A Consultative Document issued by the Director General of Telecommunications

 July 1999

1. The price controls that Oftel imposes on BT’s retail and network charges expire in 2001. This consultation document is the first stage in Oftel’s consideration of whether new price controls are needed and, if so, the form they should take. The first step in considering whether new controls are needed is to assess the extent of competition that BT faces now and is likely to face over the period of any future price control. This document outlines likely future developments which may increase competition and so constrain BT’s behaviour without the need for price controls. Oftel seeks views on both the likelihood of these developments and their implications for the need for future price controls.

2. The current retail price control covers the residential line rental and the prices of local, national and international calls and is based on the expenditure patterns of the lowest-spending 80% of residential customers. The current cap is set at RPI-4.5%. Small businesses are also protected by an assurance that a package will be available to them which is at least as advantageous as one in which call charges are no higher than those of the residential call reference tariff (the reference tariff is the one used to monitor compliance with the price cap) and in which the line rental increases by no more than RPI each year. In addition, there is a safeguard cap of RPI+0% on the retail prices of analogue private circuits at and below 64kbit/s capacity.

3. Under the network charge control, which applies to the charges BT makes to other operators for the use of its network ("interconnection"), a cap of RPI-8% is applied to three baskets of "non-competitive" services and a safeguard cap (of RPI+0%) to services which were deemed to be "prospectively competitive". The responses to this document will help Oftel to decide whether retail price and network charge controls are needed for a further period beyond 2001 and, if so, whether the current structures are likely to be appropriate for the next retail and network controls.

4. The retail price and network charge controls are distinct. However, it is important that they should be consistent, because a major part of the costs of BT’s retail businesses are the charges for the services they buy from BT Network. In addition, many of the factors which affect competition at the retail level will also affect competition to supply wholesale services to other operators. Therefore this document discusses developments affecting both the retail and network controls.

5. Some of the developments which have been identified are particularly relevant to retail customers, including both residential and business users, and others to firms in the telecoms industry. For this reason, after this summary and an introductory chapter which sets out the purpose of the document and outlines the current controls, the document is divided into two main sections. One section discusses issues most likely to be of interest to retail customers whilst another is aimed at operators. There is deliberately some repetition of material in the two sections so that each can be read as a stand-alone document. However, only the latter includes discussion of network charge control issues.

6. In March 1996, as part of the last review of BT’s retail price control, Oftel published a "competition matrix" showing the state of competition in a number of retail markets. Each market was given an index varying between 1 indicating "no competition" and 4 indicating a "fully competitive" market. Most markets were in the intermediate stages indicating some competition but with BT retaining some degree of market power. As part of the current review, it will be necessary to consider how this has changed and Oftel would welcome responses in the form of updated versions of this matrix if respondents find this convenient.

Click here for the existing version of the competition matrix.

Click here for a blank version of the matrix that can be downloaded for this purpose.

7. Historically, BT’s line rental has been below cost. Tariff rebalancing is the term used to describe the combination of line rental increases and offsetting reductions in call prices in order to bring both more closely into line with costs within the overall price cap. Although the constraint limiting any increase in BT’s exchange line rental to RPI+2% was removed in February 1996, the residential line rental has still not risen to a level where costs, including the cost of capital, are covered. One reason may be that, although the current price control gives BT a fair degree of flexibility over which prices it reduces and in particular how far it rebalances its tariffs, any increases in BT’s residential line rental must be offset by reductions in call prices and, the larger the rental increase, the larger the offsetting call price reductions required. However this would not necessarily be true in future, most obviously if there were no retail price control.

8. One effect of rebalancing is to reduce the attractiveness of operators, known as "indirect access" (IA) operators, who provide call services but not the exchange line (for which the customer retains a BT subscription), relative to direct service from BT or other competing local access providers. If BT does rebalance significantly in future, this will tend to reduce the competitiveness of indirect access operators but could stimulate investment in competing local access networks.

9. On the other hand, the introduction of carrier pre-selection (CPS) from 2000/1 will increase the ability of indirect access operators to compete with BT by removing the need to dial extra digits when making calls via indirect access.

10. The main source of competition to BT, at least for access, and for lower users, is likely to remain the cable operators. The coverage of cable networks has increased significantly over the current price control period and, at the end of 1998, just under 50% of households had access to cable companies’ networks. How far cable operators build-out beyond existing coverage is sensitive to a number of factors including the rate of tariff rebalancing and decisions about local loop unbundling, but could receive some stimulus from the desire to offer broadband services. Cable companies are also likely to attempt to increase penetration in the area already covered. Competition from cable and other direct access operators should have received a boost from number portability, introduced in 1996.

11. A number of operators are continuing to provide a competitive alternative to BT using radio fixed access (RFA). Atlantic, for example, has announced ambitious plans to expand from its current base in Scotland. RFA operators may therefore emerge as significant competitors to BT during the next price control period. Operators of regional (wireline) networks have also expanded during the current price control period and will continue to do so. However, they have so far concentrated on business customers which of course are already outside the retail price control.

12. BT’s fixed network could also face additional competition from mobile network operators as mobile penetration increases. There are now about 15.5 million mobile users in the UK. Up to now, the significantly higher price of mobiles and performance and quality differentials have limited the extent to which they can be regarded as a realistic substitute for fixed services. But some substitution has already occurred and recent price reductions, together with the prospect of further improvements in quality, suggest that this could become an increasingly important phenomenon.

13. It is also possible that the activities of resellers will increase in importance. BT now offers tariffs designed for resellers under the name "Calls and Access". In general, resellers, who continue to rely on BT’s network, may place less of a competitive constraint on BT than competing network operators. However, some types of resellers, such as least-cost routers or brokers who are able to offer customers the best deal from among the network operators, may have the effect of increasing the intensity of competition between network operators.

14. One of the main developments which is likely to occur over the rest of this price control period and the next – indeed it is already occurring – is the rapid growth in demand for data services known as the "datawave". The growth in Internet use may be the most obvious example. Most Internet traffic so far seems to have been incremental and has not substituted for other types of traffic. This can benefit price-capped services by contributing to the common costs of providing the network. The Internet can also be used for telephony in direct competition with BT. Congestion and quality problems may limit use of the Internet itself for voice, but it provides greater competition for fax and other delay-tolerant traffic. Some operators may offer packages of Internet access and telephony where the latter is provided at very low additional cost and these may be an attractive alternative to BT for customers whose use of the network for conventional telephony is relatively low.

15. In addition, new networks are being constructed which are able to exploit the cost advantages of the Internet (which arise from the way data is transmitted in packets rather than along dedicated circuits) without the congestion and quality drawbacks. These ("Internet Protocol" or "IP") networks will be able to carry voice as well as data, with quality comparable to the PSTN. Whilst some quality, standards and interconnection issues remain to be resolved, many commentators believe that these networks are likely to pose a greater competitive threat than the Internet to operators of traditional (circuit switched) networks like BT’s PSTN. Indeed incumbent operators are likely to have to respond by building IP networks of their own. For example, AT&T has already announced that this will be the last year in which it purchases traditional-type switches for its core network and BT is also building an IP network in Spain, where it is a new entrant.

16. The development of new broadband networks to provide new services may bring with it increased competition in the basic services which are currently price-capped. As noted above, new networks are likely to be based on packet-switched technologies (similar to those used on the Internet) which promise lower costs than the circuit-switched architecture of traditional telephone networks like BT’s. Cable operators are also well placed to meet demand for fast Internet access using cable modems and the assignment of spectrum for broadband wireless access networks could be a further source of entry. Convergence between broadcasting and telecommunications could lead to operators from both fields competing to provide broadband services using different technologies.

17. Some operators may wish to use BT’s network to supply broadband services. After consultation, Oftel has concluded that BT should be required to grant access to its local loop for the provision of broadband services by competing operators. The terms of access to BT’s network are crucial to its competitive implications. Oftel’s proposals for the terms of access are designed to ensure that there continue to be adequate incentives to provide competing local infrastructure. In addition, there could be an increase in competition to provide both basic and broadband services over BT’s local loop. Oftel’s proposals for access to BT’s local loop are set out in "Access to Bandwidth: Proposals for Action", available on Oftel’s website.

18. Many international telephony markets have undergone significant liberalisation during the current price control period and this has been reflected in falls in prices. This process is likely to continue, with more countries allowing access by competing operators under the aegis of the EU and WTO. Routes to these countries are likely to become increasingly competitive.

19. The arrangements for the consultation are set out at the end of this document.

1. The main purpose of this document is to set out, and invite views on, some of the likely market developments which Oftel will need to consider in deciding whether new controls on BT’s retail prices and network charges should be put in place when the current controls expire and, if so, what the scope of any future controls should be. In particular, Oftel is seeking views on the future competitiveness of the markets in which BT operates. This is necessary because it is one of the principles of regulation that price control should not be applied to competitive markets. Therefore Oftel must consider how far competition can be relied on to bring down prices and how far price control of BT’s services is still necessary. This entails an assessment of the competitiveness of the markets in which BT operates.

2. BT’s current retail price controls expire on 31 July 2001 and the current network charge controls on 30 September 2001. Time needs to be allowed for a reference to the Competition Commission (CC), if one is necessary, before the ending of the current controls. In practice, this means that the review must be completed a year or so ahead of the ending of the controls (ie in Summer 2000). Therefore, Oftel believes it must begin the review now.

3. Oftel is keenly aware that in markets as dynamic and rapidly changing as telecommunications, a lot may happen in the next few years which could have quite profound effects on the need for and scope of price controls. Some of the potential developments are discussed in this document. But there are likely to be others which it is not possible to foresee at this early stage. There are therefore advantages in basing the new control on the most up-to-date data possible, consistent with the timetable above. For this reason, Oftel has consulted separately on the insertion of a "rollover" condition into BT’s licence which would allow the existing controls to continue for the duration of any reference to the CC and until new arrangements (if they are appropriate) are in place. It will consult formally on licence modifications to give effect to this later in the summer. For this reason too, Oftel believes that the results of this consultation on future developments should not be "set in stone" but will need to be revised to take account of new market developments as they happen.

4. The current retail price control includes access (connection and line rental) and the prices of local, national and international calls. The price control is of the RPI-X type which means that the maximum increase in the weighted average of the prices of the services subject to the control is limited to X% less than the annual rate of inflation each year. The value of X in the RPI-X formula is set at 4.5%, so if inflation is 2.5%, BT’s prices (for price-controlled services) must fall by 2% on average.

5. Perhaps the key feature of the current control is that BT’s compliance with the control is assessed using the expenditure patterns of the lowest-spending 80% of residential customers. This means that the weights used to calculate the average price change are derived from the expenditure patterns of low and medium spending residential customers and so reflect the way these customers use the phone. Changes in the prices of services which lower spending customers spend most on (access and local calls) get the highest weight. In addition, small businesses are protected by a requirement on BT to offer a business package with call charges no higher than the reference residential tariff (the reference tariff is the one used to monitor compliance with the price cap) and line rental increases of no more than the annual rate of inflation.

6. This structure was considered to be the best way of ensuring that price-cap protection was provided to those who needed it most, whilst giving BT greater freedom in areas where competition was already effective. Higher spending residential and business customers already benefited from competition between BT and a number of direct and indirect access operators. They were therefore removed from the scope of the control. For most lower-spending customers, the relatively low proportion of expenditure on national and international calls meant that indirect access was not likely to be a realistic option for them. Instead, they were likely to purchase access and calls as a package from their local access provider. Therefore, competition in local access markets was of central importance for these customers, and these were not yet effectively competitive. Consequently, lower-spending residential customers continued to need protection by retail price controls. Indeed, in the past, BT had concentrated the price cuts required by the price cap on national and international calls which benefited mainly higher-spending customers. Therefore, it was expected that lower spending customers would actually receive greater protection from the new cap than under the previous arrangements.

7. Under the current network charge control regime, network services were split into three categories of competitiveness:

• Services which will be competitive during the period of the present controls;
• Services which prospectively will become competitive during the period of the controls; and
• Services which will not be competitive during the period of the controls.

Table 1.1 (overleaf) shows the view taken in 1997 of the correct category for each service. As part of the current review, it will be necessary to consider how competition for each of these services has changed and how it is likely to evolve in the light of future developments.

Table 1.1: Interconnection services and the category of competitiveness, price control 1997/2001

8. In line with the principle that regulation should not be imposed on competitive services, no controls were applied to network services in the first category. Prospectively competitive services are subject to a "safeguard cap" of RPI+0% on each service and on each time-of-day band. It was emphasised however, that it was not intended that the safeguard cap should be the binding constraint on the charges for these services. It was expected that competitive pressure would force BT to reduce the prices for these services towards the costs faced by new network operators, that is forward looking incremental costs. In considering the need for future charge controls for these services, Oftel will be reviewing the extent to which this has happened.

9. Non-competitive services were divided into three baskets, with each basket subject to a cap of RPI-8%. Call origination, local-tandem conveyance and single transit were included together in the "General Network basket". Call termination and interconnection specific services were placed in individual baskets.

10. Oftel must decide whether this structure remains appropriate or whether, as it believes, competition will have increased sufficiently for it to reduce the scope of the next controls. Is competition likely to be sufficient for retail control to be removed entirely, even for residential customers? Even if some control is still required, can additional groups of customers or services be removed from its scope? Are safeguard caps still necessary for some retail and network services? Are some non-competitive network services now prospectively competitive? These are some of the questions which Oftel is seeking to answer.

11. In the rest of this document, we outline a number of developments which Oftel believes will affect the need for future retail price and network charge controls and, where such controls are needed, their structure and scope. Oftel invites views on the implications for future controls of the factors it has identified, and on any other relevant factors.

1. Phone services for residential and small business customers are getting cheaper. Partly this is due to regulation of the retail prices charged by BT, who serve over four fifths of customers, requiring it to reduce prices each year for basic fixed line services. It is also partly due to competition from other telecoms companies seeking to attract customers away from BT.

2. But will these controls on BT retail prices still be needed in two years time, in 2001, when current regulatory controls come to an end? This is an important issue particularly for all those customers who have no choice other than BT to provide a line to their home (currently about half of all households) or business premises. If there is likely to be enough competition in the telecoms market after 2001 then it may not be necessary to have controls on BT’s prices. Oftel is therefore reviewing the question of whether there is likely to be sufficient competition to make price controls unnecessary.

3. The stages involved in reviewing whether these controls will still be needed are as follows:

4. This timetable could be extended if a rollover condition is inserted in BT’s licence. As noted in the introductory chapter, Oftel has consulted on this issue and intends to propose licence modifications later in the summer.

Stage 1, covered by this document, involves assessment of:

• how customers have fared under previous and current price controls;
• whether customers are now taking advantage of choices offered by competitors to BT and what is helping or hindering them exercise choice;
• the likely changes over the period to 2005 in the main fixed telecoms markets resulting from how BT changes its prices; and
• developments in newer telecoms technologies and services which may impact on fixed telecoms markets.

5. The current price control is the first to focus on the lowest spending 80% of BT’s residential customers. Before that, BT’s price control focused on all customers. However, BT concentrated the price reductions which it needed to make in order to comply with the price control on the areas where it faced most competition. This meant that the benefits went disproportionately to business and higher-spending residential customers. In order to prevent BT from doing this, the focus of the current control was restricted to the lower-spending residential customers who were least able to benefit from competition (with a safeguard control for small businesses).

6. The table below shows how different groups of customers have fared under the current and previous price controls. It shows the "effective value of X" (in the RPI-X formula, described in paragraph 4 of the introductory chapter) received by different customer groups over the period from 1990/91. As can be seen from their higher "X" values, the main beneficiaries of the earlier controls were business and higher-spending residential customers. Under the current control, lower-spending residential customers have done better than in the past, whilst higher-spending customers have continued to benefit from price reductions. Indeed, since the start of the current price-cap, lower-spending customers have benefited by more than is actually required under the price control. This is because BT customers receive a substantial part of their price reductions in the form of discount schemes such as "Friends and Family". The effective X is greater than 4.5% because the higher than expected take-up of the schemes meant that BT gave away more than it was required to.

7. Effective competition, which would remove the need for price controls, depends on customers:

• having a choice of efficient telecoms suppliers competing to provide services; and
• being sufficiently well informed and having sufficient freedom of action that they can effectively exercise choice among competing suppliers.

8. This section considers these issues. If customers are not well informed, or there are other barriers to switching, then they may be reluctant to change telecoms company. This may delay the achievement of effective competition and prolong the period for which price controls on BT need to remain in place.

9. To assess if telecoms customers have problems in switching suppliers it is useful to look at the extent to which customers switch currently.

10. For residential customers, 11.9 million households, or about 50% of the total, had a choice of cable or BT to provide their direct telephone connection at January 1999. Take up of cable is around 4 million, or about 30%, of these households, equivalent to about 15% of all UK households. At January 1998, there had been only 3.4m and at January 1997 only 2.2m cable subscribers so take-up has been increasing over the last few years.

11. In the business market, BT’s competitors have around 11% of total exchange lines and took a slightly higher share, 13%, of new connections in the year 1997/98 (the last year for which figures are available from Market Information).

12. Services from companies (known as "indirect access" operators) which provide calls (typically long distance and international calls) but not the exchange line are available to all households and businesses from a wide range of suppliers. Customers of indirect access operators retain a subscription to BT for their exchange line. By the end of 1997/98, customers accounting for around 3% of residential lines and 12% of business lines had switched at least some of their calls to an indirect access operator. These percentage take up figures are broadly in line with the previous four years. It should be borne in mind that indirect access customers tend to make relatively high volumes of long-distance and international calls which tend to be relatively expensive, so they will account for a higher percentage of call revenues than of line numbers.

13. Whilst these figures show an appreciable take-up of alternative suppliers, BT still remains by far the largest operator with a particularly high share of exchange lines. This may indicate the existence of some barriers to switching. What these barriers may comprise is considered below.

14. Many markets will have some barriers to customers switching suppliers. Typically these may be due to:

• customer habit or trust in their current supplier;
• difficulties for customers in making comparisons between suppliers on basis of prices (eg due to complexity) and quality of service (eg due to lack of visibility of relevant measures);
• high level of effort perceived by customers in switching back to the original supplier if the initial change proves unsatisfactory; and
• uncertainty over extent of financial commitment if lack of prepayment schemes available.

15. Each of these factors is likely to be relevant to the fixed link telecoms market for residential customers - sometimes to a greater extent than for other markets. For example, telecoms pricing is widely seen to be complex with bills payable to different suppliers being difficult for customers to predict.

16. In addition there are some aspects of telecoms markets that may create further barriers to customers switching suppliers, for example:

• requirement for a change in the physical link to the household/premises where the customer changes direct fixed link suppliers which risks an interruption in service;
• price cap regulation means that, overall, telecoms prices fall so customers see less need for change;
• customer confusion over different types of telecoms services (eg direct to the home and indirect connections);
• the belief that cable companies may be focused on TV rather than telephony; and
• low awareness of the fact that customers can keep their telephone number when they change suppliers.

17. The impact of both the general and telecoms specific barriers to customers switching suppliers are important to consider in assessing the need for future price controls on BT. If these barriers have a minimal impact on switching or are likely to be substantially addressed either by market developments or by regulatory or other initiatives, for example to improve the information available to customers, by 2001 then the need for price controls is lessened. Competition should be able to work more effectively as customers switch easily between suppliers and seek out the best deal for their circumstances.

18. Oftel is reviewing the availability of published research on these issues and considering the commissioning of research to identify both the type and the impact of barriers to customers switching suppliers. Comments are invited on the following issues:

19. Subsequent sections of this chapter discuss, in a preliminary way, some of the currently foreseeable developments which are likely to affect the competitiveness of UK telecommunications markets in the period up to 2001 and during the next price control period (assumed to be) up to 2005. Views are invited on the likely implications of these developments for competition and for the need for new price controls.

20. Table 2 at the end of this chapter gives an indication of the number of competitors which BT faces in key market segments and their market shares. Of course, market shares and numbers of firms are only two among the many indicators of the extent of competition in a market. However, the table suggests that there is a considerable variation in the degree of competition faced by BT across different market segments. The least competitive area is still the provision of access lines to lower-spending residential customers, whilst there is more competition for call services, where indirect access operators also provide an alternative to BT, and for business customers.

21. While BT’s prices for most residential customers are subject to an overall price control, BT has a fair degree of flexibility to decide which prices it reduces to meet the price control requirements.

22. Under the price cap, any increases in BT’s residential line rental must be offset by reductions in call prices and, the larger the rental increase, the larger the offsetting call price reductions required. One of BT’s main choices is therefore whether to:

• Increase the line rental and make large offsetting reductions in call prices; or
• Keep the rental approximately the same, or even reduce it, and make smaller reductions in call prices.

More recently BT’s strategy has appeared to be to keep the residential line rental roughly constant, with only very small increases in the current price control year.

23. The approach BT takes will influence how competitors respond and how competition develops. In particular, it will affect the balance between two main groups of competitors to BT, the indirect access operators on the one hand and, on the other, the cable companies and other connect-to-the-home suppliers.

24. Customers of indirect access operators retain the BT exchange line but use a competing operator for national, international and possibly local calls. This gives customers a greater choice of operator for the provision of calls and puts competitive pressure on BT to reduce call prices. However, indirect access (IA) operators do not compete with BT for the provision of exchange lines. By contrast, cable companies and other connect-to-the-home suppliers provide both the exchange line and calls made over it.

25. The main source of competition to BT for the provision of access is the cable operators. Cable operators are able to compete with BT over most of the range of residential subscribers by spend. Many customers prefer to take access and calls as a package from a single operator and, particularly for those who do not spend significant amounts on national and international calls, indirect access may not be a suitable alternative to BT. In addition, the provision of packages of cable TV and telephony, where the latter is provided at very low additional cost, may be an attractive alternative to BT for low users. For these customers, cable operators may be the only effective alternative to BT. The very lowest spenders who would not find switching worthwhile are protected by the Light User Scheme.

26. The balance between lower BT line rentals and higher call charges, as against higher line rentals and lower call charges, would have little direct effect on cable companies’ competitive position in the absence of indirect access. Both BT and the cable companies would then be able to recover, from the profits on calls, any failure of revenues from exchange line rentals to cover the costs of providing access. However, lower BT line rentals make it more difficult for cable companies to compete with indirect access (IA) operators because IA operators do not have to make up, out of call profits, for any failure of access revenues to cover access costs. Higher call prices allow the IA operators to make an increased profit margin or offer larger discounts on calls, whilst their customers also benefit from a low BT line rental. Conversely, higher line rentals combined with lower call prices puts more pressure on indirect access operators. The table below sets out the likely impact on competition of different types of price changes.

Table 2.3 – Possible impact of different types of BT price reductions on different types of competition

27. There is therefore something of a trade-off between the extent of competition which BT is likely to face from indirect access operators on the one hand and cable and other connect-to-home operators on the other. It is possible that higher line rentals, offset by lower call prices, could lead to a reduction in the number of IA operators. However, the number of competitors is only one indicator of the extent of competition and the exit of inefficient operators, that is those with relatively higher costs, could be a sign that competition is working effectively. Provided that efficient IA operators remain in the market, perhaps choosing to offer higher value services to offset lower profits on basic services, then competition could arguably be strengthened as a result.

28. It should also be borne in mind that changes in the balance of rentals and call prices affect the distribution of the benefits of any price control between different groups of BT’s customers. Higher rentals offset by call price reductions benefit mainly higher users within the price control basket whilst lower rentals with smaller call price reductions benefit lower users more. But, in the longer term, the former option could result in more customers, including lower spenders, having a greater choice of competing operators. Comments are invited on the following issues.

29. Some developments will make it easier for customers to choose another operator as an alternative to BT. One reason why more customers do not use indirect access operators may be the need to dial extra digits in order to make calls via such operators. But from 2000/01, dialling extra digits will no longer be necessary. Instead, customers will be able to choose the operator they want to carry their calls in advance, a development known as "carrier pre-selection". How significant this is depends partly on how big a deterrent the current need to dial extra digits is to customers considering using such operators. However, from the point of view of operators, the need to dial extra digits is a major drawback because customers tend simply to forget to dial them. It is therefore likely that carrier pre-selection will stimulate more aggressive marketing by indirect access operators. This could increase their competitive impact and tend to offset any reduction in their competitiveness which could result from changes in BT’s prices.

30. Until 1996, customers switching from BT to another operator needed to change their telephone number. This meant that businesses needed to change their stationery and signage, and inform their customers, and residential customers needed to inform friends and update address books etc. This was a significant deterrent to customers to switch from BT to a competitor. The introduction in 1996 of "number portability", which enables a customer to keep his telephone number when changing operator, should enhance the competitiveness of cable companies and other connect-to-home operators in this price control period and into the next. This could in turn increase the incentive for cable roll-out and investment in other networks. Oftel will examine evidence of its effect as part of this review in order to see how far this is likely to increase competition in practice.

31. BT’s prices for basic fixed link telecoms may also be influenced by competition from other telecoms companies who are providing services which, at first glance, may appear to be of a different type eg mobile, Internet and other higher speed telecoms services.

32. There are now over 15 million mobile phones in use in the UK. Prices are falling but are still above those for fixed lines, especially for calls in peak hours. The basic quality of service for mobiles – in terms of being able to make successful calls – is lower than for fixed line phones. A significant proportion, believed to be about 20%, of the total are paid for by businesses, though they may also be used for personal calls. This may affect the sensitivity of usage to price changes.

33. Technological developments, including a "third generation" of mobile phones will enable the transfer of higher speed data and ultimately video conferencing – probably in 2002 or 2003. These developments will significantly increase the overall traffic capacity of the mobile networks, particularly for voice services, which will be able to match the fixed networks for quality. And increasingly there will be bundles of services that package fixed and mobile telecoms together.

34. Customers who have both a fixed and a mobile phone may be relatively easily persuaded to switch between them for individual calls by price differentials. It may be much more difficult to persuade customers who already have a fixed line to replace it with a mobile system, particularly if they wish to use that line for data applications. However, there is already anecdotal evidence that some customers, in particular those who have never had a fixed line, are choosing a mobile as their only phone in preference to a fixed line.

35. The extent of substitution may increase as the mobile networks expand and evolve to cater for more advanced forms of data transmission and as costs fall. For example, Orange has stated that it aims to compete with and even replace the fixed network and recent announcements of significant price cuts by One-2-One were also accompanied by statements that BT’s fixed services were seen as the main target. However, the cost of mobile networks is always likely to mean that some charges are above fixed levels. This may limit the extent to which mobile services compete head-to-head with BT’s fixed network.

36. Another potential source of competition is from service providers reselling BT’s exchange line and calls over the BT network. The reseller’s customer keeps the BT line but is billed by the reseller and has no direct contact with BT. In the last year BT has developed a tariff aimed at the resale market under the name "Calls and Access". It is expected that this will be most attractive to resellers who already have an established brand name and who wish to offer fixed access along with some other product. However, for most customers, resellers using Calls and Access are unlikely to be able to offer lower prices than an indirect access operator, although they benefit from having an exclusive relationship with the customer (with an indirect access operator the customer continues to receive a BT bill for access as well as a separate call bill).

37. Oftel may also need to consider whether customers taking the Calls and Access tariff require price cap protection. For a number of reasons it seems unlikely that such regulation will be necessary. The customers for Calls and Access will be service providers, for whose business other networks are likely to wish to compete. In addition, because BT is obliged to supply service providers at a price no greater than the retail price minus costs saved by not supplying final customers, there is an explicit link between the Calls and Access tariff and BT’s retail prices. Cuts in call prices required by the price cap must therefore be passed on to Calls and Access customers, since Calls and Access prices will have to change in response to retail price changes to comply with the "retail minus" rule. Lastly, the end-users who take service from a service provider using Calls and Access will continue to be protected by the availability of BT’s price controlled services (if a new retail cap is deemed necessary) or competitive offerings from other network operators as well as competition between service providers.

38. Most commentators agree that the demand for access to the Internet will grow substantially during this and the next price control period. Again the main concern for this document is whether this will result in increased competitive pressure on price controlled services. As well as getting access to websites and sending emails, the Internet allows customers to make low cost voice telephone calls.

39. Some operators may offer packages of Internet access and telephony, where the latter is provided at very low additional cost and these may be an attractive alternative to BT for customers whose use of the network for traditional telephony is relatively low. This is one way in which the spread of Internet usage might increase competition to BT in the provision of price-controlled services. Use of the Internet itself for telephony also has the potential to increase competitive pressure on BT. However, voice services over the Internet may not be of adequate quality to compete with BT to a significant extent, although quality is likely to improve. Competition from Internet telephony is likely to be stronger for fax and other delay-tolerant traffic for which Internet telephony provides satisfactory quality of service. The important factor in determining the impact of Internet telephony will be how far lower quality is acceptable in return for lower prices.

40. In addition, new networks are being constructed which will be able to exploit the cost advantages of the Internet (by using similar technology to transmit data in packets rather than along dedicated circuits) but with a more sophisticated pricing system to avoid the congestion and associated quality drawbacks. These networks will be able to carry voice as well as data, with quality comparable to that of traditional networks. Whilst some quality, standards and interconnection issues remain to be resolved, many commentators believe that these new networks are likely to pose a greater competitive threat than the Internet to operators of traditional networks like BT’s. Indeed incumbent operators are likely to have to respond by building similar networks of their own. For example, BT is building such a network in Spain, where it is a new entrant.

41. Developments in the effective use of high speed, high capacity ("higher bandwidth") telecoms services direct to the customer are gathering pace. Oftel is currently considering how more competition can be introduced into this emerging market. As described in Oftel’s consultative document "Access to Bandwidth: Bringing Higher Bandwidth Service to the Consumer", a number of options for supplying such services exist. These may include the construction of new networks using fibre or radio technology in competition with BT as well as upgrades of BT’s existing network, which can then be used by a number of competing service providers. In its latest document on the subject, "Access to Bandwidth: Proposals for Action", Oftel sets out its conclusion that BT should be required to permit access to its local loops.

42. While these services are likely to be more expensive than standard fixed line services, operators with their own higher bandwidth networks will be able to provide basic services as well. Under Oftel’s preferred Option for access to BT’s network, operators delivering services over BT’s network will also be able to supply packages of basic as well as enhanced services.

43. At present cable operators have concentrated on providing telephony and one-way high bandwidth services (cable television). However, cable operators are also well placed to meet future demand for fast Internet access via cable modems. Thus if increasing numbers of customers want fast access to the Internet, and find it convenient to purchase in a package with other telecommunications services, this could provide a stimulus to cable operators competing with BT. Thus several competitors may emerge as providers of higher bandwidth services which may be packaged with basic services. However, it is as yet uncertain whether there is effective demand for these services. Some argue that demand will take off once a sufficient number of subscribers – a "critical mass" - has been achieved, whilst others are more cautious, arguing that the behaviour of most consumers will be slow to change.

1. In reviewing BT’s retail price cap and network charge controls, it is necessary to consider the appropriate coverage of the caps, that is, the services and customer groups which must be included. It is one of the principles of regulation that price control should not be applied to competitive markets. Competition is the best way of ensuring that customers get the best deal possible on prices, quality and innovation. Where competition is absent or less than fully effective, there may be a need for regulation to ensure that customers are not exploited. But all regulation is costly and can be only an imperfect substitution for competition. Therefore Oftel must consider how far price control of BT’s services is necessary and how far competition can be relied on to bring down prices, applying price controls only in the former case. This entails an assessment of the competitiveness of the markets in which BT operates.

2. One reason for thinking that there may be a significant change in future competitive conditions is the large-scale growth in new, particularly data, services which is already happening and is expected to continue into the future. BT has already stated that it now carries more data than voice traffic. This growth in data traffic is sometimes referred to as the "datawave". Whilst it is impossible to predict with certainty how rapidly new services will grow in importance, developments such as the Internet and the possibility of providing broadband services including on-line business information services, home-shopping, home-banking, television on demand and in-office video-conferencing over conventional copper access networks mean that the potential for substantial growth exists.

3. Services provided over the Internet and mobile telephony are areas of extremely rapid growth. By comparison, the provision of broadband services over conventional telephone networks is still relatively undeveloped. Growth in the Internet is relevant for a number of reasons. Firstly, it may provide a pointer to growth in services provided over telephone networks in future; secondly most connections to the Internet, at least currently, make use of the local access telephone network; and thirdly some Internet services, notably Internet telephony, may provide additional competition for services provided by the traditional telecoms operators. In addition, many commentators are suggesting that the construction of new data networks specifically to carry broadband services will lead to a significant increase in network competition and that incumbent operators will need to make large-scale investments in their own networks or risk being left behind. For example, AT&T reportedly intends to buy no more traditional (ie circuit-switched technology) switches for the core of its network after the end of 1999 as it moves to upgrade its network to cater for new broadband services. As discussed later, mobile telephony may become an increasingly viable substitute for fixed networks.

4. The next section of this document briefly discusses some of the requirements for a review of competitiveness. Subsequent sections discuss, in a preliminary way, some of the currently foreseeable developments which are likely to affect the competitiveness of UK telecommunications markets in the period up to 2001 and during the next price control period (assumed to be) up to 2005. Oftel seeks views on both the likelihood of these developments and their implications for the need for future price controls.

5. As a prerequisite to a consideration of the competitiveness of markets, it is necessary to define them. Oftel has set out, notably in the Competition Act Guidelines, the approach which, in common with most competition authorities it takes to market definition. It has also proposed a set of product market definitions. The main product markets identified in the last review of BT’s retail price control were as follows:

• Access;
• Local calls;
• National calls;
• IDD calls (by route);
• ISDN (basic and primary rate)
• Inland Low Capacity PCs;
• Inland High Capacity PCs;
• IPLCs (low and high capacity);
• DQ Services & Operator Assisted Calls;
• Specially Tariffed Voice Services (including Number Translation Services);
• Payphones;
• Calls to Mobiles;
• Value-added data services;
• Broadband switched mass-market services;
• Vision circuits;
• Telex;
• Supply of equipment; and
• Mobile telephony.

6. These definitions will be reviewed. In addition, these markets will be segmented by customer group. A key feature of the current retail price control is that it is based on the expenditure weights of the lowest-spending 80% of residential customers, on the grounds that these were the group of customers least able to benefit from competition. The appropriateness of this segmentation will also be reviewed in the light of competitive developments.

7. In March 1996, as part of the last review of BT’s retail price control, Oftel published a "competition matrix" showing the state of competition in these markets. Each market was given an index varying between 1 indicating "no competition" and 4 indicating a "fully competitive" market. Most markets were in the intermediate stages indicating some competition but with BT retaining some degree of market power. As part of the current review, it will be necessary to consider how this has changed.

Click here for the existing version of the competition matrix.

Click here for a blank version of the matrix that can be downloaded for this purpose.

8. The main indicators of competitiveness which Oftel will consider are:

• The structure of the market;
• Market shares: in particular, changes in the pattern of market shares over time;
• Barriers to entry and the history of entry and exit in the market;
• Price history of the market and profitability of firms in the market;
• How firms compete in the market; and
• Other relevant factors.

9. Table 1 at the end of this chapter gives an indication of the number of competitors which BT faces in key market segments and their market shares. Of course, market shares and numbers of firms are only two among the many indicators of the extent of competition in a market. However, the table suggests that there is a considerable variation in the degree of competition faced by BT across different market segments. The least competitive area is still the provision of access lines to lower-spending residential customers, whilst there is more competition for call services, where indirect access operators also provide an alternative to BT, and for business customers.

10. It should also be borne in mind that the requirement on BT to maintain geographically uniform retail prices for basic services means that the benefits of competition in some areas can be spread to other areas where choice of operator may be more restricted. The uniform tariff has the effect of making BT decide whether to respond to competition in a particular local area on the basis of the effects on BT across the country as a whole. However, this makes it less likely that BT would respond to competition in one area by cutting prices because the cost to it of doing so is increased by the requirement to cut prices in other areas as well. Whether price control is necessary would then depend on whether the state of competition in the more competitive areas was sufficient to constrain BT’s price nationally. It would not be necessary for there to be full competition to supply basic services in all areas individually, given the uniform tariff requirement for these services.

11. Under the current charge control regime, each interconnect service offered by BT is treated as being in a separate market. The introductory chapter described how services are split into three categories of competitiveness.

12. Call termination is expected to remain non-competitive. Competition in call termination tends to be limited because, once a customer has chosen to become connected to the access network of a particular operator, anybody wishing to call that customer would have no choice but to terminate the call on the chosen operator’s network. The customer choosing the network is not generally faced with the cost of terminating the call and so does not choose a network to connect to primarily on the basis of termination charges.

13. The implication is that there will be very little change in the competitive position of call termination. This is despite developments which may increase the number of direct access competitors.

14. The competitive status of general network services, consisting of call origination, local-tandem conveyance, single transit, IDD conveyance, inter-tandem conveyance and inter-tandem transit, will be sensitive to future developments. Competition for these services generally arises only if there is an alternative network infrastructure. So factors affecting the extent of this alternative infrastructure will have a direct impact on competition. As BT routes all its retail traffic over its own network, any development which takes retail traffic away from BT increases the amount of "footloose" traffic and hence the size of the addressable market for wholesale competitors. This increases the incentive to invest in infrastructure in competition with BT’s network. However, there may be trade-offs between incentives to invest in certain types of infrastructure. In particular, some factors which increase incentives to invest in trunk networks may not have the same effect on investment in the local loop.

15. If an operator wishes to interconnect with BT it has to pay for interconnection circuits, data management amendments made by BT and for BT’s administrative costs associated with running the interconnect service. Because of BT’s market power in the provision of access, the ability to interconnect with BT is crucial to the business of both indirect access and direct access operators.

16. This means that the competitive status of interconnect specific services will be affected by developments that impact on direct access competition. So one would expect rebalancing, number portability, cable roll out and alternative direct access technologies to make the service more competitive. Equally, one would expect CPS and LLU to make the service less competitive. However, it is not expected that competing operators’ need to interconnect with BT will reduce sufficiently to make these services competitive in the next price control period.

17. The Directory Information services currently controlled by a safeguard cap are:

• Entries on to BT’s Number Information Service (OSIS);
• Agency DQ;
• Supply of BT Phonebooks; and
• Customised Phonebooks.

18. In response to a competition investigation by Oftel, BT ceased to charge OLOs for placing information onto OSIS. From 1 April 1999 BT introduced a payment to OLOs for their information. These developments make the current safeguard cap on OSIS obsolete.

19. As of April 1999 BT has replaced DAS with a new database called "Pathfinder". Access to Pathfinder will be charged on a different basis to access to DAS, per look-up instead of per terminal, thus rendering the safeguard cap on access to DAS obsolete. Oftel has decided that although Pathfinder is a direct replacement for DAS, the change in charging structure leads to Pathfinder being treated as a new service. As with all new services, this means there is no control of the charge. However, the potential for the charge to be controlled in the event of anti-competitive behaviour remains.

20. In September 1998 Oftel published a Statement entitled "Provision of Directory Information Services and Products" which, amongst other things, aimed to open up the directory information markets to competition. As a result of the Statement, BT has given operators and service providers access to downloads of the core database. It has also resulted in service providers being allowed access to Pathfinder on the same terms as OLOs. It is hoped that these changes will ensure the development of a competitive market for product databases. The speed of development will determine the treatment of Pathfinder under any future charge control.

21. The final market is for the provision of directory services and products to customers. This refers to standard services for "supply of BT Phonebooks" and "customised Phonebooks". These charges are currently subject to a safeguard cap, which was put in place before the Oftel review of Directory Information Services and pending the conclusions of that review. It was decided in that review not to unbundle the retail supply of phonebooks from line rental. Phonebooks are given to BT customers as part of their line rental so customers view them as being free. This prevents competitors recovering their costs from the provision of phone books alone. It is therefore unlikely that competition will emerge for the provision of phone books. It will therefore be necessary to consider as part of the review of the price controls whether the charge made to OLOs for these standard services should be capped.

22. The following paragraphs discuss a number of anticipated future developments which may have implications for competition at both the retail and network levels.

23. An important source of competition to BT is indirect access. Indirect access customers retain the BT exchange line but use an indirect access operator for national, international and possibly local calls. Indirect access thus gives customers a greater choice of operator for the provision of calls and puts competitive pressure on BT to reduce call prices. However, IA operators do not compete with BT for the provision of access.

24. Tariff rebalancing - the combination of increases in BT’s line rental with offsetting reductions in call prices in order to bring both more closely into line with costs - is important because of the immediate impact which it has on the profitability of indirect access competition and hence on competition from direct operators too. The availability to indirect access (IA) operators of call origination on BT’s network at cost-based interconnection rates, combined with an "unbalanced" residential line rental, that is, one which is below cost and therefore subsidised from profits on calls, creates an arbitrage opportunity which indirect access operators are able to exploit. The gap between call origination charges and retail call prices, and hence the margins available to IA operators, has grown since the start of the current retail price cap. Before about 1996/7, only national and international calls could profitably be offered by IA operators and only to business and higher spending residential customers. It is now likely to be profitable for IA operators to offer local calls and to serve all customers except the lowest four deciles of residential customers. The ability to offer local calls is seen as a valuable marketing device and therefore of particular importance. Competition from IA operators should therefore have increased since the current price cap was set.

25. The extent of future competition from IA is to some extent dependent on the rate at which BT rebalances. In particular:

• If BT rebalances by increasing rentals at RPI+2% (the assumption used in setting the current price cap) or more then the ability of indirect access operators to profit from simple arbitrage should be significantly eroded by 2001/2. This is because BT’s price cap requires rental increases to be offset by call price reductions which reduce the arbitrage opportunity available to IA operators;
• If rentals increase at RPI+0% then the arbitrage opportunity reduces in size but only very slowly; and/or
• If the line rental is assumed to be kept constant in nominal terms (RPI-RPI) then, in the absence of intervention, the size of the arbitrage opportunity goes on increasing.

26. Because of the high weight which rentals have in the current price control basket, it may be profitable for BT to rebalance as slowly as possible, if it would otherwise have to pass call price reductions to customers outside the basket which are larger than it would wish to make, and provided this does not result in too great a take-up of indirect access among its customers. The use of prior-year revenue weights also tends to give an incentive to concentrate price reductions on slower-growing services, in this case access. This is consistent with BT=s current strategy which appears to be to keep the residential rental roughly constant in nominal terms, with only very small increases in the current price control year. Therefore, the opportunity for arbitrage by indirect access is likely to persist at least to the end of the current price control period. If so indirect access should continue to be an appreciable competitive force. What happens in the next price control period may depend on the structure of the new cap, as described below.

27. It needs to be borne in mind in setting the next retail price control that the structure of the cap and hence the incentives it gives to rebalance will partly determine the strength of competition which BT faces. Direct access operators are dependent on the profits from calls to cover the costs of access, unlike IA operators, and are pressing for changes to the approach taken to the recovery of access costs in the retail price and network charge controls. These would result in more of the costs of access being borne by indirect access operators or their customers. Oftel has said it will consider whether action is needed to address these concerns as part of the present price control review.

28. If Oftel does structure the new cap to give BT a greater incentive to rebalance, or otherwise reduce the arbitrage opportunity available to IA operators, then IA operators’ margins will come under pressure. This could force some IA operators to leave the market. Therefore, in considering the scope of the new price controls it should be borne in mind that rebalancing could tend to reduce the extent of competition from IA operators, perhaps to the point where once again only national and international calls are offered and only higher spending customers are targeted. However, the number of competitors is only one indicator of the extent of competition and the exit of inefficient operators could be a sign that competition is working effectively. Provided that efficient IA operators remain in the market, perhaps choosing to offer higher value services to offset lower profits on basic services, then competition could arguably be strengthened by rebalancing.

29. Turning to network services, rebalancing (or any other measure which may result in a greater share of the costs of access being borne by indirect access operators), by stimulating investment in competing local infrastructure, for example additional roll-out by cable companies, could potentially lead to greater competition in call origination and local-tandem conveyance. However, only BT currently provides call origination as a wholesale service, though it is possible that others may choose to offer indirect access over their networks in future.

30. The impact of rebalancing on competition to provide interconnection services at the national and international level is less certain. Arguably, for a given value of X in the retail basket (ie as long as the price cap is not tightened), the impact of rebalancing should be broadly neutral although the balance between direct and indirect competitors would be altered.

How is the extent of competition to BT from IA operators likely to change over the next price control period? What does this imply for the scope of price controls?

Is the current balance between indirect access and cable and other direct access operators the right one to bring about sustainable competition to BT? Would some other combination be better? How should light users by protected if BT were to raise the line rental, offsetting this by reductions in call prices?

31. The introduction of carrier pre-selection (CPS) from 2000/01 will increase the attractiveness of, and therefore the extent of competition from, IA operators. CPS will remove the need to dial extra digits in order to make calls via IA operators. How significant this is depends partly on how big a deterrent the current need to dial extra digits is to customers considering using IA operators.

32. Some estimates available to Oftel suggest that the deterrent effect by itself is not very great, particularly given the widespread use of memory button phones and PBXs, and is equivalent to only a small percentage discount on the call price. However, from the point of view of operators, CPS will remove one of the main perceived drawbacks of indirect access, the tendency of customers simply to forget to use indirect access. It is therefore likely that CPS will stimulate more aggressive marketing by IA operators. This could increase its competitive impact and tend to offset any reduction in IA operators’ competitiveness which could result from rebalancing, discussed above.

33. The result of the introduction of CPS is likely to be a reduction in demand for alternative direct access services. This will tend to reduce local access competition. On the other hand, the stronger competitive position of indirect access is likely to increase demand for alternative national and international network operators and so will increase competition in IDD conveyance, inter-tandem conveyance and inter-tandem transit.

34. The main source of competition to BT for the provision of access is the cable operators. It was noted in the last price control review that competition to BT should increase over the price control period as the cable companies roll-out their networks, spreading choice of access provider to larger numbers of customer. At January 1999, about 50% of households had access to cable companies’ networks and, of these, about 30% had taken up cable services, so that about 15% of UK households then took cable services.

35. Cable operators are able to compete with BT over most of the range of residential subscribers by spend. Many customers prefer to take access and calls as a package from a single operator and, particularly for those who do not spend significant amounts on national and international calls, indirect access may not be a suitable alternative to BT. For these customers, cable operators may be the only effective alternative to BT. In addition, the provision of packages of cable TV and telephony where the latter is provided at very low additional cost, may be an attractive alternative to BT for low users. Cable companies’ average revenue per line from calls is about £35 per quarter (£46 for CWC, though this includes revenues from CWC’s indirect access customers) for residential customers compared to £28 for BT which suggests that cable customers are drawn from a wide range of expenditure levels. The very lowest spenders who would not find switching worthwhile are protected by the Light User Scheme.

36. The rate of rebalancing is also relevant to cable roll-out. Low or zero rates of rebalancing mean that it is likely to be more profitable to operate as an indirect access operator than to build a competing network. There is therefore something of a trade-off between the extent of competition which BT is likely to face from indirect access operators on the one hand and direct access operators on the other. How far cable operators will build out beyond the 72% coverage currently achieved, given the removal of their exclusive ability to deliver broadband services, is unclear. However it does not follow from the ending of exclusivity that all incremental investment will be unprofitable and will therefore cease. In practice there may be something of a pause while cable operators concentrate on increasing penetration in the areas already covered. Future roll out is also sensitive to decisions on local loop unbundling but may be stimulated by the desire to offer broadband services, both of which are discussed below.

37. It is also possible that the numbers of individual cable operators will continue to be reduced as a result of consolidation through merger. Because cable franchise areas do not overlap, it could be argued that these mergers do not reduce the degree of competition to BT and indeed they could increase it if the result of the merger is a more effective, more efficient operator. Oftel will continue to examine the competitive implications of any merger proposals on a case-by-case basis.

38. Any increase in direct access infrastructure competition will have a positive effect on competition for general network services. In principle, competition for call origination and local-tandem conveyance should increase because there is an alternative direct access operator. However, at present only BT is obliged to offer call origination as a wholesale service. Competition for other general network services will grow as less traffic travels BT end to end.

39. It was noted in the last price control review that "over the course of the price control period, BT would also begin to face competition from regional networks such as Norweb, Torch and Scottish Telecom and new operators such as Ionica and other operators which could provide radio-based networks. In certain areas (notably the City of London) business customers are already able to choose between a number of customers".

40. Competition from regional operators has indeed increased. Operators besides BT, CWC, KCL and the cable operators had more than 150000 lines by the end of 1997/8. However, the vast majority of these are business lines; only 4000 or less than 0.2% of the total for all operators are residential lines. This reflects the strategy of the regional network operators of concentrating on business customers in their regions. Thus the choice available to business customers outside London (where competition was already well developed) is likely to increase as regional networks continue to develop, although it is less clear that this will benefit the bulk of residential users.

41. The failure of Ionica, which did promise to provide a competitive alternative for residential customers, appears to be a setback for radio fixed access as a competitor for BT, at least for narrowband services. Scottish Telecom is still operating a similar technology in Scotland but has put a halt to further roll-out. One other operator, Atlantic, continues to provide a competing voice access network using RFA technology. However, it is at present confined to Glasgow, Aberdeen and the East Coast of Scotland although it has ambitions to expand into other areas and perhaps ultimately to become a national operator. So although Atlantic or other operators using RFA technology could become an increasingly important source of competition to BT during the next price control period, it is not clear when, or in what areas.

42. The third RFA operator, originally known as Liberty but now as Tele2 plans to offer an IP-based wireless local loop access service aimed at business users and homeworkers and intended to provide high-speed data communications rather than basic voice calls. It is currently operating a system in the Reading area at prices undercutting BT’s, apparently with some success in gaining profitable customers. Tele2 is reportedly aiming to cover 60% of the UK population by 2003. It may therefore also emerge as a significant competitor during the next price control period, at least in some areas and for customers wanting high-speed data services. Company literature suggests that the service is seen as an alternative to private circuits or ISDN connections rather than the PSTN.

43. The introduction in 1996 of number portability, which enables a customer to keep his telephone number when changing operator, is another significant development which will enhance the competitiveness of alternative access providers in this price control period and into the next. The costs of changing number when changing operator, for example as a result of the need for businesses to change stationery and signage, and to inform customers, and for residential customers to inform friends, update address books etc were a significant deterrent to customers to switch from BT to an alternative access provider. Number portability should therefore encourage more customers to switch, where of course there is a competitive alternative. This could in turn increase the incentive for cable roll-out and investment in other networks. Oftel will examine evidence of its effect as part of this review in order to see how far this is likely to increase competition in practice.

44. By raising the demand for alternative direct access networks, number portability should increase competition for call origination (in principle) and local-tandem conveyance. It should also raise demand for alternative national and international networks. It could only fail to do so if all of the customers switching from BT had been using indirect access and in this scenario the impact would be neutral. This increases competition for the remaining general network services.

45. The four mobile networks are in principle a further source of competition for BT in the provision of both access and calls. The importance of mobile telecoms has increased markedly in recent years. Subscriber numbers doubled between 1 January 1997 (6.8m) and 1 February 1999 (13.6m) and by 1 May had reached nearly 15.5m, a penetration rate of more than 26%. Forecasts are for further large-scale growth with penetration expected to pass 50% in the early years of the next century and to increase to more than 70% by 2012. However, so far the competitive impact of this growth on the fixed market has been limited and mobile phones have for the most part been seen as complementary to the fixed network.

46. In principle, mobile services could substitute for fixed services to a significant extent. GSM networks can cater for both voice and data, and provide many of the enhanced services available on the fixed networks. In practice, the extent of substitution has so far been limited by the higher prices and lower quality of mobile calls. Whilst reductions in some off peak tariffs have reduced them to levels comparable with those of BT’s fixed network, a significant mobile premium has remained in most peak rate tariffs. In addition to the higher prices, mobile speech quality and data throughput are lower, in building coverage is patchy and dropped calls are much more common than on the fixed network. Mobile performance is inferior for fax machines and modems. These performance and price differentials have so far restricted the extent of substitution of fixed services by mobile and its competitive effect on fixed tariffs.

47. A distinction can be made between substitution on a call-by-call basis and complete replacement of the fixed exchange line by a mobile phone. Customers who have both a fixed and a mobile phone may be relatively easily persuaded to switch between them for individual calls by price differentials. It may be much more difficult to persuade customers who already have a fixed line to replace it with a mobile system, particularly if they wish to use it for data applications. However, there is already anecdotal evidence that some customers, in particular those who have never had a fixed line, are choosing a mobile as their only phone in preference to a fixed line.

48. The extent of substitution may increase as the mobile networks expand and evolve to cater for more advanced forms of data transmission and as costs fall. For example, Orange’s public position is that it aims to compete with and even replace the fixed network and recent announcements of significant price cuts by One-2-One were also accompanied by statements that BT’s fixed services were seen as the main target. Networks may be upgraded to provide high speed circuit switched data (HSCSD) and packet data transmission, using the GSM General Packet Radio Service (GPRS). Licences to provide new "third generation" services to be launched early in the next century will offer a higher bit rate capability (up to 2 Mbit/s), catering for applications such as high resolution video which would not have been feasible using GSM. These developments will significantly increase the overall traffic capacity of the GSM network, particularly for voice which will be able to match the fixed networks for quality. Improved quality and lower off peak mobile tariffs may mean that the fixed line will be used increasingly for data traffic while the mobile will increasingly be regarded by some customers as the principal voice communication medium, although enhanced mobile networks will increasingly be able to provide data as well. Some analysts are stating that they expect significant substitution to take place within the next five years. However, mobile operators’ costs are expected to remain above fixed levels. Consumers may well be prepared to pay some premium for the additional convenience of mobiles but full substitutability of residential fixed lines by GSM is unlikely as long as a significant price premium remains.

49. Rather than a substitute for fixed telephony, it is possible that customers will continue to wish to purchase both fixed and mobile services and that some will prefer to do so as an integrated package. As long as BT remains the only network operator with nationwide coverage, the current regulations are likely to continue to be needed to provide indirect access to BT’s fixed network for the mobile and other operators and resellers. Fixed-mobile integration, that is the provision of both fixed and mobile services in a single package, does not necessarily reduce the need for regulation of BT’s charges.

50. In principle, mobile networks could also provide an additional source of competition in the provision of wholesale network services. The higher costs of mobile networks mean that this is unlikely to be significant in practice. However, if significant fixed-mobile substitution does occur at the retail level, then this could increase the size of the market addressable by competing long-distance wholesale conveyance providers, as mobile operators would not automatically route their long-distance traffic over BT’s network.

51. To sum up, traffic patterns to date have been such that mobile operators have chosen to maintain relatively high peak call charges to cover network costs and to reduce off-peak charges to close to fixed levels in order to increase utilisation outside peak-hours. This may already have led to some substitution of mobile for fixed traffic and it is likely that we will see fixed-mobile substitution to an increasing extent over the period of the next price control, at least for certain calls and at certain times. However, the cost of mobile networks is always likely to mean that some charges are above fixed levels. This may limit the extent to which mobiles become a fully effective substitute for fixed lines.

52. BT is required to offer access to resellers. In the last year it has developed a tariff aimed at this market under the name "Calls and Access" and is now proposing similar products for multi-line customers and for ISDN. It is expected that this will be most attractive to companies who already have an established brand name and who wish to offer fixed access along with some other product. The terms which BT is obliged to offer are "retail minus", that is based on BT’s retail prices minus any costs saved by not serving the retail customer. The Calls and Access tariff is also "rebalanced". Resellers using this tariff therefore appear at a disadvantage to IA operators at least as far as price goes, although they benefit from having an exclusive relationship with the customer (with IA the customer continues to receive a BT bill for access).

53. However, offerings such as Calls and Access increase consumer choice and allow competitors to BT in other markets to have non-discriminatory access to BT’s network. In addition, some types of resellers, such as least-cost routers or brokers who are able to offer customers the best deal from among the network operators, may have the effect of increasing the intensity of competition between network operators. They may be able to overcome some of the disadvantages which retail customers face in finding the best tariff from among a large number of possibly complex alternatives and in switching between operators. This is less likely to be a factor for network services purchased by other operators.

54. The "calls and access" package offered by BT will have little direct impact on competition for network services. However, to the extent that such packages induce more retail customers to remain on BT’s network, there could potentially be a negative impact on competition for all general network services.

55. Oftel may also need to consider whether customers taking the Calls and Access tariff require price cap protection. For a number of reasons it seems unlikely that such regulation will be necessary. The customers for Calls and Access will be service providers, for whose business other networks are likely to wish to compete. In addition, the link to BT’s retail prices will ensure that cuts in call prices required by the price cap are passed on to Calls and Access customers. Lastly, the end-users who take service from a service provider using Calls and Access will continue to be protected by the availability of BT’s price controlled services (if a new retail cap is deemed necessary) or competitive offerings from other network operators.

56. Most commentators agree that the demand for access to the Internet will grow substantially during this and the next price control period. Again the main con