R Padayachie: Government Technology Conference

3 November 2006

The road ahead - regulatory challenges in a converged or converging
environment

Director of proceedings and executives of SITA,
Delegates present,

My input on the topic you have chosen may be separated into two broad
areas:

1. defining what regulation will look like in a converged or converging
market
2. enforcement and monitoring and enhancing regulatory capacity and
credibility.

1. Defining regulation in a converging market

This task requires an assumption of a general sector move towards Internet
Protocol (IP) networks and services or so called "Next-Generation Network (NGN)
and services". NGN can be understood as broadband networks that allow
integrated data, voice and video services through the deployment of IP. NGN is
an umbrella term for the infrastructure that enables advanced new services that
mobile and fixed network operators have started to offer while continuing to
support existing services.

There is a growing body of literature on the shape of the challenges and
regulatory questions that NGN regulation poses. Central to this is
understanding the technical changes and issues that will inform the regulatory
challenges as well as reframing the debate on regulation in an IP network and
services framework; the regulatory issues are not as much about policy
direction but more about ensuring that policy and regulation remove inhibiting
factors affecting IP networks and services from a regulatory point of view.

There are three basic assumptions underpinning the NGN challenge:

a. The development of IP and NGN will increasingly undermine currently held
rationales for asymmetrical approaches to regulation across sectors.
b. Technological development will always outpace regulators and the advent of
IP networks and services is no exception.
c. The development of NGNs and an enabling regulatory framework is inextricably
linked to a country's national broadband policy. It is countries that have
extensive broadband penetration and usage such as South Korea and Iceland,
where next generation networks and services have been able to flourish and
where the critical mass necessary for the network effects associated with
economic growth are most likely to be realised.

The International Telecommunications Union (ITU) Trends Report for 2006
shows a sobering reality of broadband penetration in Africa such that the
continent accounts for less than 0,1 percent of subscribers worldwide. This can
be partially explained by the fact that few African countries including my own
have developed coherent broadband policies. Few African countries have yet to
launch high speed Internet services at all and where broadband development is
occurring, it is usually led by the monopoly Telco, who still controls access
to its network for downstream competitors. By contrast the highest broadband
penetration rate reported by the most recent Organisation for Economic
Co-operation and Development (OECD) broadband study is in Iceland at 26,7
subscribers per 100 inhabitants. I cite these figures merely to assist in
setting the broader policy challenges facing African regulators and policy
makers in the transition to NGNs and the diffusion of IP based services in
particular.

Broadband subscribers by region, 2004:

* Asia-Pacific � 41,6 percent
* Europe � 27,8 percent
* Latin America and Caribbean � 2,4 percent
* North America � 27,9 percent
* Africa � 0,1 percent
* Arab States � 0,2 percent

(Source: ITU world telecommunication indicators database)

Factors affecting IP network and service development

The numerous challenges for regulation in a converged era can essentially be
split into policy and structural concerns and implementation issues. The former
can be characterised as legacy regulatory questions that pertain to
competition, market structure and policy absence or failure, usually with
respect to broadband policy as well as regulatory capacity or ability to keep
pace coupled with the ever present affordability problems. Structural market
issues relating either to continued monopoly power in certain markets or as a
result of vertical integration across several market segments will need to be
urgently resolved by regulators to attract the necessary private capital
required for investment in NGNs. Ironically, the countries that have very under
developed broadband networks may however have an opportunity to leapfrog
orthodox reform methods and adopt newer and more appropriate strategies for
rapidly deployable, lower cost and higher efficiency networks. But these
opportunities may be constrained in terms of government policies and
obligations to investors and existing operators.

Affordability remains a critical issue. The importance of a nexus between
broadband policy and IP network development cannot be overstated. All
communication services in Africa generally tend to be high or higher than in
those countries with more mature markets. The very fact that new IP services
can be run over NGNs and that there may be savings from IP network topologies,
does not change the economics of supply and demand. Generally in low
competition markets prices are higher and usage is more limited inhibiting
further economies of scale that would drive prices down. The big policy,
structural questions need to be addressed to ensure that they facilitate
broadband growth, outside of silo services, like Asymmetric Digital Subscriber
Line (ADSL) which are then protected through legacy rate regimes distorted
interconnection frameworks and access restrictions. As such it is worth
restating that in order for consumers to begin to reap the benefits of NGN
services, greater attention needs to be paid to creating competitive
environments. (This has particular relevance in the GovTech forum to issues of
municipal broadband and the responses by municipalities to try and address
broadband communication costs. A trend that is replicated in many
jurisdictions; note Toronto Hydro's latest initiative to use streetlamps to
place infrastructure for a broadband network in the city of Toronto that will
have a commercial element to it).

In regard to competition, in the South Africa context it is important to
note the benefits provided by the licensing approach contemplated in the
Electronic Communications Act (ECA). Ad hoc regulatory intervention is unlikely
to bring down communication prices in South Africa in a material manner unless
one changes the entire cost structure of the business and the sector. This is
not plausible. The speedy introduction of more facilities based competition
will be the best regulator of price for the sector. The ECA envisages a
situation in which the two public switched telephone network (PSTN) providers
can effectively become full scale mobile telecoms operators.

With regard to the implementation issues, a number of challenges arise from
various related questions that need to be answered going forward:

1. In a managed liberalisation model common to so many African countries the
first question affecting IP networks and services is whether to legalise new
services and for countries where already legalised, what aspects of these new
services require regulation? As with all regulation, enforcement questions also
arise.

2. How much regulation is necessary, when is it appropriate and how should
it be effected to best create an enabling environment? For example, should IP
network development be regulated or allowed to evolve organically? Will the
organic model ensure that capital investments are protected, or is regulatory
oversight required? Or will opening markets simply induce appropriate risk
assessed investment?

3. In terms of what to regulate should one for example distinguish voice
including Voice-Over-Internet Protocol (VoIP) as a service still needing to be
treated with a distinct set of policy, legislative and regulatory
provisions?

4. In terms of sequencing, is it more urgent to address legacy competition
issues to prevent entrenching already unfair advantages in the market? Is it
just incumbents that might need regulatory oversight or new entrants as well
and if so, to what degree?

Setting aside for a moment the question of adequately regulating the
dominance of incumbents many of these challenges emanate from an attempt to
regulate new services in a legacy fashion which gives rise to numerous possible
questions.

For example: Is consumer interest regulation required ensuring fair pricing
and quality of service or can regulators abandon this objective if markets are
fully competitive? How should numbers be allocated? Or interconnection
regulated, if at all? Where broadcasting has traditionally attracted a more
interventionist regulatory approach than telecommunications, at least with
respect to content, how should regulators approach Internet Protocol Television
(IPTV)? Can or should we regulate content in a converged "always on" world
where content delivery models are entirely different from what we have known?
Are platform independent approaches to regulation possible? And importantly how
is the public interest to be promoted in the context of this rapidly evolving
IP enabled communications environment?

Breaking down some of those challenges, I offer some of the main ones, but
caution that the list is not exhaustive. We know that IP network and services
are in an evolutionary phase and regulators are beginning to grapple with some
of the questions. No one yet has the answers except that it is imperative that
regulators "get the framework right" which also links to the issues of
credibility, reduction of investment risk and increased certainty.

Licensing

Licensing is a regulatory mechanism used to control access to a market and
to stipulate rights, responsibilities and obligations for licence holders. In
the IP network context, it is clear that countries need to move away from
service and technology specific licensing which restricts the way in which
technology is used and can prevent operators from benefiting from economies of
scope. This requires a shift away from traditional vertically integrated
licensing towards more horizontal licensing that better reflects the technical
and logical separation of the core, access and service layers on NGNs. Where
national policy has not yet matured organically regulators and policy makers
need to consider classes and types of licences to be used and whether this will
require a new set of licensing criteria and conditions to be developed or
whether existing ones will suffice. Simply reforming the law in this regard is
not enough; the actual implementation of the licence terms and conditions must
create the enabling environment the policy seeks to achieve.

In India, the Telecom Regulatory Authority of India's (TRAI's) consultation
has found that PSTN operators appear to be migrating to IP based transport
leaving the access part of the network (historically narrowband) unchanged.
Regulators need to ensure through licensing and other enforcement mechanisms
that the introduction of new networks by incumbents does not have the
unintended consequence of blocking competition from new and smaller players. As
mentioned earlier, however, a facilitative broadband policy environment is
critical for IP based networks to flourish, as full NGN transition is highly
dependent upon the success of alternative technologies such as cable television
(TV) and WiMax. As such access to the local loop through unbundling to avoid
the duplication in the access network is a regulatory necessity.

Numbering

Numbering allocation has historically been aimed at fixed and mobile
traditional networks. In the IP environment with multiple service offerings
most notably VoIP, new market players and VoIP providers must have access to
numbering resources. At the same time numbering, naming and/or addressing
schemes will need to encompass legacy, transitional and NGN services and
associated directory services will need to be developed. This is critical for
users of those services to have any connectivity and communicate with users and
subscribers of any other network. While a number used to indicate geographic
location, VoIP services now allow for "virtual" numbers. This enables persons
living in one area to be assigned the dialling code of another to enable calls
to appear to be "local" in origination. Regulators will have to review their
numbering plans to ensure that it can support new resources while ensuring the
interoperability of new and existing services and emergency services. This may
also necessitate a potential specific number allocation scheme to differentiate
IP telephony from PSTN numbers. Finally, attention will have to be paid to
telephone number mapping (ENUM) developments as its eventual maturation will
require national implementation.

Spectrum assignment

Spectrum challenges emanate from the fact that incumbent operators have
preferential access to frequency spectrum for the provision of fixed,
fixed-mobile and mobile services. Spectrum is limited and many of the key
assignments have already been made.

IP allows alternative technologies like WiMax and high speed broadband
Internet access over wireless connections. As WiMax can be used over long
distances, this technology is an effective last mile solution or alternative
for broadband to the home. Such technology also facilitates a digital divide
bridging opportunity in low density areas where technical and economic factors
make broadband deployment very challenging. It also allows fixed broadband
access in urban areas where copper quality is poor or absent. Policy makers and
regulators must ensure that legacy assignments do not hamper the introduction
of new services and entrants.

In South Africa various new players are seeking to offer WiMax services on a
national scale. Other licensed operators such as the under serviced area
licensees and the new national fixed line entrant licensed in 2005 require
access to the 800 MHz band (822 � 830 MHz - channel 65) for the deployment of
non broadcasting services, essentially, Code Division Multiple Access (CDMA).
The regulator is currently undertaking a public process to assess the viability
of new operators sharing this band for non-broadcasting purposes with the
incumbents in the band. At the same time the process is mindful of wanting to
encourage innovative applications without causing harmful interference to other
services co-existing in the same spectrum.

Consumer issues and protection

As with traditional circuit switched networks, IP networks and NGN migration
is not likely to change the need for regulatory oversight to ensure adequate
consumer protection. Issues that will require attention but are not limited to
quality of service, emergency services, location information, rights and
presence management, portability, operators' liability, privacy and security.
Some specific risks are posed to interoperability by the development of new
services on a competitive and proprietary basis, risks that existing telecoms
with a large share of the Internet access market could degrade quality so that
real time services and applications do not work adequately over the Internet in
trying to retain voice traffic on the PSTN, various risks to privacy, call
tracing for unlawful purposes and fraud.

Some regulators are beginning to grapple with these challenges as they
arise. Others have embarked on a broad consultative process to ensure that the
issues are addressed in advance of full NGN migration.

The Office of Communication (OFCOM), like TRAI have also acknowledged the
importance of a co-regulatory approach with regard to consumer issues, and
British Telecoms (BT) have already undertaken to participate in a new body
which could address these issues on an industry wide basis. Ultimately consumer
issues need to be kept central to any approach to IP network and services
migration. Keeping consumers well informed on the migration process at all
times, allowing for public participation and attempting to meet consumer needs
will assist greatly. In the interim regulators and policy makers must give
detailed attention to the quality issues for Internet access, e.g. ADSL
contention ratios and appropriate Quality of Service (QoS) parameters need to
be defined.

Security risks, privacy and law enforcement

The PSTN and mobile network is a closed network with controlled security and
privacy. IP based VoIP is open architecture where vulnerabilities, threats and
risks for communication security exist in various network elements.

These networks thus pose particular security challenges requiring special
measures. In the current global political climate this may also require a
reconsideration of the arrangements for State lawful surveillance and many
countries make network and services licences conditional on those networks
being surveillance enabled. Possible questions include where and in what form
should interception of communications be undertaken in multi proprietary
standards service architecture with dynamic, indeterminate routing? How should
the costs implications associated with these requirements be dealt with? This
cost implication is more acute in IP based networks, which unlike circuit
switched, do not have a dedicated path through the network. Monitoring and
surveillance on these networks may require additional costs and investment in
special equipment which can also serve as a barrier to entry.

Emergency Services

Emergency priority services which are supported in the PSTN environment are
critical for obvious reasons and cannot be compromised in the new IP or NGN
environment.

The main challenge in the IP space is the nomadic nature of VoIP. There is
no certainty about locational data which is essential to the accurate response
for emergency calls where the person is unable to convey that information.
Geographic numbers convey information and are used for routing such calls to
the nearest "emergency centre". With VoIP there may well be a complete
separation between the provision of the voice service and the transport of the
voice data. There is at least currently no way of conveying location
information about the user calling an emergency service between internet
service providers (ISPs) and VoIP providers. However in the future technical
solutions like that of Intrado are expected to emerge and be able to convey
data about static, mobile and out of area numbers. Regulators will have to
consider the types of provisions that have to be made to ensure accurate
location information.

2. Enforcement and monitoring and enhancing regulatory capacity and
credibility

Ultimately many of the challenges going forward require significant skills
and human capital at the policy, law making and implementation levels. As is
generally the case the role of the regulator and the role of the market have to
be carefully balanced. Regulation must be used to bring certainty but if it is
premature or prescriptive it can pre-empt the role of the market in determining
the shape and form of competition. On the other hand intervention that is too
little or too late, can result in their being no competition at all. Efficient
and effective market development may not in all cases require new policy
interventions or national regulatory intervention at all. Some policies may
just require some modification while others may require new and targeted policy
decisions. As is the case generally with telecommunications reform, the
sequencing and degree of reform is critical.

The persistent problem of attracting and retaining skilled personnel is a
key issue in the context of NGNs. This can only intensify under the pressure to
effectively and efficiently regulate IP networks and NGNs. Many of the existing
personnel within regulatory bodies are accustomed to neat lines between
telecommunications, broadcasting and spectrum management. Convergence has
presented enormous challenges for this silo organisational design.

More than ever, regulatory agencies remain skill incubation centres for the
sector and we all face major skill losses to industry. Rather than delve into
the complex nature of this problem, it may be more worthwhile to highlight the
increased importance of knowing what skills reside in-house and how to
effectively use public consultations and other regulatory resources to top up
information gaps. Even where human resource (HR) constraints are absent some
regulators appear to be hinting at greater co-regulatory approaches with
stakeholders based on the common interest held in addressing the problems and
challenges at hand. Regulators also need to be mindful of the fact that the
full migration to IP based networks and services can create some concern for
operators and service providers in terms of business models and operations. The
regulator needs to ensure that any such transition is as smooth as possible
with clear forward-looking interconnection regime and adequate industry
involvement.

Examples where joint consultations, including joint industry working groups
are under construction include OFCOM which has established "NGN committee";
TRAI in India has also proposed to establish "NGN Expert Committee (NGN eCo)".
This is in no way an abdication of oversight responsibilities but ways to
manage better the complex regulatory challenges NGNs pose. These forms of
co-operation allow the regulator to engage and better understand business
models, drivers and propositions and how best to avoid regulatory induced
failure.

Beyond NGN deployment, it is important for regulators to have watching
briefs of international developments particularly regarding standards and
numbering. There are also limits to national legislation in an environment
where there is separate service, network and access provision.

While this can make regulation much more difficult, it will also lead to the
increasing offering of services from one location to users anywhere in the
world and increases in cross border consumption of services from an operator in
one country to consumers in another. National and even regional legislation
where it exists may be inadequate and may need to be addressed.

Greater co-operation with other regulators and placing these matters on the
agenda of regulatory associations will also be of great importance.

Finally, it is not only the regulator and policy makers that face these
challenges. New challenges exist for incumbent fixed and mobile operators and
new entrants. The latter's challenges are very similar to the ones the
regulator faces but also challenging for the new entrant and ISP is going to be
the degree and intensity of competition on and between layers. IP networks
provide the opportunity for third parties and service providers to develop and
provide value added customer services over the networks owned by other
operators. As such the separate transport, control and application layers also
enable different operators to compete with each other in different layers.

As these layers should be open, competition is likely to be fierce,
providing great consumer welfare benefits and opportunities to innovative
service providers. But a clear and fair, transparent and well implemented ex
ante competition regime will have to be in place for new entrants to be
guaranteed timely entry to the market. Quantum is unpredictable at this stage
but there will also be increased additional costs for compliance with specific
consumer focused issues such as QoS or security requirements for emergency
services and lawful interception.

These may serve to be prohibitive for smaller players. Finally, there is
always concern by providers that a rigid and inappropriate application of
current regulations may limit consumer benefits.

It's a challenging time for mobile and fixed incumbents who see both threats
and benefits to IP network and services migration. There may be savings and
efficiencies in the architecture but mobile operators for example, who have
invested considerable sums in third generation (3G) licences, are also
threatened by the potential of VoIP voice substitution. Users of 3G networks
can now use mobile phone to make voice calls using Skype (to Skype subscribers)
at the cheaper data rates. This will particularly affect the high price
international roaming business under review in many jurisdictions. This
presents a dilemma for the regulator/policy maker, whether to focus on
protecting the investment of the mobile operator and delay cost saving
innovations for consumers or to allow wireless enabled mobile VoIP to flourish?
And these questions are particularly pertinent for regulators and policy makers
in developing countries where the choice is often between defending a
government owned telco incumbent (for financial and social reasons), and the
distributive policy aspiration of making cheaper communications available to a
wider number of people particularly in rural areas.

It is still unclear how telecommunications networks will develop at a
technical level over the coming years, but it is clear that IP migration and
ultimate full NGN transition represent a logical commercial development for all
operators as they increase efficiency, facilitate service and pricing
innovation and allow for lower backbone transmission costs.

In conclusion let me make the following remarks.

All of us here accept the reality that we that we have to function in an
increasingly globalised digitally networked economy. A world in which at the
touch of a keystroke we are able to traverse vast distances between us,
crossing borders and closing time zones by the transmission of information,
data and images in a matter of seconds. Our ability to communicate and to be in
touch is almost instantaneous and ubiquitous.

What this means is that if we are to be a leading nation in the knowledge
economy in a globalised world we have to be up to it, up there and competing
against the best in the world.

It means that we have to overhaul our capacities as a nation in the various
arenas of policy and regulation, infrastructure, human capital development,
industrial enterprise, research innovation and development and public and
private sector investments in Information and Communication Technologies (ICT).
What is needed is nothing short of a revolution in ICTs to make South Africa a
winning and leading nation amongst the world's best.

This challenge is a task that falls on the shoulders of every one of us who
are here, and also those who are not here and are at home. We have to mobilise
the passion and commitment of a massive array of forces both within government
and in the private sector and in the general populace as a whole. We will have
to assemble a critical quantum of social forces, develop a mass momentum of
consciousness and activism amongst the masses of our people and generate a
higher level of awareness and consciousness of the need to participate in the
digital economy and to be connected within a broad movement of "information
technology (IT) for all". It is this critical force that must be the vanguard
and lead an IT revolution in South Africa.

I have raised my hand and thrown my hat in the ring. The question is, will
you join me? Judging from the vibes that I pick up from the buzz of your
three-day conference I have no doubt that you will!

I thank you!

Issued by: Department of Communications
3 November 2006
Source: Department of Communications (https://www.doc.gov.za/)