« Previous | Contents | Next »
Listen
1. Executive Summary
This study was commissioned to investigate the scale of the broadband availability in Scotland; review the technical options that may be used to increase this availability figure; and to provide recommendations on how the Scottish Executive may wish to take action.
The economic and social importance of access to affordable broadband has been stated by many sources and identified as a method of promoting general economic interest for both businesses and residences alike. Broadband availability in Scotland is amongst the highest levels in the developed G7 countries, and is an excellent reflection of the combined efforts of the public and private sectors.
However, this is little comfort for those without broadband, who become increasingly frustrated with broadband claims, advertising campaigns and increasing levels of provision within the well-covered areas. This level of frustration is likely to increase as the overall coverage levels extend towards 100% availability.
Our study demonstrates that it is extremely difficult to predict if a telephone line will support broadband ADSL services. Availability is down to a complex mix of technical variables, which can change over time. The main measure of availability is the length of the line as the ADSL signal is attenuated with length until it becomes unworkable. However, research has shown that even for some very long lines, local conditions can mean that a service is still achievable; conversely for some shorter lines, a service may not be available. In the end, the only absolute way of knowing is to order a service and try to provision.
However, our research has shown that according to BT estimated figures, which have been supplied to Mason under a Non Disclosure Agreement, there are predicted to be - in the very worst-case scenario - up to 32,500 lines out of 2,483,000 lines in Scotland that are unable to receive a basic 512kbps DSL broadband service. A number of these will be addressed through normal BT upgrades of the network which leaves between 7,000 and 24-25,000 which are affected by poor quality lines and distance from the exchange. This represents up to 1% of the total lines.
Geographically they are spread all over the country although there will be evidence of clusters where local conditions affect a number of lines in the same way. Whilst 512kbps is often regarded as the benchmark for 'basic' broadband, it is possible that some of these affected lines could receive a lower 256kbps service depending on local conditions. Although this would provide some benefits to users over dial-up, it would be provided at the very extremes of service and consequently may be affected by reliability issues, and due to its reduced speed, could not be considered as a broadband service.
Our analysis of the data held by the Scottish Executive on known problems in obtaining service, has identified some 47 such clusters throughout Scotland. This is based on a 'demand-led' database of c.750 entries; extrapolating to the figures provided by BT is likely to result in more clusters and/or larger clusters.
There are two technical options to provide broadband services: either by wired means or by wireless. Wired broadband is reliant on the provision of ADSL over BT's copper local loop: the wires from exchange to the customer premises. If the local loop will not support this service, as BT indicates that for between c.7,000 and 25,000 customers, then the only option is for BT to replace part, or all, of the copper local loop serving that customer. BT does have a continuous programme of copper local loop investment, and makes priority decisions based on level of return and scale of the local problem. In a number of cases, however, this investment does not make business sense for BT. Without investment in improving the local loop to these customers, they will not receive ADSL broadband services.
The alternative to wired broadband is wireless broadband. BT is not a player in providing broadband wireless access to address fixed broadband requirements; alternative suppliers must be sought. Wireless broadband comes in a number of flavours. Considering the options available here and now, the use of unlicensed wireless ( WiFi) is well established in providing local distribution of broadband within a community. The issue has always been around obtaining cost-effective backhaul from the community to the Internet and providing a sustainable service provision in a market where the numbers are just so low and dispersed, commercial provision alone is not sustainable. The technology represents the best alternative however, and cost-effective backhaul options are starting to become established through the use of faster DSL services.
Touching briefly on other wireless options, the much-touted WiMAX is not yet well established, and has higher implementation costs than WiFi, meaning it is unlikely to be immediately attractive as an in-fill solution to established ADSL areas. Other wireless options may arise in the future, but the implementation costs will always be a challenge for infill, due to frequency spectrum costs.
Satellite broadband could remain part of the mix of solutions, and is most suitable to address the very remote locations where even wireless solutions are not feasible. It is however not ideal technically, due to limits in its performance, and it does come at a much higher cost.
We conclude, therefore, that the following technical options are potentially available for consideration: improve the BT copper loop, or implement WiFi based networks with suitable cost-effective backhaul, and with satellite possibly as a fallback. If implemented, these options are likely to need public sector support at some level, with wireless and satellite requiring substantial resources - approximately £20m for wireless coverage in the non-served areas and much more using satellite.
Should the Executive wish to go further than the current level of availability, we have provided an Implementation Model, which is designed to address the infill issue through a realistic staged process. In the first instance, this involves the public sector working closely with BT to identify and address areas where the most obvious solution is for improvement in the local loop - those that are business as usual for BT or those where public sector support is possible through existing mechanisms. For example, we understand there may be scope to do this under the Executive's existing contract with BT. Our cluster analysis and categorisation would be a very useful tool - in addition to any confidential information held by BT that the Executive obtains - in this dialogue, if it were pursued.
After this first stage of the Implementation Model, the Executive has a further decision point. Should the Executive wish to go even further than the above steps, suitable subscriber clusters could be offered to the market, to be addressed by interested suppliers through a competitive process, to ensure the lowest cost to the public purse and least intervention funding. Where this fails, and no supplier is available to meet the need, or the intervention level is judged too high, a final fallback solution of individual grant assistance could be made available.
However, pursuing such procurement will take 6-9 months or possibly longer and will require resource commitment to manage through the procurement and the implementation, as well as entailing significant costs.
Finally, why take action when Scotland is so far advanced in availability compared to other countries? Perhaps those who cannot obtain service should be left in the hope that technology developments may allow the market to make that final push to 100% in due course? There is the clear option for the Executive to 'do-nothing' at this stage. The cost and effort to provide a solution to the last few will be disproportionately higher due to its very nature. Any intervention will also be complicated as the demand will shift - BT will address some areas, whilst new out of reach subscribers will come forward changing the shape of cluster - and the supply of service is at the very margin involving low numbers over a wide geography.
However, we believe that whilst it is clearly a political decision on the best use of public funds, there is now a level of expectation in place through a number of past announcements, that the Executive will propose some solution. There is no purely commercial offering available outside BT to address these areas, and with little commercial return, this is unlikely to change. The risk of the do-nothing approach is that the pressure will remain and likely intensify on the Executive.
We have set out the arguments on both sides, and conclude that the implementation model proposed, with defined stages, represents the best option, as it combines pure market delivery, intervention where judged sensible, and still leaving the do-nothing as part of the final solution.
« Previous | Contents | Next »