Universal service in an IP-enabled NGN environment
The Authors
Patrick Xavier, Convernor, Economics for Business at the Faculty of Business, Swinburne University of Technology, Hawthorn, Australia.
Dimitri Ypsilanti, Head, Telecommunications Policy Section, Organisation for Economic Co-operation & Development (OECD), Paris, France.
Acknowledgements
This paper draws on work the authors did towards preparation of a paper for the OECD in December 2005 entitled “Rethinking universal service for a NGN environment” and a paper for the International Telecommunications Union (ITU) in March 2006 entitled “What rules for universal service in an IP-enabled NGN environment?”
Abstract
Purpose – The purpose of this paper is to identify concerns developing in regard to current approaches used to define, deliver and fund universal service obligations (USOs) for telecommunications due to developments in competition, new technology, e.g. wireless, convergence and next generation networks (NGN), including voice over internet protocol (VoIP). To stimulate thinking about the policies required to deliver USOs in an IP-enabled NGN environment.
Design/methodology/approach – The paper examines the nature and scope of USOs, including the concept of “functional internet access” to address questions about whether in an NGN environment: present USOs are sustainable; USOs need to change; and whether USOs should include access to broadband? The paper also examines alternative/complementary sources of funding for USOs.
Findings – With falling revenue (due to increasing competition and declining prices) occurring alongside strong demand for funds to deploy NGN, telecommunications operators are less able to provide USOs sustained by cross-subsidisation. Universal service funds, too, may be under pressure. Also there is likely to be an uneven migration to NGN since it is likely that NGN will be deployed first in more profitable, densely populated areas and then only later (if at all) in relatively less commercially viable rural/remote areas. The paper concludes that present USOs can be maintained in an NGN environment. It concludes that since USOs is an evolving concept, there should be regular systematic reviews of whether USOs should be extended to include broadband. The paper concludes that increased funding from general taxation revenue is warranted and would link decisions concerning the nature and scope of USOs more closely with financial responsibility for such decisions.
Originality/value – The paper stimulates rethinking about whether, and if so how, arrangements relating to USOs need to change? In adopting a longer term perspective of universal services, the paper may be a valuable complement to the reviews of USOs being conducted in various countries (since short term decisions should be consistent with longer term developments).
Article Type:
Research paper
Keyword(s):
Communication technologies; Transmission Control Protocol/Internet Protocol; Information services.
Journal:
info
Volume:
9
Number:
1
Year:
2007
pp:
15-31
Copyright ©
Emerald Group Publishing Limited
ISSN:
1463-6697
1. Introduction
1.1. Background
The past decade has witnessed a period of significant market liberalisation, competition, technological change and restructuring in the telecommunications sector. The rapid technological change is continuing and, indeed, the changes ahead may be even more significant. Among the major drivers of this change are wireless technology, internet protocol (IP) technology, and the convergence of media, computing and telecommunications.
Prior to convergence separate services depended on dedicated networks. The public switched telephone network (PSTN) was designed for person-to-person voice communications. Broadcast networks were optimised for one-way delivery of radio or television. And the internet was designed for non-real-time transportation of packets. These networks and services are now converging and the digitised bits flowing over the networks are “co-mingling”. A progressive replacement of the PSTN's voice-based, circuit-switched networks by IP-enabled “next generation networks” (NGN) appears to be gathering momentum and spreading. For example, IP-based video, telephony, and triple-play services are gaining popularity in an increasing number of countries. IP-enabled NGN is shifting from separate PSTN-and IP-networks to unified networks based on IP with “packet-based” multi-service platforms (in which “voice” is only one of a range of available services). After decades of traditional telephone service, there are signs of a fundamental transformation in voice services.
The transition to NGN is expected to bring many benefits making available an extensive range of innovative new services, greater control and personalisation and ease of migration between services. But there are also some concerns about these developments, including concerns relating to universal service.
One concern is that the migration from PSTN to NGN is likely to be uneven in both developed as well as developing countries. It is expected that the more profitable densely populated urban areas will be served first, with less commercially viable sparsely populated rural and remote areas, served later (if at all). If a policy of requiring that the same price be charged to customers in urban and rural areas is maintained (so-called uniform geographic pricing), this is likely to accentuate the uneven migration to NGN.
Another concern is over whether present universal service obligations (USOs) can be preserved in an NGN environment, not only in terms of availability, affordability and accessibility, but also in regard to quality and reliability of service. For instance, the quality of service and features of PSTN voice services that consumers have become accustomed to (such as the ability to make emergency calls with caller location information) might be degraded in an NGN environment with voice delivered through voice over internet protocol (VoIP). While as yet a relatively small portion of the total voice market, VoIP is expected to grow sharply and pervasively.
There are also concerns that VoIP provided by internet service providers (ISPs) over the internet at significantly lower (distance unrelated) prices will threaten the established revenue base for many traditional telephone operators (that typically depend heavily on higher prices for longer distances calls, and especially international calls). Voice revenue still accounts for the major part of total operator revenues and hence is crucial for funding investment in new technologies and services as well as for funding universal service. But revenue in the fixed voice sector is declining because of falling prices, reflecting intensifying competition, the widening use of low price pre-paid calling cards, competition from mobile, and the growth of broadband (resulting in a reduction in revenue from dial-up internet traffic) as well as broadband-based VoIP services.
Another concern (in both developed and developing countries) is the substantial funding requirements to support the costly migration to NGN, especially when viewed against the backdrop of the falling call revenue noted above. Some analysts are suggesting that fixed and mobile market participants have little choice but to invest in new technologies in order to reduce costs and position themselves in a converged environment. Indeed, operators are already beginning to offer portfolios of services, with different combinations of low-cost voice (including mobile), internet access and audiovisual content, to attract and retain customers.
With such technological, commercial, revenue erosion and investment funding developments, cross-subsidy practices to support universal service programs may become increasingly unsustainable. Indeed, an increasing number of countries are already considering the use of universal service funds (USFs) in order that the burden of universal service obligations (USOs) can be shared more equitably and flexibly among market participants. But as competition erodes the revenue base of telecommunications operators, especially but not just fixed line incumbents, USFs, too, are under pressure.
In view of such developments, questions are being asked as to whether an approach to USOs that was framed for a voice focused, circuit-switched, legacy network is still the appropriate approach. Indeed, in view of the significant competitive, technological and service changes taking place, there is increasing recognition that a broad review of USOs is required, including their coverage, how they are financed, and who is to be responsible for providing them.
Previous work
Some countries and international organisations have already begun rethinking the nature and extent of universal service, for example Australia (Australian Department of Communications, Information Technology and the Arts, 2004); Japan (MIC, 2006); the USA (United States Congressional Budget Office, 2005); the UK (Ofcom, 2005); the European Commission (EC, 2005); the OECD (OECD, 2006); and the International Telecommunication Union (Xavier, 2006).
1.2. Objective of paper
The objective of this paper is to consider the implications of an IP-enabled NGN environment for universal service. The paper does not dwell on the universal service problems faced in particular countries due to specific USO mechanisms used in those countries (e.g. in the USA or in the EU member states). Rather the paper is concerned with the impact of competition, technological developments and convergence on USO arrangements of concern to all countries (developed as well as developing). More specifically, the objectives of this paper are to:
- Identify problems that are developing for current approaches used to deliver USOs resulting from developments in competition, new technology (e.g. wireless, NGN, including VoIP) and convergence.
- Consider what universal service implies in a broadband-based IP-enabled NGN environment; should present USOs be maintained in an environment with many competing technologies?; should universal service be focussed on access to infrastructure rather than on access to services (e.g. voice)?
- Identify developments in NGN that are increasing the importance of broadband access; should the scope of universal service obligations be “upgraded” to include broadband?
- Consider initiatives to promote availability, affordability and accessibility of telecommunications in an IP-enabled NGN environment.
- Consider issues pertaining to the funding of universal service in an NGN environment.
- Stimulate thinking about the policies required for universal service in an IP-enabled NGN environment.
1.3. Structure of paper
Following this introduction, section 2 considers the essential features of NGN, and the implications of IP-enabled NGN for developing countries. Section 3 examines the implications of NGN on present USOs, such as voice, quality of service, directory assistance and public payphones. It also examines how “affordability” and “accessibility” might be addressed in an NGN environment. Section 4 examines the meaning of “functional internet access” (that is part of the EU's Universal Service Directive) since in an NGN environment, broadband will be required for access to the full range of NGN services. This raises the question about whether “functional internet access” in an NGN environment means broadband. The answer arrived at is “no”, or at least “not yet” at this relatively early stage of broadband development. But broadband availability and take-up is expanding rapidly (at least in developed countries) so that regular systematic reviews of this issue are warranted. Section 5 adopts “less constrained” rethinking about USOs in an NGN environment where market forces are unleashed with effects monitored and assistance provided to support affordability and accessibility where needed. Section 6 appraises the funding options for USOs in an NGN environment concluding that there should be a shift towards government funding. Finally, Section 7 concludes the paper.
2. USOs in an NGN environment
2.1. Essential features of NGN
The PSTN was designed to carry only voice. As demand for data communications developed, the incumbents built new “overlay networks” designed specifically to carry data traffic. And as network technology developed, the number of networks too multiplied, with many operators running several different network platforms (ATM, IP, Frame Relay, ISDN, PSTN, X.25 etc.). This multi-network approach resulted in management complexity, operational inefficiency, reduced economies of scale, high cost maintenance issues, and duplication of capital expenditure. The attraction of moving to an NGN from such a multi-network system is that it is expected to result in one single network platform capable of supporting all traffic types with lower costs of providing advanced services such as VoIP, broadband and multimedia applications. This is expected, in turn, to lead to reduced charges and greater innovation in services for consumers.
There are two types of network deployment currently referred to as NGN. The first is the deployment of fibre into the local loop, either to the incumbent's street cabinet (± max 1 km from the customer premises) in conjunction with VDSL (2) deployment or the deployment of fibre all the way to customer premises (typically apartment blocks rather than individual houses). These have been referred to as “access NGNs”. This access evolution is being driven primarily by strong demand for increasing bandwidth to support a growing variety of multimedia services. One of the drivers is triple-play services and, in particular, home entertainment, including television services, gaming and video on demand. Video can also be embedded in other applications as, for instance, in video telephony or video conferencing.
The second type of network deployment is the replacement of legacy transmission and switching equipment by IP technology in the core, or backbone, network. This involves changing telephony switches and installing routers and VoIP equipment. These have been referred to as “core NGNs”. A next generation core is in substance a converged IP infrastructure capable of carrying voice, video and data services over the same physical network – in essence the evolution from a “one network-one service” approach to a “one network-many services” one.
IP-enabled NGN may permit a clear separation between network facilities and services (such as data, pictures, audio, video and private and public voice communication). The access plane provides the infrastructure, i.e. the access network, between the end-user and the transport network. This access plane may be wireless or wireline, and it can be based on different transmission media, e.g. copper wires, cable television and fibre optic. Technologies in the access plane can be circuit-switched or packet-switched. The access network is connected to network nodes at the edge of the backbone (core) network.
Layering
The layered network model implicit in an IP network is depicted in Figure 1. As Figure 1 illustrates, IP services depend on a physical infrastructure layer. For instance, VoIP service providers cannot deliver their services without infrastructure provided by facility-based carriers. And since access to the full range of NGN services will be possible only with broadband internet access, broadband will be even more important.
Since the technological means to achieve higher bandwidth can occur via several access technologies, this could enable platform competition in the access network although it is still too early to tell if this will occur. Since the emerging technologies will differ in capabilities and costs, “platform-based competition” can be asymmetric in nature with different platforms offering varying service capabilities and levels of competitiveness depending on demographic and other factors. Thus, within a specific marketplace, competition may favour only a few or even a single platform and, similarly, only a few or even a single NGN provider. In these circumstances, the effect of such asymmetric platforms on the competitive environment would appear to be somewhat uncertain (ACCC, 2005).
2.2. Migration from PSTN to NGN
The expected migration from PSTN to NGN could itself raise universal service concerns. For instance, the transition from PSTN to NGN is unlikely to take place evenly across customer groups or geographical areas. The more profitable customers are likely to be the earlier movers to NGN networks. As traffic migrates to IP networks there will be fewer customers generating PSTN revenue from voice service. Customers remaining on the old network are likely to be clustered in poorer locations and demographic groups. Moreover, the migration from PSTN networks could increase average per line costs of existing networks and lead to a deterioration in quality of service. At some stage in the future, the PSTN could become uneconomic to maintain and their closure would become a possibility (unless obliged to remain in operation due to a USO). Indeed, closure of PSTN legacy networks would be a distinct prospect since the cost reduction benefits from migration to an NGN network can only be appropriated by an operator that is not burdened by the costs of also having to continue to maintain the PSTN network. As a result, the transition to NGN could raise significant universal service issues. There may be pro-competition reasons why the regulator may force the incumbent to continue operating the PSTN at least for some time (ACCC, 2005). But even where the PSTN is not permitted to be “switched–off” there could be concerns that the migration to NGN may create a “digital divide” in regard to the quality and range of services between those with NGN access and those still using the PSTN.
2.3. Implications of IP-enabled NGN for developing countries
Developing countries that are able to develop NGN infrastructure will be able to close the technology gap with developed countries. For instance, the Indian telecommunications regulator considers that:
In the Indian context, NGN offers scope for meeting an important national objective of rural connectivity. It may be possible that with optimal network planning and innovative applications, NGN access could provide affordable converged services (multimedia including voice, e-education, e-employment, e-health, e-governance) in small towns and rural areas at lower costs (TRAI, 2006, p. 23).
However, as Figure 1 indicates, the availability of IP-enabled infrastructure is a precondition for the provision of NGN services, including VoIP. And the development of this infrastructure would be challenging for developing countries, especially those where the majority of the population has poor or no access to telecommunications services. This may raise concerns about whether the transition from PSTN to NGN could widen the gap between (developed and developing) countries as well as within a country. Deploying NGN infrastructure to sparsely populated rural/remote areas of a developing country is likely to be more costly and less commercially attractive than in densely populated areas. But if NGN migration takes place only (or much faster) in urban areas of a country, it may widen the technology gap between urban and rural instead of bridging it.
On the other hand, there is a more optimistic view that the dim prospects for commercially viable rural telecommunications service may be exaggerated, even for developing countries. Proponents of this view suggest that there is increasing evidence that untapped rural and remote markets can be surprisingly vibrant, especially where appropriate regulatory conditions exist. For example, in India, the Telecommunication Regulatory Authority has pointed to the popularity of cable TV in India (which has connections far in excess of fixed line telephones) as an indication that even poor people in rural and remote areas will manifest a capacity to pay for a service they value (TRAI, 2006). Thus the development of converged services packages facilitated by NGN, such as triple-play networks, may offer a better value proposition and lower priced access to basic voice communications service than the traditional model of fixed network extensions for traditional telephone service (Melody et al., 2005). In addition, investment in IP-enabled infrastructure could be attractive since, as noted earlier, such infrastructure promises significant potential savings, e.g. in operating costs. Moreover, it may provide new opportunities for extending network access in rural areas in an economically sustainable manner.
Market forces and technological development
The technological change on the horizon is contributing to this less pessimistic view since a “revolution” in rural universal access is seen to be increasingly likely. This revolution will stem from the new suite of wireless technologies such as WiFi and WiMax (which promises telecommunications coverage over a radius of about 50 km) that could provide internet access and voice service cheaply to rural and under-served communities. Such developments can assist in making rural and low-income markets profitable, affordable, and sustainable. However, this also requires an environment in which market forces can facilitate innovation and creative business initiatives. A concern that universal service programs do not impede or distort the entry of new technology and services is particularly legitimate when considering the deployment of NGN. This involves encouraging rather than resisting the erosion of barriers and artificial distinctions among technologies, services and markets.
3. Impact of NGN on present USOS: voice; quality of service; emergency services; directory service; and public payphones
How might the transition to NGN impact on present USOs?
3.1. Voice
Curently USOs commonly require that all can access voice service at acceptable quality of service at an “affordable” price. The entry of VoIP is raising some challenging issues in this regard. While VoIP is still only a small portion of total voice calls, analysts are predicting an increasing switch to VoIP. Indeed, some analysts expect that in time, almost all voice communication will be through VoIP. What are the USO implications of this shift?
In NGN, broadband subscribers can access VoIP at low prices, bundled with subscription. However, those without broadband will be dependent on voice service from PSTN (which is likely to be degraded as migration to NGN proceeds) or wireless, and this could raise USO concerns related to availability or quality of voice service. (In theory, any internet connection could be used, including dial-up, but dial-up is not widely used to provide VoIP services.)
Quality of VoIP service
Voice services can impose more stringent demands on the network than other services. For instance, voice operates within relatively tight bounds of total latency, imposed jitter, signal distortion and data loss parameters, while other services (e.g. data) are more tolerant of various forms of vagaries in network service responses. Thus VoIP services may offer different features and capabilities by comparison with the features of PSTN-based services that consumers have come to expect as standard. For instance, there are problems associated with VoIP that are not present for PSTN calls, including reliable access to emergency calls, interruptions to normal service and reliance on the power supply to maintain the service (ACIF, 2006). Moreover, as with any internet-enabled system, VoIP can suffer from a number of technical problems related to using the internet such as jitter on the line, access difficulties, virus attack, security, etc. Thus there is concern that the widespread adoption of VoIP could result in a degradation in the quality of voice service.
Emergency services
The possibility to make emergency calls and to route the call to the nearest authority (fire department, police, hospitals, etc.) is a mandatory element of telecommunications service in a number of countries. Also caller location information is increasingly becoming a requirement for both fixed and mobile telephony. Because VoIP calls can be made from any internet connection, a major question to be faced is how to provide emergency services with information regarding the caller's location, which must accompany emergency calls. In VoIP it is possible to maintain the positioning and routing information for emergency calls from a fixed location. However, in nomadic use, at the present level of technological development, the caller position information cannot be made available with the emergency call.
The approach taken by a number of countries to address the calls to emergency service is noteworthy:
- USA. In the USA, calls to 911 with traditional phones provide emergency service dispatchers with the caller's number and address. Internet-based phone providers have limited access to the systems connecting those calls to primary emergency lines and location information is not always available. In May 2005, the FCC adopted rules requiring interconnected VoIP providers such as Vonage to provide emergency E911 service.
- Canada. In April 2005, the Canadian Radio-television and Telecommunications Commission decided that VoIP service providers had to offer emergency 911 services comparable to incumbent carriers. In its decision, the CRTC said VoIP service providers must notify customers about any limitations to their emergency services, as well as make sure their subscribers acknowledge they are aware of limitations. The CRTC also requires that all VoIP providers provide ongoing customer notification during service provisioning, by issuing warning stickers to be placed on telephone sets, through any subsequent advertising and billing inserts. The CRTC distinguished between three types of VoIP service that will be required to offer emergency services. These include fixed VoIP service where users can only place a telephone call from the location where their service is being provided, nomadic VoIP where calls can be made from any location that offers internet access, and foreign exchange VoIP service, which allows users in one exchange to receive telephone calls dialled as local calls in another exchange. Fixed providers must offer either enhanced or basic 911 services, while nomadic or foreign exchange must simply provide an interim solution with basic 911 service.
- EU. The European emergency number “112” can now be used free of charge throughout the EU, from fixed or mobile phones. The aim is to enable EU citizens travelling within the EU to call the emergency services using a single number. In the EU, operators are obliged to provide caller location information to emergency services, to the extent that this is technically feasible (EC, 2006). And in March 2005 the EC further stated that, from then on, the Commission would regard the provision of caller location as technically feasible, since it is available in a majority of member states (EC, 2006).
- Hong Kong. In Hong Kong, the Telecommunications Authority concluded that IP Telephony providers supplying the broadband connection to the customer or service providers that assign their customers with numbers from the Hong Kong Numbering Plan should be required to provide customers with free access to emergency services. For other services, provision of free access to emergency services is optional (OFTA at www.ofta.gov.hk).
Could the market be left to determine prices and conditions in the context of quality of service?
A basic rule for regulatory policy in the dynamic NGN environment is to focus more on the market and on consumer choice. This would mean refraining from heavy-handed regulation wherever possible. Not all broadband access is the same. For example, cable, fibre and DSL technologies have significant bandwidth advantages over WiMax, BPL, and VSAT. In addition, cable, DSL, and fibre work best in high population density areas and are less economic otherwise. Thus, in the future, the quality of access, not merely the availability of access, may become a major consideration in setting policy.
Mandating VoIP to have the same quality of service and features (such as caller location information for emergency services) would increase costs and barriers, especially for new entrants and could make it more difficult for them to compete (e.g. against the incumbent). In this view, it might be preferable for regulation to ensure that consumers are aware of the differences in quality and features and also of prices allowing them to choose. Certainly, for markets to work well, consumers need good information and any deficiencies of VoIP should be well understood by users of the service.
Some regulators (e.g. Ofcom) are advocating provision of adequate information on service capabilities so consumers can make “informed choices” (e.g. numbers with more digits for services not meeting minimum requirements). This would enable consumers to trade off quality of service against price. Regulators could encourage the industry to develop self-regulation, including industry codes, to implement consumer protection and education measures i.e. labelling of equipment and services and marketing practices in order to help ensure that consumers are provided with sufficient and clear information about the limitations and capabilities of IP telephony services before contracts are signed. A VoIP service that does not comply with QoS standards, could be labelled a “secondary service” with requirements for full disclosure of the VoIP service limitations to customers by the VoIP service provider. Examples include: service suspension during power outage on customer premises; the need to update location information on relocation; end-to-end quality of service may not be guaranteed where the customer acquires his own broadband connection to access the services; and unavailability of number portability. Where providers are able to use quality as a differentiator in an NGN environment, markets for low quality and for high quality of service could then develop in accord with customers' QoS requirements.
3.2. Directories and directory enquiry services
In many countries, e.g. the EU member states, the universal service obligations in regard to directory information includes:
- the provision of a comprehensive printed directory of subscribers to all end-users free of charge and at least one a year;
- the provision of a comprehensive national and international telephone directory enquiry service to all end-users, including users of public pay telephones; and
- the obligation to keep a record of all subscribers of publicly available telephone services in the member state, including those with fixed and mobile numbers.
Telephone customers are accustomed to ready access to a directory and directory assistance service. Will such services be provided in an IP-enabled NGN environment without the need for regulation?
In some countries, the directory assistance market has been opened up to competition but this competition relies upon information being provided from communication providers that allocate numbers to both small businesses and residential customers. The requirement to collect and pass on data is distinct from the requirement to provide access to directory services. For instance, Ofcom believes that in relation to the former, the requirement should continue to be a necessary part of the USO (Ofcom, 2006a). In regard to the latter, Ofcom points out that the UK market for directory enquiry services was liberalised in 2003 and currently has 120 providers in addition to BT. However, according to the UK National Consumer Council, the result has not been good for consumers, having led to confusion and higher costs that are often not clear to consumers (EC, 2006). In Italy, too, the removal of directory enquiry service from the USO in September 2003 is reported to have led to a doubling of prices, while the complete liberalisation of this sector in October 2005 has not yet produced a significant decrease in prices (EC, 2006).
If considered necessary, regulators could oblige all telecommunications providers, including mobile and VoIP providers, to publish the relevant number/data of their clients, in such directories without any increase in the price charged for it. In Hong Kong, the Telecommunications Authority decided that provision of directory enquiry service and printed directory to customers free of charge should be a mandatory requirement for IP telephony providers supplying the broadband connection to the customer. Such service providers should incorporate the directory information of its customers such as names and telephone numbers into the unified directory database. It would be voluntary for other service providers whether to offer the directory enquiry service and printed directory to their customers (www.ofta.gov.hk).
3.3. Public pay phones
With the widespread availability and enhanced affordability of mobile services, the use of public payphones has seen a dramatic decrease. Accordingly, usage revenue has fallen, increasing the subsidy cost of providing and maintaining public payphones. For example, in Norway, the revenue raised from public payphones was at NOK 300 million (approximately 40 million Euros) in the mid-1990s, by 2000 it had fallen to NOK 154 million (19 million Euros), and by 2004 it was NOK 70 million (9 million Euros), while estimates for 2005 indicate revenue at NOK 42 million (5 million Euros) (Telenor, 2005). In Australia, Telstra disclosed in February 2006 that it had been experiencing a significant fall in revenue from payphones and unveiled plans to reduce the number of its 32,000 pay phones across the country by about 5,000 (www.telstra.com.au).
This has led to the suggestion that a USO provision of a “wide spread national coverage” of public payphones is less necessary, at least in developed countries. In developing countries payphones will continue to provide a key service and remain critical for universal access. But even in developed countries, public pay phones can continue to be important in case of emergency calls when, for instance, using mobile telephony is not possible (e.g. due to no battery, low battery, loss of the handset, temporary or permanent limited coverage, etc.). Moreover, accessibility to public pay phones in locations of “social relevance” (such as hospitals, prisons, places where there is no mobile coverage or where mobile usage is forbidden) has raised arguments that such payphones might be left within the scope of a USO. In addition, high roaming charges and problems with networks make public pay phones a useful facility for those travelling abroad.
If operation of payphones were no longer a USO, it is likely that many loss-making ones would be removed. A particular concern is that these payphones may well be the ones located in rural and remote areas that are probably amongst the most needed. Rather than discontinuing pay phones, ways might be sought to limit the financial costs of maintaining pay phones. For instance in Germany, Deutsche Telekom has introduced “basic phones” with reduced functionality (including free emergency calls) decided upon in consultation with the public. Moreover, some argue that rather than eliminating public payphones from the USO they could be improved by incorporating public pay phones as part of a telecentre providing various facilities, including internet access. These telecentres could also provide a public access point to the IP-enabled network.
Alternatively a public pay phone box that is not covering its costs might be subjected to a predetermined review process, and then allowed to be removed or replaced, e.g. with a much simpler, cashless, emergency only call box, which would be easier to maintain and less of a target for vandalism. For instance, in the UK, the universal service providers have to publish plans to remove the last remaining public pay phone from a particular local call area then consult on those plans for 42 days during which time local public bodies can veto a proposed removal (Ofcom, 2006b).
Play or pay
Another alternative is for the government (or local government) to pay a subsidy for the provision and maintenance of payphones on the basis of a competitive tender. Yet another alternative is to give new entrants the option of providing their own USO services, to a value equivalent to that of their former (or prospective) USO contributions. New entrant operators might consider it more advantageous to subsidise loss making payphones if they could do this directly and advertise that they were doing this (in the public payphones they provide) rather than simply be required to subsidise an incumbent's loss making payphones (with the incumbents often getting all the credit). (This rationale would apply equally to allowing new entrants to themselves provide other types of USO services, e.g. voice and data through wireless or satellite technology, rather than be compelled to contribute to a USF that then subsidises the USO provider – often the incumbent.)
3.4. Supporting “affordability” and “accessibility” in a NGN environment
How will the migration from the PSTN to NGNs affect the affordability and accessibility provisions of the USO? To what extent will the market provide solutions to affordability and accessibility and with what implications, if any, for regulation? If “availability” is largely addressed (or at least prospectively so) what could be needed in an NGN world are specifically targeted schemes to support “affordability” and “accessibility” for the needy. Examples of such schemes are already in use and could be extended in an NGN environment.
Assisting “affordability” and “accessibility” through targeted programs
Some countries are already using targeted programs on a small scale, such as Lifeline and Link-up in the USA, and so-called “low user” schemes in the UK. Some schemes offer a concession on certain charges to eligible old-age, disadvantaged, disabled or low income consumers for basic telecommunications services. Discounts are offered in respect to connection charges, monthly access charges and usage charges so that the rate of growth of a “lower quartile bill” is contained. In the UK, the universal service provider is required to offer special services to customers with disabilities including text relay (that translates voice into text) for people who are deaf or hard-of-hearing, special format telephone bills for people who are blind or partially sighted, and a priority fault repair service.
In the EU, designated undertakings may also be required to provide tariff options or packages to consumers that differ from those provided under normal commercial conditions, in particular to ensure that those on low incomes or with special social needs are not prevented from accessing or using the publicly available telephone service. However, concern has been raised that people entitled to these tariff packages are not necessarily aware they exist.
Services for the disabled
It is likely that some of the new communication services made possible by NGN will improve the ability of disabled users to communicate in an effective way, choosing the service which best suits them. For example, text relay services can be provided as instant messaging service; speech to text conversion by unified messaging; conversation based sign languages by means of webcams and videophones; public pay phones could be built to accommodate wheel-chair users; the hard-of-hearing could be assisted through the use of amplification systems; and the sight impaired by installing special keyboards.
The open specification of new, IP-based networks provides a significant opportunity for providers to design and deliver special services independent of the usual telecommunications operator product development process. New networks could provide an opportunity for service providers to adopt a “design for all” approach and ensure that services are accessible to all users from the outset. Thus, the needs of disabled users could be accommodated early in the design process so as to reduce the cost of providing special services later on. In the UK, a number of respondents to Ofcom's (2004) “Strategic review of telecommunications: phase 2 consultation” argued that Ofcom should extend USOs to telecommunications equipment as well as services, including requirements that the needs of disabled people be built into NGN from the design stage.
However, new technologies could also bring challenges for users with disabilities. In the European Union, one of the sectors targeted by the EU's i2010 initiative are people with disabilities, estimated to constitute about 15 per cent of the European working population (EC, 2006). This is considered to be a significant proportion of the population, making it important for products and services to be accessible to these end-users. Action is encouraged rather than mandated.
4. “Functional internet access” in an IP-enabled NGN environment
4.1. Could USOs cover only access to telecommunications infrastructure rather than services?
The present system of USOs has its origins in the time when both access and services were only available from a single supplier. With NGN there is a separation between networks and services. The development of IP technology allows delivery of a number of previously separate communications services using only one transmission platform. Thus, internet-based communications could allow consumers greater choice in choosing service providers. In an NGN scenario of greatly increased availability and affordability of (cheap) voice service (now only one of many applications provided on networks) should the focus of universal service move towards provision of an affordable broadband access link
There is considerable disagreement the answers to these questions.
One view (e.g. put by KPN in its response to the EU Consultation on universal service) is that users will not be interested in access only since for users it is the services (communication, content, transactions) they can use. In this view, access to infrastructure as such has no value for the user. In its submission to the EC consultation on universal service (EC, 2005), Vodafone (2005, p. 4) argues that:
The focus should be on services, because social exclusion derives from lack of access to services rather than from not having access to specific communications infrastructure. For example, the satisfaction of the demand for real-time, two-way communication between individuals can be met by several technologies other than simply voice communications over a fixed line. In a data environment, it is even more important to retain the focus on the services that are considered essential for participation in society, rather than any particular delivery infrastructure. There is a high risk of market distortion if there were a bias through intervention towards a particular type of infrastructure access.
On the other hand, the UK Department of Trade and Industry declared in its submission to the EC consultation on universal service (EC, 2006) that:
We are somewhat attracted to the idea of moving towards an environment where it is the access (to perhaps broadband connection) that is specified as an obligation rather than a specific service.
Ofcom concurs that there may be a case for separating the two elements but raises a number of issues that are pertinent to this decision (EC, 2006):
- It may be premature to conclude that the provision of services will de facto ensure the availability and affordability of services to the standard required by the USO.
- An assessment of affordability would need to look beyond the cost of the services themselves and include the cost of devices required to connect to the service. This may be an additional barrier for some vulnerable customers.
Ofcom's position seems sensible since it does seem too early to ascertain the extent to which the availability, affordability and accessibility of services defined within the scope of universal service will be met only by access to infrastructure in an NGN environment. A related issue is the question of broadband access.
4.2. Should broadband be part of USOs?
Broadband is a precondition for access to the full range of NGN services, including high quality VoIP services. Thus, broadband penetration is likely to be a key issue as many converged services enabled by NGN can be delivered only through broadband. Thus the question arises as to whether the scope of the USO should be upgraded to include broadband?
The EU has already included data access (albeit at low speeds) in the definition of USO with a “functional internet access” provision in its current USO Directive. Does “functional access” in an NGN environment necessitate an upgrade to broadband access?There seems a widespread view that a USO for consumer broadband services is not warranted at the present time
Moreover, a conclusion that broadband should not be part of USOs does not necessarily argue that the government should not be supporting broadband deployment and take-up through various demand and supply side measures. Only that this should not be done through a “blunt”, blanket, USO approach that could distort competition and investment incentives (Xavier, 2003). This is not an idiosyncratic approach since the EU, for instance, allows government assistance initiatives outside the scope of the universal service definition as long as they are not funded out of a USF (that telecommunications operators are obliged to contribute to).
5. Less constrained rethinking about USOs in an NGN environment
So far, the consideration of USOs in NGN has been constrained, with the concern being the ability to preserve current USO provisions. Technological change is increasing the potential for universal “availability” of telecommunications services over the next decade, including in rural/remote areas. With such changes ahead, it may be time to think in a less constrained way about when it would be time to unleash market forces, monitor what happens and address equity concerns of “affordability” and “accessibility” through targeted cost-effective schemes? This approach would be consistent with “forbearance” of regulation in a dynamic, competitive and convergent communications sector. It would be consistent with increasing incentives for investment. And such rethinking about USOs would recognise the impact of USOs in distorting competition and discouraging entry, including the fact that subsidisation programs:
- can restrict or distort competition and availability/choice of technology in rural/remote areas;
- can discourage potential market entrants facing the prospect of competition against a subsidised provider (offering services at prices below costs);
- can suppress market signals and, accordingly, market development
- can, where they are not means-tested (as in the case of USOs), advantage those not needing subsidies;
- with the “one size fits all” approach taken by USOs, can be wasteful (since consumers have widely differing preferences); and
- raise costs for other consumers.
The recognition of these inefficiencies of USO programs may serve as a reminder that regulation should be the minimum necessary and suggest some basic rules for USOs in an NGN environment:
- Primary reliance should be placed on market-based approaches; USOs policy should not lead the market but should monitor and address concerns arising from market outcomes.
- Subsidies should be kept to the minimum necessary.
- Regulation needs to facilitate changes, not retard them; should not impede entry of new technologies (technology-neutral); should provide a conducive environment for infrastructure; needs to be wary of being heavy-handed and be disposed towards “light-touch” regulation.
The results of such an approach should be closely monitored, with problems of affordability and accessibility (if they occur) addressable through targeted subsidies. It is important that in a multi-platform NGN environment, consumers be able to choose the service provider and technology (including broadband) suitable to their needs.
6. Funding USOs in an NGN environment
Apart from operator contributions, universal service could be funded from a range of sources, including:
- a tax on each telephone number;
- a connections-based system;
- a surcharge on incoming international calls;
- from general taxation revenue;
- contribution from local government and other government departments;
- contribution from spectrum auctions (3G), spectrum pricing and privatisation; and
- donor funds (for developing countries).
6.1. A tax on each telephone number
A fixed fee for each telephone number assigned, whether fixed line or wireless, would be relatively easy to administer, readily understood by consumers, promote telephone number conservation and has the ability to raise predictable amounts of revenue through relatively small imposts on each number. Such a fixed fee would make the funds supporting universal service competitively and technology neutral since all providers using numbers from a national numbering plan can be included (including wireless, cable and VoIP providers).
Moreover, a small, broad-based tax is unlikely to distort economic decisions significantly. The tax may create incentives to adopt technologies that bypass the tax, but such incentives are likely to be small if the tax is a small one. However, some services using telephone numbers – for instance, paging devices, or low-cost pre-paid cellular phones – generate relatively modest revenue per month; a phone number fee could heavily tax such services and cause people to shift away from them. Another concern about a flat fee on each number is that it is a “regressive tax” since low users and the poor would pay as much as high users and the rich. But this concern might be addressed by exempting the poor, including “lifeline users” etc., from paying the numbers tax. There are other concerns such as the potential for bypass. A telephone number is not essential for voice communications as Skype and some other VoIP operators are demonstrating. Such bypass could become a significant problem if there is increased use of this system.
6.2. A connections-based system
A “connections-based” system is one through which any connection to a network, whether it is a data or telephone network, is taxed, with revenue channelled towards universal service programs. While a connections/line approach is similar to a numbers-based system in that contributions to universal service would be based on the number of connections, it raises a number of questions regarding the definition of a connection, including simple residential telephone lines and high-capacity business lines with dozens/hundreds of voice-grade equivalents.
Another option is some sort of hybrid approach that seeks to combine a number of schemes, e.g. a combination of a levy on each number and usage charge. A numbers-based plan could miss the revenue from high-capacity data pipes without some sort of connections allowance. Alternatively, funds could be raised through a supplementary levy on consumers' (residential and business) bills. This is done in other sectors such as the airline industry where “air passenger duty” (UK) or airport taxes (Australia) are levied on each ticket with funds raised used to support the airline sector. Making the contribution to the USO fund transparent makes it easier for the public to assess the costs of universal service.
6.3. A surcharge on incoming international calls
In April 2005, Jamaica issued an order that required its domestic telecommunications carriers to impose a surcharge on all incoming international calls: three cents per minute for terminating on a fixed network and two cents per minute for terminating on a mobile network (USTR, 2006). Jamaica claims that the purpose of this surcharge is to fund its e-learning Project – a universal service program for building broadband access for schools and libraries in Jamaica.
6.4. Financing through general taxation revenue
Economic analysis provides support for financing USO through general taxation revenue on the basis that it is likely to be less distortive (United States Congressional Budget Office, 2005). The economic (“excess burden”) cost of raising a dollar in general revenues (in terms of distortions to the supply and use of factors of production) is generally less than the economic cost of raising a dollar from sector-specific taxes, which tend to distort consumers' choices by affecting the prices of goods and services. Indeed, a number of economic analysts have recommended that future universal service support should come from general tax revenues (e.g. Hazlett et al., 2004).
In Australia, the government-appointed Independent Regional Telecommunications Inquiry recommended (Rec 9.5) that the government should provide funding for future service improvements in regional, rural and remote Australia, rather than imposing financial obligations on the telecommunications industry:
The Inquiry does not consider industry subsidisation of future sharing arrangements is appropriate. It considers industry funding to meet the costs of non-commercial telecommunications needs would impose a significant financial burden on the industry, and would negatively affect investment incentives, not just in regional Australia but nationally. Ultimately, it would also impact negatively on prices paid by consumers for telecommunications services. Government funding is preferred by the Inquiry (Australian Regional Telecommunications Inquiry, 2002, p. 249).
And a little later the report states:
It is appropriate for Government to directly fund its social and economic telecommunications policy objectives, as it does other policy priorities (Australian Regional Telecommunications Inquiry, 2002, p. 250).
The Australian Government's response was to “accept the principle that support for non-commercial service improvements in regional Australia should be provided transparently by government, and should aim to promote competition and minimise market distortions”
A major concern regarding government funding is that the predictability of subsidy amounts is an important characteristic of a subsidy and funding scheme and that the competing demands for government funding could make such funding more uncertain in the longer term and susceptible to change. However, there is a strong argument that this is as it should be. That is, that continued support for telecommunications subsidies should be regularly justified against such competing demands.
6.5. Contribution from local government and other government departments
Support for telecommunications is widely rationalised on the grounds that it can help deliver improved service to education, health, agriculture, e-government and telecentres in rural and remote regions. This suggests that a number of government agencies might contribute to a USF to support telecommunications access. Allocating responsibility for delivering program outcomes to key spending ministries can have other benefits such as a shared sense of ownership across all participating ministries and local government. In addition, telecommunications operators can form partnerships with local government agencies to help ensure the initiative contributes to local economic development.
6.6 Contribution from spectrum auctions (3G), spectrum pricing and privatisation
Some of the proceeds from telecommunications licence fees, including spectrum pricing fees, could be contributed to a USF. Also, part of the proceeds of spectrum auctions might be allocated to USO purposes. In addition, a proportion of the proceeds of privatisation of telecommunications operators could be allocated to support universal access and universal service programs. For instance, in Australia, 5 per cent of the proceeds from privatising the second tranche of Telstra's shares was allocated to improving conditions in rural areas based on the rationale that this constitutes an equitable sharing. In a submission to the DCITA's review of universal service
For pragmatic reasons, including political realities, it might be sensible to draw on a combination of funding sources, especially in the short term. However, in the longer term, there seems a strong case for universal service to be funded from general taxation revenue. Since it is the political process that will decide on what the warranted level of universal service will be, it seems appropriate that the case for continued funding of universal service programs be assessed against arguments in favour of competing government expenditure programs such as those for health, education and housing.
While political gains flowing from universal service programs can be preserved/gained at the expense of operators and/or consumers, restraint over universal service (needed to stimulate innovation and cost-effective USO programs) is less likely. Government funding would link decisions concerning the nature and scope of universal service more closely with financial responsibility for such decisions. This would help prevail against excessive growth by installing in-built incentives to cap (or at least to restrain) political disposition for universal service expenditure.
7. Conclusion
With falling revenue (due to increasing competition and declining prices) occurring alongside strong demands for funds to deploy NGNs, telecommunications operators are less able to provide universal service sustained by cross-subsidisation. USFs, too, may be under pressure. Also, there is likely to be an uneven migration to NGN since it is likely that NGN will be deployed first in more profitable, densely populated areas and then only later (if at all) in relatively less commercially viable rural/remote areas. This could raise concerns that problems of universal service and the “digital divide” could worsen, especially for some developing countries still struggling to expand infrastructure and penetration of PSTN and mobile service.
The first question addressed in this paper is whether present USOs can be maintained in an NGN environment? The conclusion reached is that they can be maintained. However, certain features such as the ability to make emergency calls with caller location information may not be available with VoIP. Such concerns could be addressed by a regulator mandating an equivalent quality of service. The problem is that this could raise costs and barriers to entry for service providers using new technology. An alternative approach could be to ensure that customers are adequately informed about the differences in quality and to allow them to decide.
Another question addressed is whether and, if so, how the scope of USOs might need to change? Access to the full range of NGN services, including VoIP, requires broadband access. Therefore, should the scope of USOs be “upgraded” to include broadband? The conclusion reached is that at this relatively early stage of broadband development and take up, there are sound reasons to be wary of using a “blunt” standardised USO approach. But broadband availability and take-up is expanding rapidly (at least in developed countries) so that regular systematic reviews of this issue are warranted (especially when NGN have become pervasive) since universal service is an evolving concept.
Looking further ahead, there is also need to think in a less constrained way about universal service for an NGN environment. In short, in an NGN environment, to what extent can unshackled market forces be depended upon to resolve USOs concerns? Such rethinking may serve as a reminder that regulation should be the minimum necessary and that USOs policy should not lead the market but address concerns arising from market outcomes.
Finally, funding. The paper argues in favour of increased funding from general taxation revenue that would link decisions concerning the nature and scope of universal service more closely with financial responsibility for such decisions. This would prevail against excessive growth by installing in-built incentives to restrain political disposition for widening universal service expenditure. Certainly, while political gains flowing from universal service programs can be gained at the expense of operators and/or consumers, restraint over universal service (needed to stimulate innovation, best practice and cost-effective USO programs and to minimise the distortions that arise from excessive USO programs) is less likely.
Figure 1Architectural layering in NGN
References
Australian Communications Industry Forum (ACIF) (2004), Policy and Regulatory Considerations for New and Emerging Services, final report, Australian Communications Industry Forum, Canberra, July, .
Australian Communications Industry Forum (ACIF) (2006), “Industry discussion paper QoS-based VoIP service interconnectivity”, ACIF, Canberra, March, .
Australian Competition & Consumer Commission (ACCC) (2005), “A strategic review of the regulation of fixed network services”, an ACCC discussion paper, Australian Competition & Consumer Commission, Canberra, December, .
Australian Department of Communications, Information Technology and the Arts (2004), Review of the Universal Service Obligation and Customer Service Guarantee, Australian Department of Communications, Information Technology and the Arts, Canberra, .
Australian Regional Telecommunications Inquiry (2002), Connecting Regional Australia, Canberra, November, .
European Commission (EC) (2005), On the Review of the Scope of Universal Service in Accordance with Article 15 of the Directive 2002/22/EC, COM(2005)203, EC, Brussels, 24 May, .
European Commission (EC) (2006), European Electronic Communications Regulation and Markets 2005, (11th Report) COM(2006)68 final, EC, Brussels, .
Hazlett, T.W., Bazelon, C., Rutledge, J., Hewitt, D. (2004), Sending the Right Signals: Promoting Competition through Telecommunications Reform, a report to the US Chamber of Commerce, US Chamber of Commerce, Washington, DC, 22 September, .
Japan Ministry for Information and Communications (MIC) (2006), Agenda for Consideration Concerning a Framework for Competition Rules to Address Progress in the Move to IP, MIC, Tokyo, February, .
Melody, W., Sutherland, E., Reza, T., “Convergence, I.P. (2005), “Convergence, IP telephony and telecom regulation: challenges and opportunities for network development, with particular reference to India.”, paper prepared for infoDev based upon the issues examined at a Workshop on Convergence, VoIP and Regulation, sponsored by infoDev in association with Telecommunications Regulatory of India (TRAI), New Delhi, 11 March, .
Organisation for Economic Co-operation & Development (OECD) (2006), Rethinking Universal Service for an NGN Environment, OECD, Paris, .
Telecom Regulatory Authority of India (TRAI) (2006), “Consultation paper on issues pertaining to next generation networks (NGN)”, Consultation Paper No. 2/2006, 12 January, .
Telenor (2005), Submission to European Commission's Consultation On the Review of the Scope of Universal Service in accordance with Article 15 of the Directive 2002/22/EC, COM(2005)203, 24 May 2005, Telenor, Fornebu, July, .
UK Office of Communications (Ofcom) (2004), “Strategic review telecommunications, phase 2, consultation document 2004. annex K, ‘Universal service: future scope and funding’”, available at: www.ofcom.org.uk, .
UK Office of Communications (Ofcom) (2005), “Review of the universal service obligation”, 10 January, available at: www.ofcom.org.uk, .
UK Office of Communications (Ofcom) (2006a), “Regulation of VoIP services, statement and further consultation”, February, available at: www.ofcom.org.uk/consult/condocs/voipregulation/voipregulation.pdf, .
UK Office of Communications (Ofcom) (2006b), “Statement ‘review of the universal service obligation’”, 14 March, available at: www.ofcom.org.uk, .
United States Congressional Budget Office (2005), Financing Universal Telephone Service, United States Congressional Budget Office, Washington, DC, March, .
United States Trade Representative (USTR) (2006), Results of the 2006 Section 1377 Review of Telecommunications Trade Agreements, United States Trade Representative, Washington, DC, 6 April, .
Vodafone (2005), Submission to European Commission's Consultation On the Review of the Scope of Universal Service in accordance with Article 15 of the Directive 2002/22/EC, COM(2005)203, Vodafone, Newbury, 24 May, .
Xavier, P. (2003), "Should broadband be part of universal service obligations?", info, Vol. 5 No.1, May, pp.8-25.
Xavier, P. (2006), “What rules for universal service in an IP-enabled NGN environment?”, paper presented at the ITU Workshop on Next Generation Networks, Geneva, 23-24 March 2006, available at: www.itu.int/osg/spu/ngn/documents/Papers/Xavier-060323-Fin-v1.pdf, .
Further Reading
Australian Communications Authority (2004), “Regulatory issues associated with provision of voice services using internet protocol in Australia”, discussion paper, Australian Communications Authority, Canberra, October, .
European Telecommunications Platform (2006), On the Technology, Business Models and Regulatory Aspects of NGN, European Telecommunications Platform, Brussels, 17 January, .
UK Office of Communications (Ofcom) (2004), “Strategic review of telecommunications, phase 1, consultation document. 2004”, available at: www.ofcom.org.uk, .
UK Office of Communications (Ofcom) (2005), “Next generation networks: further consultation”, 30 June, available at: www.ofcom.org.uk/consult/condocs/nxgnfc/, .
Corresponding author
Patrick Xavier can be contacted at: pxavier@swin.edu.au