Box 4. Network separation
In 2003, the OECD published a paper on structural separation. At the time, problems faced by new entrants in
obtaining access to the PSTN of incumbents led to calls for structural remedies on incumbents, and in particular the
separation of the local loop from service provision. The report argued that the costs of structural separation would
more than likely outweigh any benefits that it may provide. Lately, with the increase of investment in fibre in the local
loop, there has been renewed interest in some OECD countries, and by the European Commission, in functional
separation.
The first case of functional separation is the BT Openreach model in the United Kingdom, where the incumbent, BT
created an operationally independent unit aimed at ensuring that all the telecommunications industry, including other
parts of BT, have fair and equal access to the local and backhaul networks. Openreach offers access to wholesale
products (LLU, wholesale line rental, etc.) to new entrants on the same terms as BT offers its own retail entity. BT
Openreach signed a number of undertakings with the regulator, OFCOM, and set up an Equality of Access Board
aimed at monitoring compliance with undertakings and the code of Practice of Openreach and ensuring that it meets
requirement to provide products and services on an “Equivalence of Inputs” basis. The United Kingdom has found that
the introduction of functional separation has led to significant new investment from new entrants in that it increased
confidence that the regulatory system will address anti-competitive behaviour. The incumbent has also increased
investment viewing that functional separation has increased regulatory certainty. In Sweden, TeliaSonera has decided
to establish an infrastructure subsidiary selling its products on equal terms to TeliaSonera’s wholesaler customers and
the company’s own operations. In contrast to BT, TeliaSonera separated the whole network – including core and
access – from the provision of services. New Zealand’s incumbent also is in the process of implementing a separation
plan.
The renewed interest in separation by some regulators is the result of slow or non-existent parallel infrastructure
development in their countries. In addition, the move to fibre has raised the question of how many fibre access
networks a given market can support. The proposal by the European Commission, which is opposed by some EU
countries, is that functional separation should be a last resort measure available to the regulator.
The European Regulators Group (ERG) has issued an Opinion
63
in a response to the Consultation on the review of the
European regulatory framework that functional separation could be considered as a new remedy in the forthcoming
review of the EU’s regulatory framework. The ERG adds that regulators would have to judge the costs and benefits of
such a remedy and would need to base their decision on completed market reviews covering the full remit of the whole
market. ERG believes that functional separation reinforces and complements the existing remedies ensuring that
regulators can intervene where particularly non-discrimination behaviour, which cannot be addressed through other
remedies, takes place, thus providing a supplementary tool for regulators. ERG points out that the remedy of
functional separation has to be solely within the discretion of the national regulatory authority to decide upon its
applicability.
Others have argued that the current rules already apply to discriminatory behaviour and that so far, there is no clear
proof of benefits of functional separation and its potential efficiency in respect to increasing competition. They also
believe that there may be several risks associated with such a decision, such as recreating a monopoly on the local
loop and reducing the incentive to invest in future technologies. Those opposed to functional separation also argue
that it is very difficult to define the perimeter of the network that would be subject to functional separation, bearing in
mind the very quick evolution of technologies and of markets. Opponents also argue that the adoption of functional
separation incurs high implementation costs.
Compared to functional separation, there is little support for implementing structural separation which envisages a
complete separation, including ownership, between a company providing access (network infrastructure) and the
service company. A structurally separate company would resemble some municipal networks (e.g. CityNet in
Amsterdam which has a minority municipal ownership and only provides dark fibre to the market). Separation, either
functional or structural, is viewed by its supporters as limiting regulations aimed at behavioural remedies as it
eliminates discriminatory behaviour by network owners/operators; creates more efficient competition; and removes
cross-subsidies. Among OECD member countries for the moment there is ongoing discussion on using functional
separation as an additional remedy. However, if existing rules do not work well for NGA then functional separation
may become a last resort measure. Therefore, it is important, as a first step, that the benefits and costs of such a
measure are well documented so that policy makers can take appropriate decisions.
DSTI/ICCP/CISP(2007)2/FINAL
32
Dostları ilə paylaş: |