OECD - Fibre to the home - developments and investment

Developments in fibre technologies and investment DSTI/ICCP/CISP(2007)4/FINAL

The aim of this paper is to provide a broad overview of developments in optical fibre technologies in the last mile, that is, between subscribers’ premises and telecommunication switches and in the backhaul part, from the switches to the core of the network. Issues related to the deployment of last mile fibre networks, their costs and related regulatory issues are also raised. The timeframe for the paper is the period between 2010 and 2020. This is the period when the new generation of hybrid fibre and all-fibre networks will have matured and rolled out on a large scale.

The significant increase in demand by users for higher broadband capacity is leading Internet Service Providers and telecommunication operators to place emphasis on how to meet the requirements for network capacity. In the last 20 years the focus was on using fibre in the backbone part of the network, up to the local Main Distribution Frames. In the coming years the focus will be on bringing fibre ever closer to the end-user in order to be able to deliver the desired bandwidth. For the period 2010-2020 speeds of 50 Mbit/s downstream and 10 Mbit/s upstream may be required to enable the parallel consumption of services (HDTV, radio, videoconferencing, security etc.) over the network. These speeds are significantly higher than the current OECD definition of broadband at 256 kbit/s, but are necessary to allow the end-user to enjoy a full range of services in parallel and to allow competition between the providers of these services over the network.

The paper evaluates various technologies (wireless, hybrid and all-fibre) available to roll-out high speed first mile networks with speeds of 50 Mbit/s. It becomes clear that though wireless technology will be very important, it will not be the dominant technology to connect homes and businesses to broadband networks. Both in performance as in investment the wired technologies have the advantage. Wireless will be used to bridge the first meter, but not the first mile.

Hybrid copper-fibre networks (also known as Fibre to the Node/Curb) use the existing copper networks (cable, telephone and electric) to bridge the distance from the end-user to the fibre node, which is situated closer to the end-user than traditional exchanges. The speeds available for DSL connections are dependent on the distance between customers and the switch, with speeds deteriorating rapidly with distance so that high bandwidth, for example at 50 Mbit/s has a range limited to 450 meters which in most countries would cover only about a tenth of the population. Thus, in order to come within reach of customers high speed fibre networks are being brought to the curb or node (street cabinets). Cable networks, which are being upgraded in a number of OECD countries, may have an advantage because of having a higher maximum speed than DSL, but this is often outweighed by the shared nature of cable networks which means that the more users using the network at the same time, the less bandwidth is available per user. Although Broadband over Powerline technology is often cited as a potential competing technology to cable and DSL, there has been little large scale implementation to date of this technology and it is therefore hard to assess its potential in the market.

With fibre to the curb there is no need for main distribution frames which several incumbents have indicated that they will dismantle once they have completed fibre roll-out. However, at present in most OECD countries which have local loop unbundling, unbundling takes place at the main distribution frames. An important debate is therefore to determine the strategy for existing new entrants using unbundling. Subloop unbundling in competition with an incumbent does not seem to be a viable option on a large scale. This leaves networks that are in competition with the incumbent with the option to either move ahead and invest in all-fibre networks or to be content with the role of service provider and use the network of the incumbent (or an available competing network) through wholesale broadband access.

Fibre to the home is perhaps the most future-proof technology in that it can handle most bandwidth intensive applications. There are different topologies for fibre networks and the way they are built influences the way they can be open to multiple service providers and local loop unbundling may not be effective under certain network configurations. This could mean that incumbents may regain market power or, at best, if cable is available a duopoly situation may emerge.

When looking at the business cases for a move to higher bandwidth networks, it becomes clear that for cable and ADSL2+ networks much of the investment to support the higher speeds has already been done. For VDSL2 and FTTH networks a significant investment still needs to be made. For incumbent telecommunications networks the investment in VDSL2 seems clearer than for new entrants and smaller competing networks. Incumbents can use savings in the operational expenditure of their network, together with other savings (e.g. sale of MDF locations) to support the roll-out of the network. Whether or not a new entrant will move up the ladder of investment and invest in new VDSL2 or FTTH networks is dependent upon local conditions.

The paper provides a financial model of a FTTH-network as an indication of what the costs of such a network might be and how these costs would be distributed between the various cost elements. Though the model is dependent upon a local situation in the Netherlands, it becomes clear that the costs are significant and that the monthly costs per subscriber are highly dependent upon penetration rates. The substantial costs involved raise questions as to the level of competition which will emerge in the market. Sharing of costs, especially the civil engineering costs and wiring of buildings, can have benefits and reduce investment risks.

The role of the government in the stimulation of the roll-out of these networks can be in three ways. The role of a stimulator is defined here as removing the barriers that may impede the investment and roll out in new networks. The role of producer is defined as actually investing in new networks and the role of regulator is limited to the government’s role as a telecommunications regulator trying to guarantee a competitive marketplace. Whether or not the government will have to perform any of these roles is dependent upon the local situation. A well-defined policy, which is discussed with industry and other stakeholders, with clearly stated goals and timelines, can help identify where bottlenecks are and which areas may be unprofitable. On the basis of such a policy the government can base its decisions to stimulate or to intervene.

Governments, especially municipal governments, can play an important role in facilitating the roll-out of fibre infrastructure. This role can be in providing or facilitating rights of way, and if necessary in joint public-private partnerships in infrastructure development.