rules for universal service and the calculation of retail prices, e.g. the RPI-x formula (retail price index minus the politically decided price reduction). An important argument, on the other hand, against the existence of scope and scale advantages in common regulatory organizations is the specificity of the regulatory assignments in the different fields, and the need for a sufficient level of special competences to understand and regulate the different markets. However, in a number of especially smaller developing countries the arguments in favor of scope and scale advantages are strong, as the human resources with the necessary specific qualifications are scarce. This calls for regulatory organizations of either a convergent, multi-sector, joint infrastructure and content, or joint competition and sector-specific character, or a combination of two or more of these. However, the general arguments for and against the different models still apply and must be weighed against the scope and scale advantages. Capture is another important issue to be considered. Regulatory organizations can either be captured by policy-making entities or by dominant market players, or by both. This is the reason why not only formal, but also real, independence is needed in order to ensure the highest degree of transparency and accountability possible. However, it cannot in any abstract manner be determined whether a single industry regulator, a convergent regulator, or any of the other regulatory models best ensure independence. In relation to policy authorities, it cannot in general be determined whether it is better to be related to one, two or more ministries. This very much depends on the specific circumstances. A third issue concerns the advantages and disadvantages of a competition between regulatory agencies. As in the case of market players, competition between different agencies may have beneficial effects hindering regulatory complacency. It cannot, of course, be a direct competitive relationship, as the different agencies cover different regulatory areas. However, a kind of yard-stick competition is possible with indirect competition between agencies. Again, it is difficult in general to determine whether the forces of creative competition

will be bigger than the forces of destructive competition[2]. This depends on the circumstances in the specific cases. Yet another factor is diversification of risk vs. better coordination[3]. This factor is closely related to the abovementioned point on competition between regulatory agencies. If there are different regulatory entities, there may be more experimentation in the regulatory practices and a greater diversification of risks. Furthermore, the specific regulatory organizations can be closer to the market realities in the different market segments. On the other hand, joint regulatory organizations will lead to better coordination and a higher level of certainty in decisions across market segments. One-stop-shopping is an often heard and good argument for joint regulatory organizations. Market players will not have to go from regulator to regulator, either not knowing where to approach the system of regulatory organizations or needing to approach more than one organization to settle a matter. However, there is a counter-argument, which is that joint regulators may seem more opaque to market players because of the diversity of the issues they deal with. The last point listed deals with the possible transfer of regulatory practices if different regulatory organizations are merged. This is a concern often aired, namely that if merging a regulatory area with little regulation with an area with more heavy regulation, the outcome may be that both areas will be regulated more stringently, even if the intention was to decrease the level of regulation. An example would be the transfer of existing telecoms regulations to the Internet area. Finally, two issues relating to the functional assignments of regulators and their independence deserve mentioning. Regarding functional assignments, a great number of issues are raised in connection with new ICT technologies and convergence developments such as network security, protection of privacy, protection of intellectual property right etc. The question is whether some of these issues should be part of the assignments of ICT convergence regulators. Again, a definite answer cannot be given, but to the extent that content issues and infrastructure issues are dealt with by different regulatory agencies, a content related question like intellectual property right should also be left out of the regulatory assignments of an ICT regulator. With respect to the independence issue, the only question to be mentioned here is that if ICT infrastructure regulation should be seen in the broader perspective of the general information technology and society policies of a country, it could be argued that close ties between the policy-making entities and the regulatory agency would be beneficial. However, close organizational ties are in no way necessary to ensure the overall goal of situating the ICT regulations in a broader policy context. This can be ensured at the policy level when developing the overall ICT related policies. ENDNOTES [1] More information on these issues can be found in, e.g., Jean-Jacques Laffont: ‘Regulation and Development’, Cambridge University Press, 2005 and Anders Henten, Rohan Samarajiva and William H. Melody: ‘Designing Next Generation Telecom Regulation: ICT Convergence or Multisector Utility?’, WDR, 2003, from which most of these points and the subsequent discussions are inspired. [2] These are the terms used by Jean-Jacques Laffont, ibid, page 222. [3] This point is directly inspired by Jean-Jacques Laffont, ibid, page 223.

3.5.3 Organizational leapfrogging
It applies to a number of developing countries that their regulatory organizations are weaker in terms of regulatory resources compared with the economically more developed countries. The implications are that the regulatory organizations will not be able to implement, for instance, time consuming and heavy market analyses (for instance the new EU market analysis scheme) and cost calculations (for example Long Run Incremental Cost calculations) as in a number of developed nations. Even though specific regulatory organizations in the telecom field in most cases do not have more than 10-15 years of experience in economically developed nations, regulatory organizations in developing nations have an even shorter history. They are at an early formative stage, but this also means that they can more easily be shaped to the convergence developments in the markets in an appropriate manner if there is the necessary political will. There is, in other words, a possibility for leapfrogging some of the developments that regulatory organization have gone through or are still going through in economically developed nations. The most important thing in this connection is to put emphasis on the facilitating aspects of the assignments of regulatory organizations. The prime purpose of the regulators should be to support and keep open as many and different possibilities for the spread and use of communication technologies. This implies to create a level playing field in terms of competitive possibilities; it also implies an innovative use of frequency resources, and it implies a support for the communication technology initiatives by, for instance, local and other communities. There should be more focus on proactively supporting technology take-up and use than on a reactive watchdog role. The activities related to universal access and service promotion thus have a central position in the range of regulatory assignments of regulators. This means that other regulatory areas concerning interconnection and limited resources, including frequencies, and the possible licensing practices of regulators should be seen in the light of expanding access to communication resources. This can be seen as a re-conceptualization of the relations between the different fields of regulatory assignments. Regulatory activities must be directed towards an increased take-up and use of communication technologies by means of the facilitation of the activities of as many commercial and non-commercial actors as possible[1]. ENDNOTES [1] This approach can, for instance, be found in a report by Seán Ó Siochrú and Bruce Girard, ‘Innovative Technologies and Community Ownership: A New Mode of ICT Access for the Rural Poor’, UNDP, April 2005. The approach is also the basis for the 3d cycle of research activities of the World Bank supported World Dialogue on Regulation for Network Economies on ‘Diversifying Participation in Network Development’, see http://www.regulateonline.org

3.5.4 Key points and recommendations

s

s

s

Convergence of telecommunication, IT, broadcasting and other media has led to many initiatives around the world in the direction of convergent regulators or regulators covering a larger range of infrastructure areas. The relationships between sector specific regulation and general competition regulation are also on the agenda. The important lesson to learn is that the processes of convergence have to be taken into consideration in the legal provisions and their implementation. However, the optimal specific organizational design of regulatory institutions depends very much on the specific national circumstances. Apart from the developments in technology and in the markets, there are also specific organizational issues to be taken into consideration when determining the best design of regulatory organizations such as economies of regulation, capture, competition between agencies, risk, one-stop-shopping, and transfer of regulatory practices. With respect to most of these organizational issues it cannot be determined in advance which organizational forms will perform best. Again, the specifics of different national contexts are important. However, it can be concluded that scope and scale economies of regulation are generally more related to the methods and form of regulation than the content and domain of regulation. Furthermore, the greatest concern raised in connection with merging regulatory organizations is the possible transfer of regulatory practices from areas with heavy regulations to areas with lighter regulations. The regulatory organizations in developing countries are often at an early formative stage. This means that these organizations have extremely limited inherited expertise and established practices, but it also means that they can more easily be shaped to the convergence developments in the markets in an appropriate manner if there is the necessary political will. There is, in other words, a possibility for leapfrogging some of the developments that regulatory organization have gone through or are still going through in economically developed countries.

3.6 Key Points and Recommendations
Based on the potentials in recent and forecast technology developments and the experiences with the present basic paradigm and its results in terms of market developments and technology spread, it is timely to consider what the contours of a new regulatory paradigm will look like.

3.6.1 Diversification of participation
s

The overall purpose of the new regulatory paradigm must be to open as many paths to network development as possible and resist attempts to control or restrain these developments unless there is clear evidence of harm to the public interest, a relatively rare circumstance. This implies promoting the diverse range of different business models and organizational forms. Telecom has traditionally been dominated by large companies benefiting from economies of scale and scope in single networks. These kinds of players will continue to play important roles in the area, but regulation should also seek to promote other organizational forms, including alternative operators, and more demand-led initiatives, such as end-user organized networks. New technologies now permit many diverse participants to be part of expanded networks where the full economies of scale and scope can still be realized. Among the different areas of regulation, universal service/access stands out as the central issue in present regulatory developments in developing countries. This does not mean that other regulatory issues are less important. Interconnection is of crucial importance and so is the management of limited resources. Rather, it means that the different regulatory issues should all be guided by the aim of increasing access. Technology leapfrogging involves the technical aspects of implementing new technologies in existing technological environments; it involves the economic, including financial aspects, the power and broader social interests related to existing and new technology systems, and involves a wide range of other socio-economic factors. The prospects of technology leapfrogging in the ICT area seem relatively good in relation to backbone infrastructures as well as access infrastructures and the services delivered, for instance the fast growth of mobile telephony in comparison with fixed line telephony in many developing nations. The overall recommendation is to take advantage of such potentials for technology leapfrogging by developing a regulatory framework that favors an open technology environment and technology change. Successful technology leapfrogging requires open markets and credible regulation. Without institutional change there can be no real progress with technological leapfrogging. Infrastructure and/or service competition has become an important issue. The background is that infrastructure- or facility-based competition is often seen as more sustainable than service-based competition, where alternative operators competing with the incumbents have to rely on the infrastructure elements provided by incumbents. The best policy, however, is to support all kinds of competitive strategies. Instead of viewing service- and infrastructure-based competition as alternative policy options, an optimal policy is to consider them as complements. The reason is that infrastructure and service markets partly are different market segments with different market players. It is important to implement the basic principle of cost based regulation, without adopting the complex and costly methods, such as LRIC, employed in some of the large developed countries. Knowledge of cost characteristics and benchmarks will facilitate the establishment of reasonable retail as well as wholesale prices, which will encourage competition and network investment. Furthermore, cost based regulation can also be used as a tool for promoting technology changes. Knowledge of the costs of different technology solutions can be used to shape regulations so that market players have incentives to take up new technology solutions. In developing countries, the problems of quality of service (QoS) are among the biggest issues in telecom. Often quality is poor on all

s

s

s

s

s

parameters. This applies to the PSTN technology and also applies to Internet technology. The areas where regulation can have a role to play are in connection with service level agreements (SLAs) in relation to interconnection between dominant market players and competitors. Furthermore, QoS for end-users can be secured by way of regulatory provisions. Users will often experience that the QoS does not correspond to the promises made, for instance with respect to transmission speed or the quality of VoIP services. There are, however, good reasons to hesitate with respect to strict regulatory interventions in the field of QoS on the Internet. One reason is the very dynamic character of the Internet and the continuously changing technology solutions used. Another reason is that the Internet-environment mostly is competitive, and that regulatory intervention, accordingly, should be light-handed.

3.6.2 Policy integration
s

s

s

s

s

s

Communications regulation must be seen in the broader context of policy measures related to the development of ICT infrastructures and services. However, the principle of independent regulation must be upheld in order to secure a reliable, stable and accountable regulation. Most regulatory provisions will be similar in the vast majority of countries. There will, however, be some differences, especially with respect to the emphasis on the different areas of regulation. The mix of policy initiatives and the more specific features of the different kinds of policy initiatives will vary. Regulation should contribute to opening the sector for innovations and be seen as part of a broader national strategy for innovation of the communication area. The issue of innovation clearly illustrates that the relationship between technology change and regulation goes two ways. It is not solely a question of adapting regulation to technology changes, but regulatory changes and flexibility are necessary in the promotion of technology and organizational innovations in the area. There is also a two-way interrelationship between competition and innovation. This means that not only will competition enhance innovative market offerings; innovations will also increase competition. There is, furthermore, a two-way relationship between network and application, service and content development. Network developments will lead to the implementation of new applications, services and content, and the demand for applications, services and content will act as a driver for network capacity expanding initiatives. Apart from the establishment of Internet specific organizations with responsibility for the development of the Internet including standards, the most outstanding feature of the new standardization environment is the establishment of a wide range of specialized forums for specific technologies with standardization as one of their major tasks. One of the results is that the influence of smaller and/or economically less strong nation states is weakened and that national public authorities in many countries only have very limited influence on standards for telecom networks. This, however, does not mean that standardization activities are no longer relevant for national regulatory authorities, but it does mean that the focus and direct influence of national authorities have shifted from the network aspects towards the application and service aspects. There is an important public task in promoting the use of standards and specifically open standards in and between public institutions, thus influencing the use of standards not only between public institutions and citizens and private enterprises but also among citizens and private enterprises. Another important area for public policy and regulation is the choice between intra- and inter-standard competition. This choice is very complex and there is no one right answer. It depends on the specific technology area and the specific situation. However, there is presently a preference for allowing/promoting more inter-standard competition, e.g. in relation to the uses of unlicensed radio frequencies and in relation to the trading of radio frequency licenses. A final critical issue is the national character of type approval processes. This often creates very lengthy type approvals, which, in addition, may be subject to corruption. One of the ways to speed up type approval and to limit the problems of corruption is to implement a system of mutual recognition. Public-private partnership (PPP) is often part of a privatization discourse. However, it can easily be disentangled from this discourse and made part of a development of the potentials and problems in public-private cooperation in the build-up of infrastructures and the delivery of services. The issue of public-private partnership should be seen in light of the aim of developing a coherent policy for the establishment of national infrastructures and service provisions. The fact that communication services increasingly are subject to mechanisms of supply and demand renders it more relevant to examine the ways in which public sector initiatives can help build infrastructures, in the cases where private operators are unable to develop a market, for instance in geographically peripheral or poor areas. Instead of limiting the partnership arrangements to the public sector and private companies, there should be room for other individual and collective actors, including civil society organizations, non-governmental organizations resulting in Multi-Stakeholder Partnerships (MSPs). Societies today are far more dependent on technology mediated communication than before. Furthermore, the security issues related to the present communication systems including Internet are more wide-reaching than security issues in the period where telephony was, by far, the dominating service. This combination poses a significant challenge, where regulation has a role to play. The important issue to be dealt with in connection with network and information security is to construct the appropriate combination of technology and legal measures, taking the broader market and norms environment into consideration. Regulators are increasingly approached in security issues, but there are other organizations looking into these matters, for instance the national Computer Emergency Response Teams (CERTs). There can be many models for the organization of security work, partly or to a larger extent, involving communication regulators.

3.6.3 Convergence and regulation
s

Focus in the discussions on the implications of technology and market convergence on the regulatory provisions has been partly

s

s

s

different in different parts of the world. In the economically developed nations, primarily the US, emphasis has been on the competition enhancing aspects of technology convergence. Furthermore, the growth of new services and new business areas is considered to be important. In developing countries, emphasis is mostly on new access possibilities with new technologies. In relation to technology divergence vertically, the proposal to separate infrastructure and services has been put forward in developed as well as developing countries, and there is no discussion that it would contribute to solving the issue of the transfer of market dominance in the infrastructure area to the service area. The less clear questions, however, are who the infrastructure provider(s) will be and what the incentive structure for extending the infrastructure will be. The recommendation is for developing countries to take advantage of the new access and competition potentials created by new technologies and to limit horizontal integration of companies to the extent that it hinders competition. Furthermore, vertical integration should in general not be constrained when not presenting a serious problem to competition. If analyses document that a sufficient degree of competition has been attained in a market segment, specific regulations can be abandoned and general competition rules will apply in this segment. The issue is not necessarily sector specific regulation or general competition regulation at the general level. There will be a combination of general competition rules and sector specific regulations according to the different situations in the different market segments. The concept of technology neutrality means that different technologies offering essentially similar services should be regulated in similar manners. However, exactly identical regulations may result in the advantage of one technology over another in the market. Technology neutral regulation can, consequently, include slightly differing regulations for different technology solutions in the same market segments. The technology neutrality concept has implications on most regulatory issues in telecom, including universal service/access, frequency management and competition.

3.6.4 International outlook
s

Equipment production has increasingly become international as have international backbone operations; access infrastructures, however, are mostly locally bound, while services and content can increasingly be delivered internationally. These developments have already affected and will continue to affect national regulations and the interrelationships between national and international regulations. The processes of liberalization and internationalization of communication have led to a decreasing influence of public institutions on international organizations and forums. To the extent this is considered a problem, working for the general improvement of the influence of public institutions in the different kinds of international organizations is important, and one of the ways forward is southsouth cooperation between regulators.

s

3.6.5 Organizational change
s

s

s

Convergence of telecommunication, IT, broadcasting and other media has led to many initiatives around the world in the direction of convergent regulators or regulators covering a larger range of infrastructure areas. The relationships between sector specific regulations and general competition regulation are also on the agenda. The important lesson to learn is that the processes of convergence have to be taken into consideration in the legal provisions. However, the optimal specific organizational design of regulatory institutions depends very much on the specific national circumstances. Apart from the developments in technology and in the markets, there are also specific organizations issues to be taken into consideration when determining the best design of regulatory organizations such as economies of regulation, capture, competition between agencies, risk, one-stop-shopping, and transfer of regulatory practices. With respect to most of these organizational issues it cannot be determined in advance which organizational forms will perform best. The specifics of different national contexts are important. However, it can be concluded that scope and scale economies of regulation are generally more related to the methods and form of regulation than the content and domain of regulation. Furthermore, the biggest concern mostly raised in connection with merging regulatory organizations is the possible transfer of regulatory practices from areas with heavy regulations to areas with lighter regulations. The regulatory organizations in developing countries are often at an early formative stage. This means that these organizations are relatively weak, but it also means that they can more easily be shaped to the convergence developments in the markets in an appropriate manner if there is the necessary political will. There is, in other words, a possibility for leapfrogging some of the developments that regulatory organizations have gone through or are still going through in economically developed nations.

4 Hot Topics
This section includes a list of Hot Topics, which are of particular interest for telecom regulators. Basic information on the technology as well as a description of regulatory issues and implications are described for each topic. In addition relevant links to other parts of the toolkit, Practice Notes and Reference Documents are provided.

4.1 NGN
Technology The concept Next Generation Networks (NGN) is used in two very distinctly different ways: 1) A broad concept encompassing the whole development of new network technologies, new access infrastructures and even new services, and 2) A focused concept of specific network architecture and related equipment, with one common IP core network deployed for the entire legacy, current and future access networks. ITU defines NGN as: “a packet-based network able to provide telecommunication services and able to make use of multiple broadband, QoS-enabled transport technologies and in which service-related functions are independent from underlying transport related technologies. It enables unfettered access for users to networks and to competing service providers and/or services of their choice. It supports generalized mobility which will allow consistent and ubiquitous provision of services to users”[1]. The first definition is a very broad descriptive reference covering all the current network technology trends. The second definition relates more precisely to the transition path towards a converged IP based core and access network. In the ITU definition there is a major emphasis on one of the main characteristics of IP platforms, namely the separation of network and service layers. The regulatory issues identified below arise from the implications of the specific technological changes in the NGN that will require a reassessment of how some traditional areas of regulation can best be implemented in the new environment. Regulatory issues
s

Interconnection: The deployment of NGN networks will create a demand for new types of interconnection products offering interconnecting packet switched services. These include interconnection between different types of networks and new types of unbundled network components. Separation of network facilities and service facilities will eventually allow for a complete separation of network and service interconnection. Measuring interconnection requirements in terms of minutes of traffic and leased lines will need to be replaced by measures of communication capacity, e.g. gigabits to facilitate efficiency. Regulators will need to foster this transition. Licensing: Effective competition among different infrastructures offering NGN requires a technology neutral licensing regime. Unified licensing will stimulate optimal use of technology options by operators and should be implemented at the earliest opportunity Universal service: Most universal service funding schemes will be rendered obsolete by the NGN. There will be a need to reformulate universal service obligations from being an obligation to offer a particular service (e.g. public voice) to an obligation to offer network access with a specified minimum capacity and quality. Subsidy schemes for financing universal access/service will have to be carefully designed so as not to distort market incentives to extend networks. Where subsidies are needed, competitive bidding for subsidies has been shown to be an effective way to maximize network development for minimum subsidy costs. Sector specific regulation: With networks carrying all kinds of services, both distributional (broadcast) and communicational, telecom sector specific regulation is becoming increasingly obsolete. In many countries, the regulation of telecom networks has therefore been extended to include all kinds of communication networks. NGN put this issue acutely on the agenda. Policy makers must establish a comprehensive convergence regulation framework for regulation to be effective. Infrastructure vs. service competition: The deployment of NGN involves an increasing separation between the core infrastructure and the services provided over it. There can be increased market entry possibilities for new service providers in all countries if regulators permit it, as the barriers to market entry are low. Infrastructure competition is more difficult given the large-scale investment requirements, but it can still be an effective force, stimulating network and services development in most countries. Regulation will affect the specific manner in which NGN is implemented and regulators should use this opportunity to foster expanded participation in the sector. QoS: As the NGN is based on Internet protocol, traditional QoS issues migrate from the former circuit-switched related issues to Internet-related QoS issues. It will be important for regulators to monitor QoS on the NGN with a view to ensuring that consumers understand the QoS associated with different service offerings, and consumer interests are protected.

s

s

s

s

s

ENDNOTES

[1] GLOBAL INFORMATION INFRASTRUCTURE, INTERNET PROTOCOL ASPECTS AND NEXT-GENERATION NETWORKS ITU-T Recommendation Y.2001 RELATED INFORMATION Interconnection (Re) Licensing Universal Service Next Generation Networks (NGN) Regulatory implications from NGN Sector specific and/or general competition regulation Infrastructure vs. service competition Quality of service

4.2 Wi-Fi
Technology The brand Wi-Fi is originally licensed by the Wi-Fi Alliance (http://www.wifi.org) to describe the underlying technology of Wireless Local Area Networks (WLAN) based on the IEEE 802.11 specifications. The IEEE 802.11 standard was published by the Institute of Electrical and Electronics Engineers (IEEE) in 1999. Several variations of the standard have been published since - the best known is IEEE 802.11(b). The Wi-Fi standard uses the unlicensed Industrial, Science and Medical (ISM) band. In the absence of licensing barriers, and because of the simplicity of the technology and its cost effectiveness, Wi-Fi networks have developed rapidly. Indoor coverage of 50 to 100 meters is normal and, depending on the standard, capacity bit rates of 11 to 54 Mbps (in some proprietary versions even more) are possible. In the point-to-point architecture, the Wi-Fi networks can have wider coverage (depending on the level of amplification in the antenna and the mast size, up to about 30km), which enables the creation of wider area network infrastructures. This is very important in developing countries, where no legacy telephony or cable networks exist. In developed countries, Wi-Fi is mainly used as a complementary local area infrastructure to enable flexible connectivity to different access networks for business as well as private users. In this case, Wi-Fi is the local extension of a broadband infrastructure. Furthermore, specific businesses offer Wi-Fi as public and commercial hot spots. Some cities are planning to offer Wi-Fi zones where there will be complete coverage within a specified area. Regulatory issues
s

Licensing: Licensing rules for Wi-Fi must be defined as simply as possible, so that they enable flexible use of Wi-Fi by network operators as well as end users. Universal service: Wi-Fi can contribute to the extension of networks to provide universal service by providing a niche component in networks provide local connections. Its application should be planned in association with WiMax technologies which provide longer transmission capabilities. Spectrum management: Wi-Fi requires access to unlicensed spectrum. This reduces licensing barriers, but the allocation of spectrum resources may still be necessary to avoid spectrum scarcity. Regulators need to ensure that the spectrum requirements for full exploitation of the benefits of Wi-Fi technology can be realized. Infrastructure competition: Wi-Fi offers the possibility for low-cost wireless access. It can provide network access to unserved areas, and can be used as a supplement or substitute for other types of access. Wi-Fi can therefore constitute a competing access infrastructure, but is most often used as an extension of broadband access and a complement to other kinds of wireless access. Business models: Most ISPs require that their broadband customers do not open their Wi-Fi access points to any other user, believing they will realize higher revenue with this restriction. The business model of these ISPs is to view Wi-Fi as a wireless extension of the DSL or cable modem connection. However, some companies or associations aim to make Wi-Fi access points available to other users so as to share the access infrastructure. This raises a consumer rights issue. In most circumstances consumers should be able to select whether or not they wish to share their Wi-Fi access points. Network and information security: There are network and security issues at stake to the extent that Wi-Fi access is not protected/blocked by encryption keys. Even if Wi-Fi access is protected or blocked by encryption, network and information security is relatively low. However, detailed regulation of QoS could potentially undermine the benefits of this low-cost access possibility. Regulators should ensure users are aware of the QoS provided, which will always be better than no service.

s

s

s

s

s

RELATED INFORMATION (Re) Licensing Universal Access Spectrum management Regulatory set-up and business models Network and information security

4.3 WiMAX
Technology WiMAX is the popular name for the IEEE802.16 standard, which is a high capacity and long range wireless broadband standard, and may become the international FWA (Fixed Wireless Access) standard. Coverage of 50km and capacity of around 70 Mbps is possible using this technology. It is, however, important to note that the capacity offered over long distances is only a fraction of the maximum capacity. WiMAX, as access technology, is more typically offered for distances of 5 to 10km. A key point is that 70 Mbps will only be achieved if the frequency bandwidth of 20 MHz is allocated and assigned by the spectrum management authorities. WiMAX is also being developed to become mobile (IEEE802.16e); mobile implementation is likely to be available after 2008. Regulatory issues
s

Licensing: In many countries, licensing rules must be updated in order to allow WiMAX operators to provide voice services and Internet access, and take full advantage of the potential of the technology. Universal service: WiMAX offers a unique opportunity for cost effective provision of rural access by new entrants, in many cases without a subsidy requirement. Where subsidy is required, WiMAX will frequently be the technology of choice, and barriers to its use should be minimized. Spectrum management: Adequate spectrum resources for WiMAX must be ensured. The refarming of spectrum resources may be desirable and in some circumstances necessary. The use of unlicensed bands lowers costs and barriers of entry, whereas the use of licensed bands enables better quality services. Spectrum licenses must be carefully designed to suit the particular needs of WiMAX operators if licensed bands are used. Infrastructure vs. service competition: WiMAX is one of the new technologies allowing for alternative operators to enter the market for broadband access provision. WiMAX may thus contribute to infrastructure competition. Regulators should consider how the new possibilities for broadband access and for enhancing competition in the broadband access market can best be developed using WiMax as well as other technologies. QoS: A significant issue regarding WiMAX is the actual quality of the network access offered, including the stability of the connection. As with Wi-Fi, QoS is likely to be lower than that provided with other technologies. Regulators should ensure users are aware of the QoS provided, recognizing that reduced quality service will always be better than no service.

s

s

s

s

RELATED INFORMATION (Re) Licensing Universal Access Spectrum Management Infrastructure vs. service competition Quality of service

4.4 VoIP
Technology Voice over Internet Protocol (VoIP) refers to the transmission of voice telephony over IP networks. VoIP can be delivered over the Internet and over the managed IP networks. The Internet is a ‘best effort’ medium, whereas in the managed IP networks it is possible to maintain

specified levels of QoS. Furthermore, gateways can be used to interconnect VoIP to the PSTN or mobile networks. This results in three main deployment scenarios for VoIP:
s

Pure IP: Here the VoIP services are implemented using Computers (or other IP terminals) as end devices. In some deployments, the regular PSTN terminal is connected to an IP converter. The communication takes place over the Internet or over managed IP networks. Skype is an example of VoIP over the Internet. IP => PSTN/mobile: In this scenario, there is a gateway between the IP and the PSTN or mobile networks, such that one can call from the IP terminal to legacy PSTN or mobile networks. Skype-out is an example of this service type. IP <=> PSTN/Mobile: In this scenario, it is possible to call from an IP terminal to the PSTN/mobile AND from a PSTN/mobile terminal to an IP terminal. Skype-in is an example of this service type.

s

s

Regulatory issues
s

Interconnection. Interconnection to the legacy PSTN networks is essential for the success of VoIP services. This interconnection is implemented by using gateways and contractual agreements between VoIP providers and PSTN operators. These interconnection arrangements should be monitored by regulators. If fair and non-discriminatory conditions for interconnection are not established in a timely fashion in the marketplace, regulators should intervene following traditional interconnection principles, as reasonable interconnection is a precondition for the successful development of VoIP. Universal Service. In rural areas the new wireless technologies will play an important role, where a combination of VoIP services and wireless infrastructures can enable a more efficient development of all communications services, including basic voice services. A particular problem related to use of VoIP for the provision of universal service is In-line powering of terminals. Traditional telephony service is designed with back-up power, and so it continues to work in case of electricity power failure. The current VoIP services/terminals are dependent on a functioning power supply. A requirement for in-line powering of terminals could put an enormous burden on the VoIP operators and slow development of service to un-served rural areas. Regulators should facilitate VoIP growth as a driver for network extension. For the longer term, the requirements for emergency communication standards and services can best be addressed by considering all the new technologies and services in the NGN, and most particularly VoIP and mobile. Numbering. VoIP services will co-exist with traditional public telephony for many years before the transition to all VoIP is completed. The rate of growth of VoIP will depend on its access to the national E.164 number plans. Any regulatory obstacles in accessing numbers can impede or slow down VoIP development. One model is to assign a new number series for VoIP services, although this would create confusion among consumers. The best model would be to assign numbers similar to the current PSTN numbers and to require number portability, so people are not forced to change their phone numbers when they want to change to a competitor offering VoIP services. Emergency call and positioning. The possibility to perform emergency calls and to route the call to the nearest authority (fire department, police, hospitals etc.) has been defined as a core element of Publicly Available Telephony Services in Europe[1]. Similar requirements are part of regulation in other countries. Location information is also more frequently becoming a requirement for both fixed and mobile telephony. In VoIP it is possible to maintain positioning and routing information for emergency calls. However this requires use of VoIP services from fixed locations. However, one of the promising characteristics of VoIP services is nomadic use. In nomadic use at the current level of technological development, the position information cannot be connected to the emergency call. This is a challenge both to the market players and to the regulatory framework. Regulators should take a lead in facilitating the resolution of these important emergency issues in future networks. QoS. With POTS, there are detailed recommendations on QoS from the ITU and in many national regulations. In managed VoIP services it is possible to provide measurable QoS, but this is more difficult in best effort services. Another important issue is the willingness of facility based operators to offer access to QoS provision to non-facility based operators. For example, a major debate in Europe and other regions is the lack of QoS provision in the wholesale Bit stream access products offered by the PSTN incumbents. Regulators will need to take the lead in ensuring consumer protection with respect to QoS. Interoperability & Standardization. Different technical standards are used to establish VoIP services. It is important to establish interoperability between these standards. The interoperability may be implemented at the technology or market levels. Also new numbering schemes like global Dialing System and ENUM may require standardization and interoperability. These developments should be monitored by regulators, and if the market players do not find adequate solutions for interoperability, regulatory measures may be necessary. Security and consumer protection. In regular telephony services the security and consumer protection standards have been defined and are generally found adequate.

s

s

s

s

s

s

Endnotes [1] For a more detailed outline of the European discussion see e.g. Communication staff working document on the treatment of Voice over Internet Protocol (VoIP) under the EU regulatory framework, Brussels, June 2004
RELATED MATERIALS

For information about the competition and price regulation aspects of VoIP, please see Module 2, section 4.2, "About VoIP" Interconnection Universal Access Numbering Standardization Quality of service Network and information security

4.5 IPTV
This section defines IPTV and examines its classification in various jurisdictions. It also addresses the issue of whether regulations and obligations typically related to broadcasters should be extended to IPTV. In addition, this section addresses the manner in which different countries are licensing IPTV as well as the impact on IPTV of network unbundling.

4.5.1 What is IPTV?
Distinguishing IPTV from other video offerings The term IPTV can cause some confusion. In narrow terms, IPTV is defined as the provision of video services (for example, live television channels, near video-on-demand (VOD) or pay-per-view) through an IP platform. However, some define IPTV services to encompass all the possible functionalities that can be provided over an IP platform. For example, some equate IPTV services with multimedia services, a category that can include television, video, audio, text, graphics, and data.[1] This encompasses not only one-way video broadcasting services but also ancillary interactive video and data services, such as VOD, web browsing, advanced email, and messaging services. The interactive services associated with IPTV allow the viewer to determine what and when to watch, and also allow the user to “teleshop” or order movie tickets. IPTV providers now commonly include in their commercial packages a personal video recorder (PVR) through a hard disk in the set-top-box (STB) or on the network, allowing “timeshifted” or “catch-up” viewing of TV broadcasts.[2] With an IP-based managed network, the service provider is able to offer a high quality of service (QoS) level and high “Quality of Experience” (QoE), as well as security, interactivity and reliability. IPTV providers are signing content agreements and developing innovative applications in order to compete with cable and satellite television. This includes striking deals for special viewing packages such as sports. Several IPTV providers have also launched High Definition (HD) television services. In Hong Kong, China PCCW recently introduced stock trading on its “Now” IPTV service. In France, Iliad’s “TV Perso Freebox” lets subscribers post their own videos for view by others. IPTV can be confused with Internet video or Internet TV, but those services are quite different. Internet video and Internet TV are both offered over the public Internet. Internet video is an unmanaged service that offers the streaming of video through the public Internet. Internet video companies include user-generated video websites like YouTube or Metacafe where users can upload and view others’ videos. Today, these services tend to lack a QoS standard and are without any real control over production quality.[3] Internet TV companies, like Joost, Babelgum, and Zattoo, tend to operate on peer-to-peer networks rather than on managed networks, and they typically offer free, ad-based services. However, their offerings are similar or identical to IPTV in several key areas. First, like IPTV, Internet TV provides professionally produced and copyright-protected video. Internet TV companies also tend to use MPEG 4, the same encoding technology used by IPTV providers, for high video quality and offer near-TV quality picture resolution. While IPTV allows subscribers to more easily switch from television to computer mode, users are increasingly able to view all kinds of video on their television sets with Internet TV.[4] For Internet TV providers like Joost that offer VOD, users can rewind and fast-forward videos, much like IPTV users that rewind and fast forward with PVR. However, Internet TV providers that stream live television, such as Zattoo, do not yet have this capability. Although limited in their service areas, both the U.S.-based Joost and European-based Zattoo have negotiated digital rights management (DRM) agreements, requiring operators to prevent end users from copying or converting copyrighted materials. DRM deals are considered a necessary component of offering IPTV.
A comparison of IPTV, Internet TV, and Internet Video IPTV Examples of U-verse (AT&T) Internet TV Joost Internet Video Youtube

Operators

Opzioni TV (Fastweb) Orange TV (France Telecom) Imagenio (Telefonica) Now TV (PCCW)

Zattoo Babelgum

Metacafe

Users Services

Subscribers only; closed network Live TV VOD Interactive services

Free, ad-based service VOD and/or live TV and Internet in multitask environment

Free, ad-based service Video clips only

Network Video Production Video Quality

IP-based platform; Managed network Professional video only

Public Internet; Peer-to-Peer Professional video only

Public Internet Amateur/user-generated video only Unmanaged QoS Low, but improving PC QCIF/CIF No copyright protections Fully accessible

Managed QoS MPEG 2 to MPEG 4, MSVCI

Managed QoS High – MPEG 4 PC Near full TV display Content is protected through DRM Trial stages only

Receiver device Resolution Copyright Status of roll-out

STB with TV or PC Full TV display Content is protected through DRM Deployed in limited geographic areas in various countries

Source: Based on IPTV – Market Regulatory Trends and Policy Option in Europe, Background Material, ITU-T IPTV Global Technical Workshop: Driving The Future Of IPTV, Document: IPTV/01, 1 November 2006, Seoul, 12-13 October 2006, at p. 7 and Telecommunications Management Group, Inc. research

Technical aspects of IPTV An IPTV operation has four components: the content source, the core network, the access network, and the end user (see figure below). [4] The content source is the video provider that owns or is licensed to sell live television programming, VOD, or other downloaded content. Live television is typically received via satellite or through fiber networks, while VOD content is stored by the network operator. Content passes through an encoder, or headend, which prepares the content for transmission on the network. The core network encodes the video streams using MPEG-2, although the use of MPEG-4 (H.264 AVC[5], Windows Media VC-1) is on the rise. Once encoded, the content is encapsulated into IP packets, and is then ready for delivery to subscribers.
An Operational Diagram of IPTV

Source: Telecommunications Management Group, Inc.

Live television is delivered via multicast, which allows many end users to receive content from one packet through efficient use of the IP network. Channels are essentially IP multicast group addresses that subscribers request to join. Unlike a cable system or an over-the-air television that “tunes” to a channel, the IPTV set-top box (STB) acts only as an IP receiver. The STB changes channels by using the protocol to join a new multicast group. When the local switch office obtains the channel change request, it confirms that the subscriber is authorized to view the content and adds the user to the channel distribution list. Therefore, only signals being watched are sent from the

local office, through a digital subscriber line access multiplexer (DSLAM), if necessary, and finally to the user. Rather than a “one-to-many” transmission like multicast, VOD is unicast, or “one-to-one.” When an end user requests a VOD product, the servers pull pre-compressed video streams and transmit them as IP packets. Typically, the local switch office uses a VOD server to stream from the server to a particular subscriber’s location. The stream is generally controlled by real time streaming protocol (RTSP), which allows the user to play, pause, and stop the program. If the video stream is delivered over a copper local loop, the IPTV provider must use DSLAM equipment to deliver IP packets to the subscriber after the content is encoded. DSLAMs are located either along the core network or access network. At the customer premises, the STB allows subscribers to select the content they want to watch and provides user control over functionalities such as rewind, fast-forward, and pause over non-live programs. The two-way functionality of IPTV services not only allows subscribers to choose their services with the press of a button, it also offers interactive capabilities, which allow a user to easily manage multimedia sessions and personalize preferences. Who provides IPTV services? The main providers of IPTV services tend to be telecommunication service providers, but cable and satellite operators are also starting to get into the game. There are two types of telecommunication providers offering IPTV: incumbent operators and newcomers. The former category includes operators such as France Telecom, PCCW in Hong Kong, China, Telefónica in Spain and AT&T and Verizon in the United States. Incumbents are offering the service over their copper ADSL (asymmetric digital subscriber line) networks or increasingly over fiber access networks. Newcomers include Iliad in France, Fastweb in Italy and Hanaro in the Republic of Korea. These new market players have often been successful by offering IPTV as part of a basic ADSL subscription. IPTV service typically offers from 40 up to 300 TV channels, as well as VOD, High Definition (HD), and PVRs. Coverage and deployment vary widely. For instance, AT&T currently only offers their “U-verse” service in select cities in a dozen U.S. states. However some deployments are having an impact. In Hong Kong, China, PCCW’s “Now” IPTV service had 560,000 subscribers in June 2007 and accounted for almost 40 per cent of all television subscribers. IPTV has also been successful in Italy and France, where conventional subscription television penetration is not as developed as in other Western European nations. Equipment manufacturers are increasingly introducing an element of IP into their STBs.[6] It is estimated that by 2010, about half of the 30 million IPTV STBs deployed in the world will be hybrid (IPTV combined with some form of digital cable, terrestrial, or satellite front end). In addition, some established subscription TV operators are combining IPTV technology and services with their existing channel packages to offer enhanced functionality, such as on-demand content.[7] For example, Premiere in Germany is planning to offer a combined satellite and IPTV service in partnership with Deutsche Telekom (DT), allowing access to DT’s IPTV offering and to its own satellite subscription service. In Japan, subscription television satellite operator Sky PerfecTV has rolled out an IPTV offering. In the UK, BT has launched a combined IPTV/DTT service (“BT Vision”) that provides traditional broadcast-based channels over DTT, alongside additional content over an IP connection. Market potential According to the DSL Forum, there were some 8.2 million IPTV subscribers worldwide in June 2007.[8] This marked an increase of 127 per cent from a year earlier. Europe leads in deployment, accounting for more than half of the world’s IPTV subscribers (see figure below). Indeed, measuring IPTV penetration as a percentage of total pay TV subscribers, four out of the top five countries are European. But IPTV has been most successful in Hong Kong, China, where it accounts for about two out of every five pay television subscriptions. Several forecasts of IPTV evolution predict subscribership ranging from 41 million to 73 million by 2011.[9] Given that IPTV is at an initial stage of market development, however, these figures should be treated with caution. Some jurisdictions such as France and Hong Kong, China have been extremely successful with IPTV, and if these experiences can be replicated elsewhere, then the figures could be much higher. Also, most major deployments have so far been limited to developed economies. The potential for IPTV in developing nations could be significant in markets lacking traditional subscription television access through cable or satellite systems. However, the attraction of IPTV has to be balanced against the high investment costs of installing broadband infrastructure. IPTV presents an opportunity for traditional telecommunication providers to offer triple play services. In addition, unlike new entrants, most major operators launching IPTV operations have the financial resources available to upgrade their networks and an existing customer base for marketing purposes. But they do face some bottlenecks that impact IPTV strategy. First, coverage is far from ubiquitous. In order to receive IPTV, high-speed broadband access is required. While many operators have launched broadband network rollouts, they have not reached nationwide coverage in most markets. Meanwhile, even some broadband systems operate at speeds too slow to support IPTV, which requires a downstream connection of at least 4 megabits per second (Mbit/s). A second issue is that some telecommunication operators already provide television service through cable or satellite ownership or partnership agreements. So they are reluctant to “cannibalize” those services.
Distribution of IPTV subscribers, by region; Leading IPTV countries, by percentage of total subscribers, June 2007

Source: Adapted from DSL Forum and regulator and operator reports.

ENDNOTES [1] See ITU-T IPTV Focus Group definition. [2] ITU IPTV Global Technical Workshop, Driving The Future Of IPTV 13 (2006), available at www.itu.int/osg/spu/stn/digitalcontent/4.9.pdf. [3] Mark Rooney, Pace Micro Technology: IPTV White Paper 3 (2006), available at http://www.iptvnews.com/__data/assets/word_doc/0016/51145/2006_IPTV_White_Paper_Pace.doc. [4] Wei Li, Hong Liu, and Yiyan Wu, Communications Research Centre Canada, IEEE: Introduction to IPTV, available at www.ieee.org/organizations/society/bt/iptv1.pdf. [5] H.264 generally provides a 40 percent saving in bandwidth over MPEG-2 encoded content, allowing IPTV providers to offer high definition (HD) television to the home. [6] Rooney, supra note 3, at 4. [7] Off. of Comms. (Ofcom), The International Communications Market 2006 99 (2006), available at www.ofcom.org.uk/research/cm/icmr06/icmr.pdf. [8] Press Release, DSL Forum, “IPTV Deployments More Than Double in a Year as Broadband Continues to Achieve Strong” (Oct. 8, 2007) www.dslforum.org/dslnews/pdfs/pr_bwfeurope100807.pdf. [9] Martin Olausson, Strategy Analytics, Strategy Analytics Forecast 41 Million IPTV Subscribers by 2011, available at www.strategyanalytics.net/default.aspx?mod=ReportAbstractViewer&a0=3259, Press Release, Gartner, Gartner Projects 49 Million by 2010, available at www.gartner.com/it/page.jsp?id=496291, and Press Release, Multimedia Research Group, MRG 73 Million by 2011, available at www.mrgco.com/press_releases.html#GF1007.

4.5.2 Classification of IPTV
Countries are taking various approaches to classifying IPTV. These approaches range from simply not addressing classification − instead focusing on competitive market entry into the video market − to denoting IPTV and its related functionalities as a regulated broadcasting service. Some countries are also developing a broad middle ground, where some services offered over IPTV platforms are considered broadcasting while others, such as VOD, are not. In the United States, for example, IPTV has yet to be classified. The Federal Communications Commission (FCC) initiated a proceeding on IPenabled services in 2004, in which it made certain determinations about VoIP and other IP services but did not decide anything on IPTV services.[1] This has not precluded FCC, however, from addressing certain perceived barriers to the deployment of IPTV services. FCC has:
s

Declined to require incumbent local exchange carriers to provide unbundled access to their hybrid or FTTH loops for the provision of broadband services; Relaxed the process for issuing cable franchises (a licensing process) to facilitate entry into the video market; and Found that clauses granting cable providers exclusive access for the provision of video services to multiple dwelling units and other real estate developments harm competition and broadband deployment and were therefore illegal.[2]

s s

On the other end of the spectrum, some countries have adopted a technology-neutral approach to classifying IPTV. For example, the Canadian Radio-television and Telecommunications Commission (CRTC), considers IPTV one of the broadcast distribution technologies available for television programming.[3] Services offered over this platform, including VOD, are deemed to be broadcasting services. IPTV

providers fall within the category of broadcasting distribution companies, and are licensed accordingly. Another approach is taken by the Republic of Korea, Singapore, and Pakistan, where IPTV has not only been specifically classified as a broadcasting service, but new categories of broadcasting licences have been established. In Singapore, for example, “broadcasting” includes the IP transmission of any television programming − either full scheduled channels or VOD − to households via a broadband connection.[4] The Republic of Korea has enacted a new law that classifies IPTV as an “Internet multimedia broadcasting” service. This is defined as a “type of broadcasting whereby various types of content, including real-time broadcasting programmes, are provided to users through television sets by way of Internet protocol that allows interactivity using fixed-line telecommunications facilities.”[5] Some jurisdictions are basing their regulatory classification of IPTV services on the degree of interactivity they allow. On this basis, many countries are distinguishing between broadcast and VoD elements of IPTV. For instance, the EU countries and New Zealand differentiate between transmission that is linear (transmitted at a scheduled time) or non-linear (content that is selected by the user and viewed when the viewer wishes). Linear programming is generally subject to broadcasting and content regulations. Non-linear content may be exempt from those regulations, as in New Zealand, or subject to some of them but not others, as in the EU. ENDNOTES [1] In the Matter of IP-Enabled Services, 19 FCC Rcd 4863, 4878 (2004). [2] In the Matter of Review of the Section 251 Unbundling Obligations of Incumbent Local Exchange Carriers; Implementation of the Local Competition Provisions of the Telecommunications Act of 1996; Deployment of Wireline Services Offering Advanced Telecommunications Capability, 18 FCC Rcd 16978, 17103-04, 17149, paras 200, 288 (2003). These rules were upheld by the D.C. Circuit Court of Appeals on Mar. 2, 2004, in United States Telecom Ass’n v. FCC, 359 F.3d 554, 564-76 (2004) (“USTA II”); In the Matter of Implementation of Section 621(a)(1) of the Cable Communications Policy Act of 1984 as amended by the Cable Television Consumer Protection and Competition Act of 1992, 20 FCC Rcd 18581 (2006); In the Matter of Exclusive Service Contracts for Provision of Video Services in Multiple Dwelling Units and Other Real Estate Developments, 22 FCC Rcd 20235 (2007). [3] Can. Radio-television and Telecomms. Comm’n (CRTC), The Future Environment Facing the Canadian Broadcasting System: A Report Prepared Pursuant to Section 15 of the Broadcasting Act (2006), available at www.crtc.gc.ca/eng/publications/reports/broadcast/rep061214.htm?Print=True. (“There are four broadcast distribution technologies available for television programming: conventional over-the-air (OTA) transmission, cable distribution, DTH satellite distribution and Internet Protocol Television (IPTV)”). [4] See Media Dev. Auth. (MDA), Licence Framework for Broadcasting IPTV Services, available at www.mda.gov.sg/wms.www/devnpolicies.aspx?sid=88. [5] Internet Multimedia Broadcasting Business Act (Republic of Korea), approved by National Assembly in 2007, and implemented in April 2008.

4.5.3 Applying content regulation to IPTV providers
Many countries have only recently started to grapple with whether IPTV providers should comply with content regulations. In the EU, for example, the European Commission recently decided to amend the Television without Frontiers Directive (“TWF”), last revised in 1997, to address the new scope of audiovisual services. In December 2007, the EC approved the Audiovisual Media Service Directive (“AVMS Directive”), which will apply to all “audiovisual media services” (that is, services providing moving images with or without sound). This includes traditional television broadcasts (termed “linear” audiovisual media services) as well as on-demand services (termed “nonlinear”). Under the AVMS Directive, both of these services are subject to a baseline set of rules (for example, rules protecting minors and promoting European productions), but traditional television services will be subject to certain additional obligations. IPTV providers will be subject to the baseline rules when they offer television broadcasting and on-demand services. But they will not have to adhere to the content regulations if they are merely retransmitting television or on-demand programming without altering the content. In certain jurisdictions, regulators have decided that IPTV providers should be subject to the same content regulation imposed on subscription television providers. For example, IPTV providers in Singapore are subject to the programming code imposed on subscription television providers. In addition, must-carry obligations apply to fixed IPTV operators in numerous EU countries, such as Belgium (in the French-speaking community), France, Sweden, and the UK (although in practice, the parties have negotiated commercial arrangements). The U.S. Federal Communications Commission, however, has yet to rule on what the regulatory status of IPTV will be and whether the must-carry rules will apply to such services. As in the EU, the Telecom Regulatory Authority of India (TRAI) has recommended not subjecting telecommunication providers offering IPTV services to content regulation for unaltered content obtained from television broadcasters.[1] TRAI did recommend, however, that IPTV providers be required to comply with the programme and advertisement code under the Cable Television Network (Regulation) Act 1995 if they obtain “broadcasting content, Internet-related content or VOD including movie related content.” Examples of this would include musicon-demand, games, or locally developed content. In addition, telecommunication service providers offering IPTV services may only show

news channels that have been approved by the Ministry of Information and Broadcasting. ENDNOTES [1] Telecomm. Reg. Auth. of India (TRAI), Recommendations on Provision of IPTV Services, at 25-26 (Nov. 28 2007).

4.5.4 Licensing IPTV providers
Regulators are taking different approaches to licensing IPTV providers. Sometimes the licensing requirement is based on the service being offered rather than the particular platform used to offer it – in other words, a technology-neutral approach. To the extent an IPTV provider is offering live television, for example, it may subject to the same licensing requirements imposed on television broadcasters. Europeans are following a technology-neutral approach that considers any television service − provided over any platform (cable, satellite Internet, ASDL, or mobile network) − a broadcasting service. In France, for example, an IPTV operator must submit a declaration to the Conseil Superieur de l’Audiovisuel (CSA), although small operators with annual programming budgets of less than EUR 150,000 are exempted from this mandate.[1] Similarly, Canada requires any television service, including VOD, provided over a managed IP network to have a “Broadcast Distribution Undertakings” licence. In Europe, however, VOD is not considered a television service due to its interactivity.
Pakistan’s IPTV Channel Distribution Service Licence Issuing Authority Primary Requirements Pakistan Electronic Media Regulatory Authority Must hold Fixed Local Loop Licence from PTA for each region to be served Company must be incorporated in Pakistan Majority of shares cannot be owned or controlled by foreign nationals or any entity whose management control is vested in foreign nationals Coverage Fee Structure Per Zone: Two Categories, A and B (Category A – 4 zones including Karachi and Islamabad; Category B – 10 zones) Application Processing Fee: Rs 20,000 (approx. USD 320) Category A licences: Rs 1,000,000 (approx. USD 16,000) per zone Category B licences: Rs 500,000 (approx. USD 8000) per zone Security Deposit Licensing Term Annual Renewal Fee 10 per cent of licence fee (refundable after 1 year after satisfactory operation) 5 years 30 per cent of the licence fee plus 5 per cent of the annual gross revenues

Source: Pakistan Electronic Media Regulatory Authority, Guidelines for Submission of Statement of Qualifications for IPTV Channel Distribution Service Licence

In countries such as the Republic of Korea, Pakistan, and Singapore, the broadcasting authority has developed new licences to IPTV services. Under the Republic of Korea’s new Internet Multimedia Broadcasting Business Act, IPTV providers must get an “Internet multimedia broadcasting” licence from the Minister of Information and Communications. In Pakistan, IPTV providers must not only obtain an “IPTV Channel Distribution Service” licence from the Electronic Media Regulatory Authority to provide service, they must also hold a Fixed Local Loop licence for the same areas (see figure above). In 2007, Singapore’s Media Development Authority (MDA) developed a technology-neutral licensing framework to facilitate the introduction of new media services like IPTV. All media service operators seeking to offer IPTV services (or any form of subscription television services) now require a licence from MDA. That agency defines IPTV as the transmission of television programming, in the form of either full scheduled channels or video-on-demand, to households via a broadband, IP connection. Using the IPTV network, service providers can also offer rich interactivity and services such as television commerce, VoIP, video conferencing, and gaming (see figure below). Singapore’s new framework, however, applied ownership restrictions under the Broadcasting Act to nationwide licensees with more than 100,000 subscribers − but not to niche licensees with fewer than 100,000 subscribers. The disadvantage of this two-tier distinction is that as licensees grow their subscriber base, they may find themselves forced to divest foreign ownership, since the Broadcasting Act prohibits a foreign entity from holding more than a 49 per cent interest. MDA is proposing a similar two-tier ownership approach for mobile TV providers.
Singapore’s Two-Tier Licensing Framework for IPTV Services Niche Subscription TV Licence Number of subscribers Licence term 100,000 subscribers 5 years Nationwide Subscription TV Licence Unlimited number of subscribers 10 years

Licence fee

2.5% of total revenue; minimum annual licence fee of SGD 5,000 will be applicable during duration of licence.

2.5 per cent of total revenue; minimum annual licence fee of SGD 50,000 per annum will be applicable throughout.

New service licencees may enjoy a concessionary rate of 0.5 New service licensees may enjoy a concessionary rate of 0.5 per cent of total revenue or SGD 5,000, whichever is the per cent of total revenue or SGD 50,000, whichever is the higher amount during first three years of licence term higher amount during first three years of licence term. Performance bond Ownership Must carry Advertising revenue Advertising time limit Content guidelines SGD 50,000, in the form of either banker’s guarantee or cash. No ownership conditions Not applicable No cap on advertising revenue SGD 200,000, in the form of either banker’s guarantee or cash. Subject to ownership restrictions set forth in Part X of Broadcasting Actb Must carry obligations for enabling access to local free-to-air channels are applicable Advertising revenue not to exceed 25 per cent of total revenue

14 minutes per hour advertising time limit applies for channels with scheduled programming (not applicable for VOD content and interactive advertising services. Subscription TV programme code applies if scheduled programmes are offered. VOD programme code applies if ondemand programmes are offered.

Source: Media Development Authority of Singapore, www.mda.gov.sg/wms.www/devnpolicies.aspx?sid=88#3

Hong Kong, China has not established a special licence category for IPTV providers. Instead, it regulates IPTV providers in the same manner as a subscription television provider, requiring them to obtain a domestic subscription television programme licence. However, as in Pakistan, such licences can only be obtained if the operator already holds a fixed network licence. India’s regulator, TRAI, has recommended that IPTV telecommunication providers be regulated under the terms of their telecommunication licence, and that cable operators be regulated under the terms of the Cable Television Network (Regulation) Act, 1995. TRAI has indicated that IPTV services provided by telecommunication operators are not the same as a cable service. This judgment was based on the technical aspects of the services and the manner in which the content is delivered to the user (cable channels are pushed to the user whereas IPTV channels are pulled by the user). TRAI recommended that telecommunication service providers holding a “Unified Access Service” or “Cellular Mobile Telephony Service” (CMTS) licence be allowed to provide IPTV services under their licences, without any further approval.[2] ISPs with a net worth of more than a billion Rupees (approximately USD 25 million) would be allowed to provide IPTV services after obtaining permission from the regulator. Similarly, cable television operators could offer IPTV services under their current authorizations. ENDNOTES [1] OECD, IPTV: Market Developments and Regulatory Treatment, DSTI/ICCP/CISP(2006)5/FINAL, at 24 (Dec. 17, 2007), available at www.oecd.org/dataoecd/11/23/39869088.pdf. [2] Telecomm. Reg. Auth. of India (TRAI), Recommendations on Provision of IPTV Services, at 18-20 (Nov. 28 2007).

4.5.5 Network unbundling
Unbundling access to local loops allows new market entrants to employ the fixed infrastructure of incumbents and provide advanced services, including IPTV and ancillary interactive services. It also expands new entrants’ potential market reach and increases IPTV service competition.[1] Several countries have required local loop unbundling, including all of OECD countries except Mexico. Many developing countries, such as Colombia, Peru, and South Africa, have introduced or proposed mandated local loop unbundling into their regulatory frameworks.[2] In many countries where IPTV already has a high market penetration (such as France, Italy and Spain), unbundling has been a key factor allowing new entrants to develop competing offers and increase IPTV penetration. In Japan, the main broadband competitor, Yahoo! BB, offers an IPTV service based on unbundled infrastructure that competes with the incumbent’s IPTV service. As incumbents plan network upgrades and build fiber-based, next-generation networks, regulators are considering whether to modify unbundling rules in order to avoid disrupting the broadband Internet and IPTV services of incumbents. For instance, European countries such as the UK have conducted public consultations to adapt their existing unbundling rules to the new fiber network architectures. Among the issues being considered is the feasibility of unbundling the last mile of fiber network architecture. The consultation also has addressed whether or not to introduce access obligations for additional local loop network elements, such as street cabinets, ducts, and the fiber itself.[3] ENDNOTES

[1] OECD, IPTV: Market Developments and Regulatory Treatment, DSTI/ICCP/CISP(2006)5/FINAL, at 15 (Dec. 17, 2007), available at www.oecd.org/dataoecd/11/23/39869088.pdf. [2] OECD Communications Outlook 2007 53-59 (2007), available at http://213.253.134.43/oecd/pdfs/browseit/9307021E.PDF. [3] See Off. of Comms. (Ofcom), Regulatory Challenges Posed by Next Generation Access Networks (Nov. 23, 2006), available at www.ofcom.org.uk/research/telecoms/reports/nga/nga.pdf.

4.6 Mobile TV
This section defines mobile TV and analyzes its classification in various jurisdictions. It also addresses the issue of whether regulations and obligations typically related to broadcasters should be extended to mobile TV. In addition, this section addresses the manner in which different countries are licensing mobile TV as well as the spectrum issues surrounding mobile TV.

4.6.1 What is mobile TV?
Defining mobile TV Mobile TV is the wireless transmission and reception of television content – video and voice – to platforms that are either moving or capable of moving. Mobile TV allows viewers to enjoy personalized, interactive television with content specifically adapted to the mobile medium. The features of mobility and personalized consumption distinguish mobile TV from traditional television services. The experience of viewing TV over mobile platforms differs in a variety of ways from traditional television viewing, most notably in the size of the viewing screen. The technologies used to provide mobile TV services are digitally based, and much of the terminology used in describing mobile TV reflects Internet phraseology. For example, the terms unicast and multicast are used in the same way they are used for IPTV. That is, unicasting is transmission to a single subscriber, while multicasting sends content to multiple users. These definitions also correspond to those given for similar Internet-based applications.[1] Technical aspects There are currently two main ways of delivering mobile TV. The first is via a two-way cellular network, and the second is through a oneway, dedicated broadcast network. Each approach has its own advantages and disadvantages. Delivery over an existing cellular network has the advantage of using an established infrastructure, inherently reducing deployment costs. At the same time, the operator has ready-made market access to current cellular subscribers, who can be induced to add mobile TV to the services they buy. The main disadvantage of using cellular networks (2G or 3G) is that mobile TV competes with voice and data services for bandwidth, which can decrease the overall quality of the mobile operator’s services. The high data rates that mobile TV demands can severely tax an already capacity-limited cellular system. Also, one cannot assume that existing mobile handsets can receive mobile TV applications without major redesign and replacement. Issues such as screen size, received signal strength, battery power, and processing capability may well drive the mobile TV market to design hand-held receivers that provide a higher quality of voice and video than is available on most current cellular handsets. Many 2G mobile service operators and most 3G mobile service providers are providing VOD or streaming video. These services are mainly unicast, with limited transmission capacity. They are built upon the underlying technologies used in the mobile cellular system itself – GSM, WCDMA, or CDMA2000.[2] An example of a technology designed to work on a 3G network is Multimedia Broadcast Multicast Service (MBMS), a multicast distribution system that can operate in a unicast or multicast mode.[3] MBMS has been designed by the 3rd Generation Partnership Project (3GPP) to provide mobile TV services over existing GSM and WCDMA cellular networks. It operates in the 5 MHz WCDMA bandwidth, and it supports six parallel, real-time broadcast streaming services of 128 kbit/s each, per 5 MHz radio channel.
Video Services over Mobile Networks Live Mobile TV Over 3G Network Examples of Operators Users Services Orange Mobile TV AT&T Wireless (using MobiTV) Subscribers only; closed network Live television VOD, instant messaging Network 3rd generation mobile networks One way dedicated broadcast network Live Mobile TV Over Dedicated Network V-Cast Mobile TV (Verizon) 3 Italia

Technology Platform

MBMS

MediaFLO DVB-H/SH DMB

Video Production Video Quality

Professional video

Managed QoS MPEG-4

Receiver device

Requires a standard 3G cellular phone

Requires a new dual-mode handset capable of receiving the broadcast signal and the cellular signal for phone calls and mobile Internet access Limited availability in certain countries; trial stages elsewhere Cost of building a dedicated network

Status of roll-out Relatively wide availability—service is available to any 3G subscriber on a network offering mobile TV Relative Limitations 3G network may not be able to support mobile TV traffic as the number of 3G voice and data users grow

Source: TMG, Inc. research

Dedicated mobile TV systems, however, can be designed to optimize the delivery of mobile TV. These systems are either totally terrestrially based, completely satellite based, or a combination of both. One of the major advantages of a dedicated mobile TV delivery system lies in the relative ease with which mobile TV content can be multicast to numerous users simultaneously. On the other hand, the disadvantages include the large capital investments in infrastructure that are required and the limited content options that are currently available, although that issue should dissolve significantly as the mobile TV market grows. Mobile TV standards for dedicated systems There have been significant advances in the development of standards used to support mobile TV transmissions and mobile multimedia by dedicated delivery systems. These include standards for:
s s s s

Digital Video Broadcasting-Handheld (DVB-H), Digital Multimedia Broadcasting (DMB), Integrated Services Digital Broadcasting-Terrestrial (ISDB-T), and MediaFLO (see box below).

These standards employ advanced modulation techniques such as orthogonal frequency division multiplexing (OFDM) and are interoperable with mobile telecommunication networks.[4] DVB-H has been identified as the mobile TV standard in most of Europe, due to its compatibility with GSM and WCDMA mobile standards. TDMB is being used in the Republic of Korea, Indonesia and parts of Europe, and a satellite version of the technology (S-DMB) is operating in the Republic of Korea. ISDB-T was developed in Japan to provide mobile TV services. MediaFLO technology is being extensively deployed in the United States for mobile TV applications. In addition to the above standards – which form the basis for Recommendation ITU-R BT.1833 − there are other mobile TV transmission technologies in various stages of standardization or deployment in countries around the world. These include DAB-IP mobile TV technology, Advanced-VSB technology, and the China Mobile Multimedia Broadcasting (CMMB) system.
Mobile TV Standards Standards that form the basis for Recommendation ITU-R BT.1833:
s

DVB-H: is based on the DVB-T digital broadcast standard and is optimized for handheld terminals. DVB-H incorporates time-slicing to reduce power consumption and to allow time for a smooth handover from one cell to another. It is designed to operate in bandwidths of 5 MHz, 6 MHz, 7 MHz, and 8 MHz, which correspond to the bandwidths used by broadcasting services around the world. Terrestrial Digital Multimedia Broadcasting (T-DMB): is an enhancement of the T-DAB system to provide multimedia services including video, audio, and interactive data services for handheld receivers in a mobile environment. It operates in a channel bandwidth of 1.712 MHz and is completely backward-compatible with the T-DAB system for audio services. ISDB-T: There are two distinct systems identified in Recommendation ITU-R BT.1833 for mobile TV: one is based on the ISDB-T-one segment and operates in bandwidths of 429 kHz, 500 kHz, or 571 kHz, and the other system is a hybrid terrestrial/satellite system Multimedia System “E” based on Digital System E of Recommendations ITU-R BO.1130 for the satellite component and ITU R BS.1547 for the terrestrial component. It operates in a 25 MHz bandwidth. Receivers are typically handheld with a 3.5 inch wide display for video and data broadcasting, in addition to high-quality audio. The satellite part of the standard provides nationwide coverage in Japan, with terrestrial gap-fillers augmenting areas that are shadowed from the satellite footprint. Media Forward Link Only (MediaFLO): is an end-to end system that enables broadcasting of video streams, audio-only streams, digital multimedia files, and data-casting to mobile devices, including handheld receivers. The system is designed to optimize coverage, capacity, and power consumption for handheld receivers. It can operate in channel bandwidths of 5 MHz, 6 MHz, 7 MHz, or 8 MHz.

s

s

s

Other mobile TV technologies in various stages of standardization or deployment:
s

DAB-IP mobile TV technology: is a variant of the ETSI DAB standard and was standardized by ETSI in mid-2006. It has a limited amount of channels compared to DVB-H or MediaFLO but as of 2006 it was the only standard that could be deployed commercially in the United Kingdom, as the spectrum needed for DVB-H was not available. Advanced-VSB technology: builds on the current North American ATSC television transmission standard to enable mobile receivers to receive television broadcasts. It is backward-compatible with current digital television receivers in the United States. The China Mobile Multimedia Broadcasting (CMMB) system: is a satellite/terrestrial wireless broadcast system designed to provide audio, video and data service for handheld receiver. The system employs high-power satellite signals and a complementary terrestrial network. The CMMB system can operate in 2 MHz or 8 MHz channels, uses OFDM modulation, and supports interactive services by cooperating with terrestrial telecom networks.

s

s

Sources: Recommendation ITU-R BT.1833, Appendix 1; Recommendation ITU-R BT.1833, Table 1; Appendix 2, Table 1; Recommendation ITU-R BT.1833, Tables 1, 2, 3; Section 4.4; Annex 4; Appendix 2, Table 1; Recommendation ITU-R BT.1833, Tables 1, 2, 3; Section 4.3; Annex 3; Appendix 2, Table 1; Recommendation ITU-R BT.1833, Tables 1, 2, 3; Section 4.1; Annex 2; Appendix 2, Table 1; and Recommendation ITU-R BT.1833, Tables 1, 2, 3; Section 4.5; Annex 5; Appendix 2, Table 1.

ENDNOTES [1] Networking Glossary, Searchnetworking.techtarget.com (February 2008). [2] Comms. and Tech. Branch, Com., Industry and Tech. Bureau, Hong Kong Consultation on Digital Broadcasting: Mobile Television and Related Issues 7 (2007), available at www.cedb.gov.hk/ctb/eng/paper/doc/mobile_TV.pdf. [3] ITU, Recommendation ITU-R BT.1833: Broadcasting of Multimedia and Data Applications For Mobile Reception by Handheld Receivers app. 1 (2007). [4] ITU-R BT.1833: Broadcasting of Multimedia and Data Applications For Mobile Reception by Handheld Receivers app. 1 (2007) and ITU-R Report BT.2049-1: Broadcasting of Multimedia and Data Applications for Mobile Reception (2005) provide further detailed information on these standards.

4.6.2 Classification of mobile TV
Given that mobile TV has only recently started being deployed, regulators have only begun to consider the possible regulatory classification of these services. Nevertheless, specific trends can be distinguished. Some jurisdictions have opted for a light-handed approach, classifying mobile TV as an information service, while others regulate it, or are proposing to regulate it, as a broadcasting service. In the United States, mobile TV services (both second generation and 3G, as well as dedicated mobile systems offering live television channels) are classified as information services and are not subject to broadcasting rules and regulations.[1] Singapore’s broadcasting regulator, the Media Development Authority (MDA), is proposing to classify mobile TV services and cellular mobile TV services (point-to-point video distribution services) as broadcasting services. MDA determined that a technology-neutral approach should mean that both types of mobile TV services would be regulated in the same manner, independent of the transmission platform.[2] 3G mobile providers in Singapore strongly oppose this determination and its regulatory implications. They believe their current licences allow them to offer such services, and they should not be regulated as broadcasters.[3] Other jurisdictions have found that existing broadcasting regulations are not applicable to mobile TV services.[4] For example, in Hong Kong, China, the Broadcasting Ordinance was drafted in the context of television reception at a specified location, rather than for an audience with mobility. Given this, Hong Kong, China is proposing two alternative approaches. The first option would provide for a selfregulatory approach. Mobile TV would not be classified as a television programming service. Instead, mobile content would be regulated in the same manner as content provided over the Internet and providers would be required to draw up and comply with codes of practice.[5] The second proposed approach would amend the Broadcasting Ordinance to include mobile TV as a new category of service (both over mobile networks and as dedicated systems).[6] Other governments have amended existing broadcasting regulation to cover mobile TV. For instance, in 2006 the Italian regulator, AGCOM, amended the 2001 digital terrestrial television regulations to extend their coverage to mobile TV services delivered over broadcasting networks (i.e., in the case of Italy’s DVB-H networks).[7] AGCOM has classified these mobile TV services as broadcasting services. Similarly, in the Republic of Korea amendments were introduced to the broadcasting regulations to include mobile TV services over broadcasting networks, in effect creating new “mobile multimedia broadcasting services” (both terrestrial and satellite).[8] By contrast, regulators in other countries have chosen to tread more lightly. The Canadian regulator, CRTC, exempted mobile TV service over the public Internet from licensing or other requirements of the Broadcasting Act of 1999 and 2006.[9] The exemption applies to operators that use point-to-point technology to deliver the service, meaning that the operator transmits a separate stream of broadcast video and audio to each end user. The CRTC determined that a variety of factors made it unnecessary − and potentially detrimental to the development of mobile broadcasting − to impose more stringent broadcasting conditions on these operators. CRTC found that point-to-point mobile TV will not have a significant impact on traditional broadcasters because of the inherent limitations of the wireless technology employed, battery life, screen size of the handset, and the type and range of programming choices offered by the mobile broadcasters.

[10] But CRTC has yet to make a determination regarding the regulation of dedicated point-to-multipoint mobile TV systems. ENDNOTES [1] Nat’l Cable & Telecomms. Ass’n v. Brand X Internet Servs., 545 U.S. 967 (2005). [2] Cable & Sat. Broad. Ass’n of Asis (CASBAA), MDA Public Consultation on the Policy and Regulatory Framework for Mobile Broadcasting Services in Singapore (2007), available at www.mda.gov.sg/wms.file/mobj/mobj.1180.Casbaa.pdf. [3] See Singapore Telecom Mobile PTE LTD, Submission to the Media Development Authority of Singapore, Policy & Regulatory Framework for Mobile Broadcasting Services in Singapore (2008), available at www.mda.gov.sg/wms.file/mobj/mobj.1196.SingTel.pdf. See also, StarHub Mobile PTE LTD, Response to MDA Consultation Paper, Policy and Regulatory Framework for Mobile Broadcasting Services in Singapore (2008) available at www.mda.gov.sg/wms.file/mobj/mobj.1200.StarHub.pdf. [4] After an initial public consultation in early 2007, regulatory authorities in Hong Kong, China have noted that mobile TV services provided over 2.5G/3G mobile networks and broadcasting networks currently do not fall within the purview of the Broadcasting Ordinance. Nevertheless, authorities have found that some form of regulation for the public good should be applicable to these services (e.g., protection of public moral and children). See Comms. and Tech. Branch, Com., Industry and Tech. Bureau, Hong Kong Consultation on Digital Broadcasting: Mobile Television and Related Issues 7 (2007), available at www.cedb.gov.hk/ctb/eng/paper/doc/mobile_TV.pdf. See also Comms. and Tech Branch, Com. and Econ. Dev. Bureau and Of. of the Telecomms. Auth., Second Consultation on Development of Mobile Television Services, paras 6.2, 6.4 (2008) (“Hong Kong Second Mobile TV Consultation”). [5] Hong Kong Second Mobile TV Consultation, at paras 6.6, 6.7. [6] Id. at para 6.11. [7] Comms. Reg. Auth. (AGCOM), Delibera n. 266/06/CONS, Modifiche al regolamento relativo alla radiodiffusione terrestre in tecnica digitale di cui alla delibera n. 435/01/CONS. Disciplina della fase di avvio delle trasmissioni digitali terrestri verso terminali mobili, available at www.agcom.it/provv/d_266_06_CONS.htm. [8] Enforcement Decree of the Broadcasting Act, art. 1-2(3) (2007) (Republic of Korea), available at http://www.moleg.go.kr/FileDownload.mo?flSeq=26454. [9] Can. Radio-television and Telecomms. Comm’n (CRTC), Canada Broadcasting Public Notice CRTC 2007-13, Feb. 7, 2007, available at http://www.crtc.gc.ca/eng/archive/2007/pb2007-13.htm. [10] Id. at para 42. The CRTC further noted that it would initiate a further proceeding to examine whether mobile TV services not delivered and accessed over the Internet would be exempt from the broadcasting conditions. It found that, at the time, much remained unknown about the potential impact of the broadcast-based or point-to-multipoint mobile technologies, particularly if they could become a substitute for conventional television. Id. at para 27.

4.6.3 Applying content regulation to mobile TV
Many countries are applying television content regulations to mobile TV providers. The EU, for example, imposed the same restrictions on mobile TV advertising that apply to television broadcast advertising. In Singapore, the Media Development Authority (MDA), which is responsible for mobile broadcasting, conducted a public consultation in which it proposed subjecting mobile TV service providers, as well as cellular mobile TV providers, to broadcasting regulation. This would put them under MDA’s programming codes for free over-the-air content, subscription content, VOD and other kinds of content.[1] Singapore also proposed applying to mobile TV the existing framework for advertising regulation. In Australia, regulation restricting minors’ access to certain content was applied to mobile premium services, including mobile portal and premium rate SMS/MMS services.[2] This regulation removed content-related provisions regarding mobile phones from the Telecommunications Service Provider (Mobile Premium Services) Determination of 2005. Meanwhile the Restricted Access System Declaration, enacted in accordance with the new Schedule 7 of the Broadcasting Service Act 1992, was applied to mobile premium services such as mobile TV. ENDNOTES [1] Media Dev. Auth. (MDA), Public Consultation on Policy and Regulatory Framework for Mobile Broadcasting Services in Singapore, at 18 (Nov. 21, 2007), available at: www.mda.gov.sg/wms.file/mobj/mobj.1167.Mobile%20TV%20Consultation.pdf. [2] Ausli. Comms. & Media Auth. (ACMA), Restricted Access System Declaration 2007, (Dec. 20, 2007). This new rule for restricting access to age restricted content becomes effective on Jan. 20 2008.

4.6.4 Licensing mobile TV providers

The licensing approaches for mobile TV resemble those for IPTV services. In several countries, the government distinguishes between content and carriage. In Singapore, for example, MDA has proposed subjecting mobile TV to the existing licensing structure for fixed digital broadcasting. Mobile TV operators would obtain both a multiplex licence and a broadcasting service licence from MDA, under the Broadcasting Act, as well as a “Facilities-Based Operator” licence issued by the Info-communications Development Authority (IDA) under the Telecommunications Act.[1] In the United States, however, a licensee operating under one of the C block (710-716/740-746 MHz) or D block (716-722 MHz) licences in the UHF band can provide “flexible fixed, mobile, and broadcast uses, including mobile and other digital new broadcast operations, fixed and mobile wireless commercial services (including FDD- and TDD-based services). . .[that] could also include two-way interactive, cellular, and mobile TV broadcasting services.”[2] In January 2008, Hong Kong, China issued a consultation paper on mobile TV, proposing to licence mobile TV services under a new category of television programme service provided in the Broadcasting Ordinance. Or, mobile TV could be regulated through general laws (as currently is the case), if industry implemented a code of practice for self-regulation.[3] Currently, mobile TV providers offering streaming-type mobile TV services (already available on 2.5 GHz and 3 GHz mobile networks) are not subject to broadcasting regulation and can offer services if they hold a mobile carrier licence.
Hong Kong, China’s Regulation of Mobile TV, IPTV, and Internet TV Mobile TV Carriage Licences Content Licences Mobile licence or unified carrier licence (proposed by regulators to replace both fixed and mobile carrier licences in future) Not currently applicable to mobile TV on 2.5 and 3 GHz networks IPTV Services Carrier licence required for conveyance of IPTV services Internet TV No licence required

Broadcasting licence required: IPTV service Exempted from the licensing over fixed network is categorized as domestic requirement under the pay TV programme service Broadcasting Ordinance

Source: Consultation on Digital Broadcasting: Mobile Television and Related Issues, UCAC Paper No. 3/2007, 26 April 2007, at p. 12, available at http://www.ofta.gov.hk/en/ad-comm/ucac/paper/uc2007p3.pdf

In January 2008, TRAI in India issued recommendations on mobile TV. Since cellular licensees were already allowed to deliver video content over their networks, the recommendations primarily addressed mobile TV licensing for dedicated broadcast networks.[4] TRAI proposed a tender process to award mobile broadcast licences in the 582-806 MHz band for Digital Terrestrial Television (DTT) and in the 2520-2670 MHz band for satellite transmission. TRAI further recommended that mobile TV operators not be made responsible for following content codes if they simply retransmitted channels without altering content. ENDNOTES [1] Media Dev. Auth. (MDA), Public Consultation on Policy and Regulatory Framework for Mobile Broadcasting Services in Singapore, at 8 (Nov. 21, 2007), available at: www.mda.gov.sg/wms.file/mobj/mobj.1167.Mobile%20TV%20Consultation.pdf. [2] FCC, FCC Auction 49: Lower 700 MHz Band Fact Sheet, available at: http://wireless.fcc.gov/auctions/default.htm? job=auction_factsheet&id=49. [3] Comms. and Tech Branch, Com. and Econ. Dev. Bureau and Of. of the Telecomms. Auth., Second Consultation on Development of Mobile Television Services, paras 6.2, 6.4 (2008). [4] TRAI, India, Recommendations on Issues Relating to Mobile Television Service, (Jan. 23, 2008).

4.6.5 Spectrum issues
The major issue facing the deployment of any mobile TV system is access to the spectrum needed to support the services. The availability and cost of spectrum will largely dictate the technology deployment options available to the operator. For example, if a potential operator wishes to deploy a satellite-based mobile TV system, there must be a satellite broadcasting allocation for the desired spectrum band in that country. If there is no spectrum available for a dedicated terrestrial mobile TV network, then the mobile operator must provide mobile TV service in bands that are already being used for more traditional mobile services. This will limit the mobile provider’s options and may affect QoS. As countries move to identify spectrum for dedicated mobile TV networks, government should consider compatibility between new and existing services. The choice of mobile TV technology should be left to the operator, as long as that technology conforms to national and international frequency allocations. ENDNOTES [1] For example, in the ongoing Public Consultation on Policy and Regulatory Framework for Mobile Broadcasting Services in Singapore, Media Development Authority (MDA) has proposed not to mandate a single standard for mobile television service in Singapore as the MDA

found it premature for a small market like Singapore to set a “particular standard when there is no obvious global champion,” as well as no need for the MDA to depart from its policy of technology neutrality. Media Dev. Auth. (MDA), Public Consultation on Policy and Regulatory Framework for Mobile Broadcasting Services in Singapore, at 6 (Nov. 21, 2007), available at: www.mda.gov.sg/wms.file/mobj/mobj.1167.Mobile%20TV%20Consultation.pdf.

4.7 Other legal and regulatory issues impacting IPTV and mobile TV
IPTV and Mobile TV face common legal and regulatory issues. This section addresses legal issues related to acquiring content, vertical integration in content markets, standards, quality of service issues, and ownership issues. It also looks at the regulatory authorities responsible for IPTV and mobile TV and provides a checklist for regulators introducing IPTV and mobile TV.

4.7.1 Legal issues related to acquiring content
IPTV and mobile TV providers need to offer desirable content if they want to attract viewers. But it’s a difficult task to locate and contract with every individual content provider, ensure that the provider has the necessary rights to distribute the material, and then manage those relationships. IPTV and mobile TV providers may also find themselves in a difficult negotiating position as movie studios and other content providers demand the largest possible audiences for maximum exposure. Content generators also seek guarantees of a secure distribution chain that will safeguard their intellectual property rights, as well as guarantees of high-quality transmission. Start-ups may find it difficult to compete with the more established media entities that deliver programming over traditional media (broadcast, cable, and satellite). The media giants have more resources, experience, and viewers. A new breed of business — the content aggregator — has sprung up to fill the gap between retail providers (that is, IPTV and mobile TV providers) and the content providers. Content aggregators act as middlemen in order to obtain the rights to content and then facilitate its distribution through their clients. Also, they typically offer security enhancements and due-diligence services, ensuring that the content provider has legitimate rights to distribute the content. IPTV and mobile TV providers get one-stop-shopping for content, as well as assistance in negotiating licence agreements and help in understanding the contours of their distribution rights. Content providers, in turn, get wider exposure for their material via outlets that otherwise may have gone untapped.

4.7.2 Vertical integration in content markets
Vertical integration gives the entity controlling both the content or programming rights and the distribution platform the ability to discriminate in favor of its affiliated video distributor (e.g., cable or satellite) to the detriment of competitors in the downstream market. Although there may be economic efficiencies that benefit consumers, such discrimination may also lessen competition and diversity in the distribution of video programming, ultimately harming consumers. In India, for example, concerns arose about having centralized control over content production and distribution, services, and platforms across different sectors. That concern led to a proposed bill, the Broadcasting Services Regulation Act of 2007, which would impose a 20 per cent cross-holding restriction between any "content broadcasting service provider" and any "broadcasting network service provider." In the United States, exclusive contracts for satellite cable programming or satellite broadcast programming are prohibited between vertically integrated content vendors and cable operators. These “Program Access Rules” were introduced to address the concern that potential competitors to incumbent cable operators, particularly satellite television providers, would be unable to gain access to the programming offered by vertically integrated cable operators. In countries such as Australia, Spain, and the United Kingdom, regulators have also imposed restrictions on exclusive content agreements between vertically integrated video programmers and distributors. These restrictions, however, have been adopted on a case-by-case basis in the context of mergers and other transactions. In Singapore, meanwhile, the lack of vertical integration between video programmers and cable distributors was a determining factor for the Ministry of Information, Communications and the Arts (MICA) to allow StarHub Cable Vision’s exclusive carriage agreements.

4.7.3 Standards

Governments need to consider which mobile TV and IPTV standards should be authorized in their country, or whether they will leave the choice of standards up to the network operators. Several standards are available for IPTV, such as Microsoft or DVB-based standards. But these are not interoperable, and they can create difficulties for consumers who may want to change service providers, a process that may require new hardware and new user interfaces. The challenge for the industry and regulators is to create open standards and facilitate interoperability. With the development of multi-platform set-top boxes, the industry can help create more choices and better use of resources. Regulations should encourage this process. In addition, regulators need to consider whether current laws and regulations on equipment certification may impose barriers to rolling out IPTV or mobile TV. For example, in India, concerns have been raised about whether cable operators’ use of IPTV set-top boxes would violate the Cable Television Networks (Regulation) Act, 1995, which bars cable equipment that does not use Indian-developed standards. Since there is, in fact, no Indian standard for an IPTV set-top box, TRAI has issued an opinion that the Act has not been violated – but it also recommended that the Bureau of Indian Standards expedite IPTV standards development.

4.7.4 Quality of service issues
Quality of service (QoS) is likely to be an issue regulators will consider. With services provided over the Internet, QoS amounts to a “best effort” standard. But with IPTV and mobile TV offerings, a provider can control QoS because it uses a privately managed network. In addition, given the importance of better picture quality with IPTV and mobile TV, it is in the operator’s interest to provide high-quality service or the consumer will go elsewhere. In Singapore, MDA was considering whether to impose QoS requirements on mobile TV, but it proposed not to mandate picture quality or performance indicators for customer service. MDA indicated that it believes imposing picture quality requirements would limit the mobile TV providers’ flexibility to determine an optimal mix of formats, and customer service mandates are not necessary in a competitive environment. However, MDA did propose minimum network coverage requirements to ensure that (1) mobile TV services are offered throughout Singapore, and (2) outdoor coverage meets a 95 per cent threshold.[1] ENDNOTES [1] Media Development Authority (MDA), Public Consultation on Policy and Regulatory Framework for Mobile Broadcasting Services in Singapore, at 6 (Nov. 21, 2007), available at: www.mda.gov.sg/wms.file/mobj/mobj.1167.Mobile%20TV%20Consultation.pdf.

4.7.5 Ownership issues
Several countries maintain ownership restrictions that may impede the development of IPTV and mobile TV services. Joint-provisioning prohibitions may bar a telecommunication provider from operating a cable subscription service. For example, Argentina forbids a telecommunication provider from offering of any type of broadcasting services over any platform, whether it is subscription-based or not. In Brazil, incumbent telecommunication providers are banned from providing cable broadcasting services in areas where they offer telecommunication services. They are permitted, however, to provide VOD after obtaining a direct-to-home licence, and they may also offer satellite television. Countries also may have cross-ownership restrictions that prevent telecommunication providers from owning cable networks and vice versa. Unlike joint-provisioning restrictions, cross-ownership restrictions do not necessarily prevent the affected firm from entering a specific market, provided its entry strategy does not involve acquiring an existing player in that market. In the context of public switched telecommunication networks and cable television services, cross-ownership limitations generally are directed at safeguarding the independence of competing service providers and fostering facilities-based competition. In India, for example, broadcasters and cable operators may own only up to 20 per cent of any satellite television company. In January 2007, TRAI issued recommendations to the Ministry of Information and Broadcasting proposing to extend the 20 per cent cap to mobile TV providers. In response to convergence, however, some governments have pursued legislation to eliminate or modify these restrictions. For example, the Mexican government introduced a “convergence agreement” that eliminated the joint-provision restrictions barring incumbent Telefonos de Mexico (Telmex) from providing television services. The government did, however, attach conditions to the agreement, requiring clear interconnection rules and number portability processes. The agreement, however, was affected by competition concerns related to joint ownership in Mexico. As a result, the Communications and Transport Secretariat and the Federal Competition Commission established a presumption that joint ownership would stifle inter-modal competition in Mexico. If an interested party can rebut this presumption and present evidence of efficiencies derived from the potential joint ownership, the Competition Commission may grant a waiver of the restrictions.[1] The Brazilian Congress also began working on a comprehensive law to eliminate the joint provisioning restriction.[2] But in other countries, cable operators have mounted strong efforts to oppose lifting bans on joint provisioning. For example, in an attempt to prevent telecommunication providers from offering IPTV, the Argentine Cable Television Association filed an appeal with the Federal Administrative Court (affirmed by the Supreme Court), which upheld the constitutionality of a law barring telecommunication companies from entering the broadcasting services market.[3]

Traditionally, there has been great sensitivity regarding broadcasting services and foreign ownership. As a result, while foreign ownership restrictions have been eliminated in many countries for telecommunication companies, they often remain in place for traditional broadcasting companies. In India, for example, the foreign ownership cap for telecommunication providers is 74 per cent – but just 49 per cent for cable operators.[4] Similarly, the Republic of Korea has different foreign ownership restrictions for the broadcasting and telecommunication sector.[5] ENDNOTES [1] See Fed. Competition Comm’n, OF. No. PRES-10-096-2005-117, Opinion Twelve (Oct. 31, 2005); Fed. Competition Comm’n, OF. No. PRES10-096-2006-102, Opinion Eleven (Jul. 7, 2006); Commns. and Transport Secretariat, Convergence Agreement, Second Accord (Oct. 3, 2006). [2] The draft law nº 29 establishes that the concessionaires operating at the switched fixed line services would be allowed to offer cable TV; except in localities where a cable TV provider launched its operations less than one year before the issuance of the new law (draft 29). In those localities, the concessionaires would need to wait one year after the issuance of the law in order to acquire a cable TV licence. [3] Cooperativa Telefonica de Libertador General San Martin before the Supreme Court of the Nation (interpreting article 45 of Broadcasting Act 22.285, as amended by Law 26.053 of Sept. 14, 2005). [4] Telecomm. Reg. Auth. of India (TRAI), Recommendations on Provision of IPTV Services, at 28 (Nov. 28 2007). [5] Foreign equity cap for broadcasters and backbone telecommunications service providers is capped at 49 percent. However, the method of calculating foreign equity different under the Telecommunications Business Act (calculation is based only on stocks with a voting right) and the Broadcasting Act (calculation is based on all stocks). Int’l Telecomms. Union (ITU), IPTV Case Study of Korean Digital Convergence, Study Group 1, ITU-D/E/57-E, at 10 (Nov. 4, 2007).

4.7.6 Regulatory authorities responsible for IPTV and mobile TV
By 2009, more than 150 countries have established independent regulatory authorities. Among these are several converged regulators, such as those in Australia, Finland, Italy, Malaysia, South Africa, Singapore, the United States, and the United Kingdom.[1] Nearly 30 converged regulatory agencies have opened their doors in past seven years.[2] The rationale for this accelerating trend is that a converged regulator is better suited to respond to an environment where distinctions based on service and network platform are becoming blurred. A converged regulator establishes a single government entity to oversee all matters involving the communication sector. Despite this trend, most OECD countries still have separate regulators for broadcasting and for telecommunications.[3] In addition, content regulation is typically addressed by a separate ministry or government authority (such as in India and Saudi Arabia) or by the broadcasting authority (Botswana, Chile, and Colombia). In India, there are two entities responsible for content regulation. The Ministry of Information and Broadcasting monitors content related to broadcasting and film, and the Ministry of Information Technology regulates content related to the Internet.[4] As noted in the figure below, many countries retain multiple government authorities responsible for the functions of broadcasting licensing, telecommunication licensing, spectrum allocation, and content regulation.
Heading Telecoms Spectrum CNC Broadcast Spectrum CNC

Country Argentina

Telecoms Carriage TextNational Communications Commission (CNC); Communications Secretariat (SECOM) Ministry of Communications, Science and Technology (MoCST); Botswana Telecommunications Authority (BTA) Ministry of Communications (MoC); Telecommunications Regulatory Commission (CRT) Telecommunications Secretariat (SUBTEL) within Ministry of Transport and Telecommunications National Telecommunication Regulatory Authority (NTRA); Ministry of Communications and Information

Broadcast Carriage Federal Broadcasting Committee (COMFER) National Broadcasting Board (NBB) National Television Commission (CNTV) National Television Council (CNTV) Egyptian Radio and Television Union [ERTU]

Content COMFER

Botswana

BTA

NBB

NBB; BTA

Colombia

MoC

CNTV

CNTV

Chile

SUBTEL

SUBTEL

CNTV

Egypt

NTRA

ERTU

Ministry of Interior (Internet security); ERTU (Broadcasting)

Technology (MCIT) France Regulatory Authority for Electronic Communications and Postal Service (ARCEP) National Spectrum Agency (ANFR) OFTA DoT Higher Council for Radio and Television (CSA) Broadcasting Authority (BA) andOFTA TRAI /Ministry of Information and Broadcasting (MI&B) (for licensing) Audiovisual Commission (AVC) ANFR; CSA ARCEP; CSA

Hong Kong, Office of the Telecommunications Authority (OFTA) China India Telecommunications Regulatory Authority of India (TRAI); Department of Telecommunications (DoT)(for licensing) Ministry of Information and Communications Technology (MoICT); Telecommunications Regulatory Commission (TRC) Communications and Transportation Secretariat (SCT) and Federal Telecommunications Commission (COFETEL) Ministry of Information Technology – IT and Telecom Division (MoIT) and Pakistan Telecommunications Authority (PTA) Infocomm Development Authority (IDA)

BA; OFTA DoT

BA Ministry of Information Technology (MIT) (Internet; MI&B(Broadcasting) AVC

Jordan

TRC

AVC in coordination with TRC SCT

Mexico

SCT

SCT; Secretariat of Public Education (SEP) Pakistan Electronic Media Regulatory Authority (PEMRA)

SEP; General Directorate for Radio, Television and Cinemato-graphy (RTC) within Executive Secretariat PEMRA

Pakistan

PTA

PTA

Singapore

IDA

iDA; Media IDA Development Authority (MDA) Uganda Broadcasting Council (UBC) UBC; UCC

MDA

Uganda United Kingdom United States

Uganda Communications Commission UCC (UCC) Office of Communications (Ofcom) Ofcom

UBC Ofcom

Ofcom; Department for Ofcom Culture, Media, and Sport FCC; local government FCC for cable TV franchises

Federal Communications Commission FCC [FCC], plus Various state-level public utility commissions (PUCs)

FCC, Federal Trade Commission (FTC)

Source: Based upon Telecommunications Management Group, Inc. research and Telecommunication Regulatory Institutional Structures and Responsibilities, OECD Paper, DSTI/ICCP/TISP(2005)6/Final, at p. 31, 32.

Operators are exposed to multiple sets of regulations and regulators, often delaying the rollout of IPTV and mobile TV services. Regulators often become embroiled in jurisdictional disputes over where their authority begins and another agency’s ends. In the Republic of Korea, for example, the Republic of Korea’s Broadcasting Commission considered converged service providers to be broadcasting companies, but the Ministry of Information and Communication argued that they were providing “value-added services.” Similar debates have arisen in Colombia between the National Television Commission (CNTV) and the Ministry of Communications over which body should regulate IPTV services. A common result of such bureaucratic infighting is that telecommunication operators cannot obtain IPTV authorizations, while their cable television rivals are free to begin offering them. In China, the Ministry of Information Industry (MII) and the State Administration of Radio, Film and Television (SARFT) share broadcast licensing responsibilities − resulting in confusion over which agency will regulate converging services like IPTV.[5] SARFT has interpreted a 1999 law as barring telecommunication operators from offering video services and has twice blocked operators from doing so. Since telecommunication operators are prohibited from controlling IPTV networks, they must enter into joint arrangements with broadcasters. For example, China Telecom, the country’s leading telecommunication operator, partnered with Shanghai Media Group, one of just four broadcasters granted an IPTV licence.[6] As governments look at how to promote new services such as IPTV and mobile TV, they should consider whether their institutional frameworks need updating. The Republic of Korea, for example, had four government authorities responsible for regulating the communication sector:
s s s s

The Telecommunications Commission, The Ministry of Information and Communication (MIC), The Broadcasting Commission, and The Ministry of Culture and Tourism.

This overlap in agencies was delaying the rollout of IPTV services. In December 2007, the South Korean government enacted a new law eliminating MIC and transferring most of its functions to the Ministry of Commerce, Industry and Energy; the Ministry of Culture and Tourism; and the Ministry of Government Administration and Home Affairs.[7] In addition, the legislation called for a unified broadcasting and communications commission, merging the Broadcasting Commission, the Telecommunications Commission and MIC’s Broadcasting Policy Office. ENDNOTES [1] ITU Telecommunications Regulatory Database (2007). [2] A “converged” agency can be defined as one that regulates multiple, previously separate regulated industries or sectors, reflecting the convergence of those sectors themselves (e.g., telecommunications, broadcasting and IT sectors). Some of these agencies may have existed for decades, but the majority were created in recent years, often through the combination of previously separate agencies. [3] OECD, Policy Considerations for Audio-Visual Content Distribution in a Multiplatform Environment, DSTI/ICCP/TISP(2006)3/FINAL, Jan. 12, 2007. [4] Telecomm. Reg. Auth. of India (TRAI), Recommendations on Provision of IPTV Services, at 20 (Nov. 28 2007). [5] MII coordinates the deployment of the public telecommunications network, the television and radio broadcasting networks. However, the State Administration of Radio, Film, and Television (SARFT) regulates broadcast content transmitted via radio, TV, satellite, cable, and the Internet and has authority to award or revoke licences to produce, distribute, and broadcast films, TV, and radio programs. [6] China Telecom provides the equipment and broadband network while the broadcaster holds the IPTV licence and contributes video content. [7] Farewell to the Ministry of Information and Communication, Telecom Korea News Service, Jan. 16, 2008.

4.7.7 Checklist for regulators introducing IPTV and mobile TV
The deployment of IPTV and mobile TV changes traditional perceptions and challenges existing laws and regulations. Both services offer enormous opportunities to provide consumers new platforms for multimedia content, enhancing competition and boosting broadband deployment.
Checklist for Regulators Introducing IPTV and Mobile TV For regulators, there are a variety of factors to consider in relation to these new services. In the case of IPTV, such factors are potentially broader due to the fact that incumbent telecommunications providers are subject to legacy regulation. Because of this, regulators need to consider the following questions:
s

What is the impact of legacy regulation on providers’ ability to offer services and on providers’ incentives to incur the significant investments and high risks associated with deploying/upgrading infrastructure to allow for the provision of IPTV services? In the case of IPTV, are there any legal or regulatory restrictions to incumbent telephone providers’ ability to provide video services within their markets (joint provision restrictions or cross-ownership restrictions)? If incumbents are not restricted from entering the video market, does the application of existing regulation, specifically issues such as access obligations to dominant providers’ network, skew the incentives for investment in deployment/upgrading of networks to support IPTV services?

s

s

Having performed this initial review, regulators might look at how, if at all, IPTV and mobile TV fall within the existing regulatory framework for broadcasting services.
s

Do the services offered by the IPTV or mobile provider fall within the definition of television broadcasting included in a country’s laws or regulations? If so, what type of regulation would be imposed on such providers? Does a mobile TV provider require multiple licences under the existing legal framework (telecommunications, broadcasting, and content)? Does your regulatory framework provide for a technology neutral approach for granting licences? How many regulatory entities claim jurisdiction over IPTV and mobile TV services? If not, can one entity be given the authority to regulate IPTV? Or mobile TV? Should laws and regulations relating to content be applicable to IPTV and mobile TV services? Are there legal restrictions that impede investment in IPTV and mobile TV services (e.g., foreign ownership restrictions)? Is extending existing broadcasting regulation to these services the best mechanism to foster their deployment?

s

s s

s s s

Source: Telecommunications Management Group, Inc.

As noted in the box above, however, governments need to look at their regulatory frameworks and institutional structures, and determine how best to facilitate the deployment of these services. Recognizing this, some countries have initiated consultations to address the regulatory issues related to these services. In addition, governments are considering whether to alter legal or regulatory restrictions that

now prevent telecommunication and video service providers from competing in each others’ markets. Hindering the development of a multiplatform, multi-service environment may result in limiting choices and benefits for consumers. In addition, governments are looking at how IPTV and mobile TV services should be regulated, including the appropriate licensing requirements and regulatory obligations. In some countries, regulators are opting to treat the offering of any television programming as broadcasting, leading them to regulate IPTV and mobile TV providers in the same manner as traditional broadcasters. However, other countries may, for policy reasons, choose to subject these new services to lighter regulation, at least for a certain period of time until the market develops. For regulators, there is no right or wrong approach. What is important is to give IPTV and mobile TV service providers some certainty about how they will be regulated. It is also important to minimize or eliminate jurisdictional debates among government agencies, as well as any onerous regulatory hurdles and cumbersome licensing requirements that may delay the deployment of new services.

Abbreviations
ADSL AMPS ATM ATSC

BBO BOO BOOT
BWA CA CATV CDMA CDMA2000 CEPT

CERTs
DAB DARPA

DBFO
DMB DSL DTH DTTV DTV DVB DVB-H DVB-T EDGE ENUM ETSI EU FCC FDM FEC FM FMC FMI FTTH FTTx FWA GPRS

Asymmetrical Digital Subscriber Line Advanced Mobile Phone System Asynchronous Transfer Mode Advanced Television Systems Committee Buy-Build-Operate Build-Own-Operate Build-Own-Operate-Transfer Broadband Wireless Access Conditional Access CAble TV Code Division Multiple Access CDMA 2000 Conférence Européenne des Administrations des Postes et Télécommunications Computer Emergency Response Teams Digital Audio Broadcasting Defence Advanced Research Projects Agency Design-Build-Finance-Operate Digital Multimedia Broadcasting Digital Subscriber Line Direct To Home Digital Terrestrial TV Digital TV Digital Video Broadcasting DVB- Handheld DVB- Terrestrial Enhanced Data rates for GSM Evolution tElephone NUmber Mapping European Telecommunications Standards Institute European Union Federal Communications Commission Frequency Division Multiplexing Forward Error Correction Frequency Modulation Fixed Mobile Convergence Fixed Mobile Integration Fibre To the Home FTTHome, FTTArea, FTTCabinet, FTTCurb, … Fixed wireless Access General Packet Radio Service

GPS GSM HDR HDTV HSCSD ICT IEC

IEEE
IP IPTV IPv4 IPv6 ISDB ISDN ISM SLA ISP ITU ITU-T LAN LF LRAIC MAN MediaFLO MOS MPEG

MSP
NGAN NGCN NGN NMT PCM PDC PDH PHS PLC POTS PPP PSTN QoS RFID SDH SDR TACS

Global Positioning System Global System for Mobile communication Hardware Defined Radio High Definition TV High Speed Circuit Switched Data Information and Communication Technologies International Electronical Commission Institute of Electrical and Electronics Engineers Internet Protocol IP Television IP version 4 IP version 6 Integrated Service Digital Broadcasting, Integrated Services Digital Network Industrial, Science and Medical Service Level Agreement Internet Service Provider International Telecommunication Union ITU-Telecommunication Local Area Network Low Frequency Long Run Average Incremental Cost Metropolitan Area Network Media Forward Link Only Mean Opinion Score Moving Pictures Expert Group Multi-Stakeholder Partnerships Next Generation Access Network Next Generation Core Network Next Generation Network Nordic Mobile Telephone Pulse Code Modulation Personal Digital Cellular Plesiochronous Digital Hierarchy Personal Handyphone System Power Line Communication Plain Old Telephony Services Public Private Partnership Public Switched Telephone Network Quality of service Radio Frequency Identification Synchronous Digital Hierarchy Software Defined Radio Total Access Communication Systems

TDCDMA TDMA UDP UHF UK US USO

VAN
VHF VHS VLF VoD VoIP

VSAT
WCDMA Wi-Fi WiMAX WSIS WWW

Time Division- Code Division Multiple Access Time Division Multiple Access User Datagram Protocol Ultra High Frequency United Kingdom United States Universal Service Obligation Value Added Network Very High Frequency Video Home System Very Low Frequency Video on Demand Voice over IP Very Small Aperture Terminal Wideband CDMA Wireless Fidelity Worldwide Interoperability for Microwave Access World Summit on Information Society World Wide WEB

Sign up to vote on this title
UsefulNot useful