Chapter 4

Assessment of market power

Preamble

4.1 The Framework Directive (FD) and the SMP Guidelines referred to in Chapter 1 for market analysis and the assessment of SMP clarify that a market shall be deemed effectively competitive where no operators in that market possess SMP. Under the new Directives Significant Market Power (SMP) has been newly defined so that it is equivalent to the competition law concept of dominance. Article 14 of the Framework Directive states:

4.2 SMP may be held by only one company in the market (single dominance) or by more than one company (collective dominance). In assessing whether SMP exists, this review takes due account of the European Commission and Oftel SMP Guidelines set out in Chapter 1.

4.3 The single dominance criteria listed in the European Commission Guidelines can be found in Oftel’s Market Review Guidelines. Oftel has also identified in its Market Review Guidelines a number of other criteria for use in assessing dominance. In the sections that follow, the retail markets are reviewed against the most relevant of these criteria.

4.4 This chapter considers these criteria first in the context of the access and call markets in the UK excluding the Hull Area and second in the context of the access and call markets in the Hull area.

4.5 While market share is an important criterion, it does not on its own, determine whether a supplier has SMP – hence the importance of considering market shares over time, the size of other suppliers and the other SMP criteria outlined by the Commission. Where market shares are discussed hereafter, it is useful to emphasise the discussion of market shares in the Competition Act guideline The Chapter II prohibition:

4.6 Given the equivalence between the newly defined concept of SMP and the competition law concept of dominance, the Director will therefore apply the same guiding principles regarding market shares as those cited in the preceding paragraph.

4.7 It is often helpful to analyse market shares both by revenue (value) and volume. This is because revenue shares will capture the effects of any premiums above the competitive (cost-based) price level that operators are able to charge. Nevertheless, a higher share by revenue is not necessarily indicative of market power. For example, the supply of more costly services than competitors would also be consistent with a higher market share by revenue than value. Where possible, market shares by both revenue and volume have been considered.

Access and call markets in the UK excluding the Hull Area

Market shares

Fixed narrowband access markets excluding the Hull area

Residential analogue access

4.8 As can be seen in the table below, BT’s residential analogue access market share by revenue is very high, in excess of 80%, and despite a few percentage points’ fall in recent years it remains persistently high.

Residential analogue percentage share of lines by revenue.

These shares relate to the UK excluding the Hull area market of 24.5 million lines.)

Source: Oftel Estimate

4.9 The Director considers that these market share figures create a presumption that BT is dominant in this residential retail market.

4.10 The existence of BT’s dominant position in this market is confirmed by the limited number of competitors and the penetration that they have achieved. For residential analogue access customers there is now typically just a single competitor to BT in each area, the relevant cable operator. Cable operators have been competing against BT for a number of years but have only achieved a combined 15% market share. The Director further believes that the UK is likely to have already witnessed the bulk of cable access rollout that is currently seen as economic by such operators.

Residential 128Kbit/s-capable digital access

4.11 BT’s ISDN 2 residential service was until mid-2001 the only service that was available to customers in this market. ntl entered the market in mid-2001 with its 128kbit/s-capable cable modem access service. Customers in its franchise areas swiftly took up its service such that ntl has developed a share in excess of 60% within the space of a year. Furthermore, ntl is continuing to grow this share at a swift pace.

Residential 128kbit/s-capable digital access percentage share of lines by volume.

	BT	ntl
99/00 Q1	100%	0%
00/01 Q1	100%	0%
01/02 Q3	85%	15%
01/02 Q4	65%	35%
02/03 Q1	49%	51%
02/03 Q2	38%	62%

These shares relate to the UK excluding the Hull area market of 400 thousand lines.

Source: Oftel Estimate

4.12 As explained earlier, BT’s residential ISDN2 service possesses a very similarly priced monthly rental to ntl’s (£23 and £24.50 respectively) and whilst BT’s service comes with a much higher connection fee (and a conversion fee); amortising these over the reasonable life of a customer would mean that BT’s charge per month might be in the region of £28-30. It is also the case that Internet access calls come at no incremental charge with the ntl service whereas they are charged on a metered basis for BT’s ISDN2 service. If these bundled ntl Internet calls are implicitly removed from the cable access charge, it can be seen that at these relative retail prices BT is charging a premium for ISDN2 compared to ntl’s cable price. So, in market shares by revenue, BT would still be likely to possess in excess of 50% of this market. However, this share is continuing to fall rapidly.

4.13 The current dynamism of BT’s and ntl’s market shares and ntl’s rapid customer growth suggest that appropriately priced 128kbit/s access services are likely to drive substantial demand.

4.14 Due to this dynamism created by the significant recent entry into this market by ntl, the Director believes it is too soon to draw conclusions that either BT and/or ntl possess SMP. It may be the case that having seen the impact that ntl’s 128Kbit/s-capable digital access service has had in this market, ntl may face a competitive response from BT in the form of a reduction in its ISDN2 prices and/or entry into this market by Telewest in its franchise areas. It is also not clear the extent to which ntl’s customers are likely to continue subscribing to this service in the future. Given this uncertainty, the Director does not believe that it is appropriate to designate either BT and/or ntl as possessing SMP in this market.

Business analogue access

4.15 BT possesses a substantial share of the business analogue access market and, whilst this share has declined in the last few years, it remains well above the presumption of dominance threshold.

Business analogue percentage share of lines by revenue.

(These shares relate to a UK excluding the Hull area market of 7 million lines.)

Source: Oftel Estimate

Business ISDN 2 access

4.16 Whilst BT faces strong competition in the provision of ISDN 2 exchange lines to residential customers from ntl, Oftel notes that BT possesses a market share in excess of 99% of this market of 1.5 million channels and that this situation has endured throughout the history of this market in the UK.

Business ISDN 30 access

4.17 As can be seen below, BT has maintained a market share of approximately 66% since 1999 in the provision of business ISDN 30 exchange lines. These market shares have been calculated from data provided to Oftel’s Market Information Group. However, as some non-cable operators do not distinguish between analogue and ISDN lines and some leased lines that are used to provide ISDN 30 go unreported, BT's market share of ISDN30 may be overstated and its share of analogue lines understated. Cable and Wireless plc (C&W) has indicated that the majority of the lines it reports as analogue lines are in fact ISDN30 lines and the Director believes that all other operators' lines are also likely to be ISDN 30 lines. Assuming that all C&W and other operator's analogue lines are ISDN30 channels, this would still give BT an approximate 63% market share of ISDN30. The Director therefore believes that BT is dominant in ISDN30 access market and that this situation has remained relatively static over the last few years.

ISDN 30 percentage share of lines by revenue.

(These shares relate to the UK excluding the Hull area market of 3 million channels.)

Source: Oftel estimate

Fixed narrowband call markets excluding the Hull area

4.18 Annex C summarises Oftel’s revenue and volume market share data for call markets in the UK excluding the Hull Area. Market shares are presented for the latest available 13 quarters.

4.19 Although operator assisted calls were identified as a separate market, this market is small and Oftel does not have revenue or traffic figures for all operators split between residential and business customers. However, in terms of assessing the competitiveness of this market, it is to be noted that all calls to the operator originating on BT exchange lines will be routed to the BT operator. Therefore, a proxy for BT’s position in OA is the number of BT exchange lines.

4.20 BT has around 81% of residential exchange lines and around 87% of business exchange lines in the UK excluding the Hull area as at 2002/03 Q1. These shares effectively remained static over the preceding 12 month period.

Residential call markets

4.21 As can be seen from Annex C, revenue market shares point to a presumption of dominance for BT in local calls, national calls, and calls to mobiles. Volume market shares also corroborate a presumption of dominance for these markets.

4.22 As outlined in Chapter 3, the Director has identified a number of residential IDD markets. These are:

i. a single market corresponding to those international routes in which corresponding wholesale markets are competitive (category A routes); and

ii. a number of smaller markets corresponding to each of the routes in which BT retains market power at the wholesale level (category B routes).

4.23 Oftel has been unable to obtain sufficiently robust retail market share information at the level of granularity required to calculate both residential and business market shares in all of the above markets. This is because in attempting to calculate market shares for each of the above IDD markets, Oftel was in many cases provided with volume data that did not distinguish between traffic originated by residential and business users. The most reliable quantitative information available, and consequently the information that the Director has chosen to base his analysis upon, has been at the aggregate level of all routes (ie category A plus category B routes), split between residential and business customers.

4.24 At the level of the combined IDD residential markets (ie category A plus category B routes), BT's volume market share has consistently exceeded 46% (as shown in Annex C), although this share has been steadily declining. Given that, in volume terms, retail traffic on "category B" routes accounts for less than 10% of the total, it is clear that BT's market share on category A routes must be in excess of 40%. It also seems likely that this figure is at least as high on category B routes, for which the underlying wholesale markets are not competitive (in the majority of cases because BT has SMP) and this makes it less likely that retail markets on category B routes are competitive). BT’s revenue market share has been persistently higher than its volume share, remaining above 57% at the aggregate level. This would be consistent with the exercise of market power, (although it is also possible that BT’s market share by revenue is affected by having a higher share of high cost routes).

4.25 The Director considers that it is likely that BT’s market share in both the "category A routes" market and the majority of the "category B routes" markets is in excess of 50% on a revenue basis, suggesting a presumption of dominance.

Business call markets

4.26 BT’s revenue share of business local and national call markets would suggest a presumption of dominance (see Annex C). However, on the basis of volumes, BT’s share of these markets is less than its share on the basis of revenue. This would be consistent with BT being able to price above rivals due to market power.

4.27 In calls to mobiles, BT’s share of business revenues has persistently exceeded 45%. This share is above the level below which the Director would consider unlikely to indicate individual dominance in the absence of other strong evidence to the contrary (ie 40%). Evidence consistent with the exercise of market power is BT’s higher share on the basis of revenues than volumes and the fact that competitor shares are generally low in this market (the next largest competitor has only 15%).

4.28 Similar caveats to those outlined in the section on retail markets apply to the quantitative information that the Director has been able to assemble on business IDD markets. However, it is clear that BT’s share in business IDD markets is less than in residential markets. Indeed, at an aggregate level (ie category A plus category B routes), BT’s market share by volume is 20% and its revenue share is 37%. The difference between these market share figures would be consistent either with the exercise of some market power or a higher proportion of BT’s revenues being derived from higher cost routes relative to competitors.

4.29 The business market for category A routes has previously been deemed effectively competitive. The Director proposes to continue with this designation in the light of the market share figures referred to above.

4.30 On the basis of the data received from operators on BT’s position in individual category B markets, the Director proposes, for the first time, not to make an SMP finding in respect of the category B business markets. However, attempts at pricing below ‘cost’ (defined as the sum of relevant wholesale charges plus retail long-run incremental cost (LRIC)), in such a retail market(s) would likely be in breach of BT’s obligations under sectoral regulation at the wholesale level and general competition law, if it has SMP in the relevant wholesale IDD market(s). (Further details on the wholesale routes where the Director has proposed to designate BT as having SMP can be found in the consultation Wholesale International Services Market Review published today.

4.31 If BT’s retail price does not cover its wholesale charges plus its retail LRIC, it must by implication be charging itself a lower wholesale rate than that levied on other operators. On routes where BT has SMP at the wholesale level, this would be unduly discriminatory because it would prevent other operators from competing by imposing a margin squeeze. Alternatively, any attempted margin squeeze could be investigated under competition law.

HHI for residential and business fixed narrowband excluding the Hull area

4.32 The Herfindahl-Hirschman Index (HHI) is a measure of market concentration. The HHI is calculated as the sum of squared market shares of all firms and therefore reflects both the size and distribution of market shares. With a large number of firms of equivalent size the HHI will approach zero. A low value of the HHI would therefore be consistent with an effectively competitive market. The HHI will increase as both the number of firms diminishes and as market shares become more divergent. A monopoly would result in a value of the HHI of 10,000.

4.33 The HHI index reported for each market ignores the aggregated category ‘others’ market share so as to minimise distortion of the index. Where the ‘others’ share is large, for example for international calls, inclusion of the square of this value will induce significant upwards bias, implying that the market is more concentrated than it is actually is. Where the ‘others’ market share is small, the upwards bias will be correspondingly small, but in order to produce a conservative estimate of concentration the ‘others’ market share has been excluded from the index.

4.34 The US Department of Justice and Federal Trade Commission Horizontal Merger Guidelines contain explicit thresholds defined in terms of the HHI. While it should be emphasised that these thresholds are for use in merger appraisal and not strictly evaluations of dominance, they provide a general indication of degrees of concentration beyond which competition is not likely to be effective.

4.35 A market with an HHI less than 1,000 is considered ‘unconcentrated’, a market with an HHI between 1,000 and 1,800 is considered moderately concentrated and a market with an HHI greater than 1,800 is regarded as ‘highly concentrated’.

4.36 As can be seen from Annex C all call markets would be considered highly concentrated on either a revenue or volume basis, with the exception of business IDD call markets considered in aggregate (ie category A plus category B routes combined). Although not reported, the HHI for the access markets considered would also point to highly concentrated markets. Using the HHI threshold for ‘highly concentrated’ of 1800, this implies that if any firm has an individual market share greater than 42% to the nearest percentage point (ie Ö 1800 » 42), that the threshold will be exceeded. For all access markets BT has a share in excess of this lower bound with the exception of residential 128kbit/s-capable digital access, where ntl exceeds the threshold (although in any case, this market is highly concentrated as only two players are present).

Profitability in access and calls

4.37 The ability to price persistently and significantly above the competitive level is an important indicator of market power. In a competitive market, individual firms should not be able to persistently maintain prices above costs (including a normal return) and thus sustain excess profits. However, it should be borne in mind that, in the short term, high profit rates can be explained by factors such as innovation and unexpected changes in demand. Conversely, low profits could indicate inefficiency of the firm rather than effective competition.

4.38 The table below summarises BT’s returns on access (connection plus rental) and call services. Although the data shown have not been split between business and residential customers, a more detailed breakdown has been made available to Oftel and used to inform the SMP assessment in respect of the relevant markets.

4.39 All data reflect end-to-end (ie retail plus network business) profitability. Retail only returns can be found in BT’s Financial Statements. End-to-end ROCE (return on capital employed) is presented below because when retail capital employed is particularly low this can make retail ROCE difficult to interpret – indeed, retail ROCE are typically greater than those shown below. Because BT is vertically integrated and will compete against a number of other vertically integrated suppliers, prices in competitive markets would be expected to be driven down to end-to-end costs (including a normal return). In the following table, return-on-sales (RoS) has also been reported for international calls to show the trend in profitability because the ROCE figures alone cannot be relied upon. (In particular, because capital employed fluctuates significantly between years and in some instances goes negative for international calls.)

BT’s financial returns

Source: derived from BT submissions to Oftel dated 1 November 2002 and 3 March 2003.

Note:

• BT's retail cost attribution methodologies for generic retailing costs often uses turnover as the basis for attribution. Where there is no direct causal relationship between the costs incurred and the services provided, BT uses turnover on the basis that the retail activity giving rise to the cost is incurred in pursuit of revenue generation. This approach may affect the results between different call services, in particular by placing a greater proportion of retail costs onto activities with relatively high revenues such as calls to mobiles.
• For the periods shown, BT’s retention for calls to mobiles was separately price-controlled. It is now within the price-control basket along with other calls.
• Negative ROCE figure for international calls reflects a negative mean capital employed (MCE).

4.40 The difference over time between ROCE and a firm’s cost of capital can be used as an indicator of the extent of any excess profits and hence the degree of competition in the markets in which it operates. BT’s current nominal pre-tax cost of capital for its systems business is set by the Director at 13.5% and provides a useful benchmark for comparing BT’s ROCE in other areas. This regulated cost of capital is applicable from 1 October 2001, prior to which it was set at 12.5%.

4.41 As can be seen from the above table, viewed in aggregate (ie business plus residential access), BT’s ROCE has typically been less than its cost of capital in only analogue access. However, this reflects BT’s particular balance between analogue access and call prices. Combining analogue access with local, national, international and calls-to-mobiles for either residential or business customer, yields a ROCE greater than BT’s cost of capital. (However, a substantial number of business calls will also originate on ISDN lines.) In ISDN2 BT’s ROCE exceeded its cost of capital in the two most recent financial years and in ISDN30 has persistently exceeded the cost of capital.

4.42 Again looking at the combination of business plus residential sales, BT has also persistently exceeded its cost of capital in all call markets, with the exception of calls to mobiles in the latest financial year (although, this is not necessarily the result of effective competition (ie lack of SMP) as BT’s retention for calls to mobiles is price-controlled). For international calls, BT’s ROCE cannot be relied upon as an indicator of end-to-end profitability, given the purchase of wholesale inputs from Concert/CNS. For this reason it is more satisfactory to consider return-on-sales, which indicates that profitability in international calls increased over the three financial years shown.

4.43 Although not shown in the above table, in residential call markets, BT has either persistently and significantly exceeded its cost of capital or, where ROCE alone cannot be relied upon, return-on-sales has increased over the three latest financial years.

4.44 Typically financial returns in call markets for residential customers (measured by ROCE or, where more appropriate, RoS) were greater than for business customers. Nevertheless, where positive, ROCE in business call markets was significantly greater than the cost of capital. BT’s operator assistance services earned more (in some cases substantially more) than the cost of capital for all years shown.

Entry barriers

4.45 The threat of potential entry may prevent a dominant incumbent from raising prices above competitive levels. However, if there are significant barriers to entry, this threat may be weak or absent.

Sunk costs

4.46 An important barrier to entry is sunk costs. Sunk costs are those which must be incurred to enter an industry but which cannot be recovered on exit. A potential entrant will only incur the sunk costs of investment in an industry if it expects to cover these sunk costs as well as the avoidable costs of production from revenues earned. The incumbent on the other hand, has already made its sunk investments and so will stay in the market as long as it can cover its avoidable costs. The incumbent may then be able to exploit this asymmetry by signalling to the entrant that, if it were to enter the market, prices would be too low to cover sunk costs. Entry would then be deterred.

4.47 Sunk costs are particularly relevant to telecommunications because a very large investment is needed to create an efficient telecommunications network and it is likely that little of this could be recovered if the entrant later decided to leave the market. The presence of significant sunk costs in both access and calls markets considered in this Review would be one factor indicating that BT has SMP in those markets.

Economies of scale

4.48 Significant economies of scale and density characterise telecommunications networks. Given the scale and configuration of BT's fixed narrowband network, it will typically incur lower transmission costs and, in general, (though not always) incur fewer switching stages than competing network operators. This will often enable BT to provide a given end-to-end call at a lower overall cost than any other network operator.

4.49 The effect of such economies of scale is that an entrant would need to take a large share of the market if it is to compete. In order to gain such a large market share, an entrant is likely to have to price well below the incumbent, which would make it more difficult to recover sunk costs. The presence of economies of scale in both access and calls markets considered in this Review would be one factor indicating that BT has SMP in those markets.

Economies of scope

4.50 A further cost advantage from which BT is able to benefit is economies of scope which arise due to the existence of common costs in the supply of different services. (Common costs are incurred when supplying a group of services jointly but which would not be reduced if the undertaking ceased supplying one of these products.) Economies of scope mean that the unit costs of producing a given service will be lower, the greater the range of other services which share costs with that service.

4.51 Given BT’s presence in virtually all significant telecommunications services, it is able to recover common costs from a range of services. This can act as a barrier to entry for competitors without the commercial or physical capability of offering such a range of services. In order to enjoy comparable economies of scope as BT, an entrant would need to enter all relevant markets simultaneously, which increases the costs of entry (a number of these costs being sunk – for example, network roll-out and advertising). The presence of economies of scale in both access and calls markets considered in this Review would be one factor indicating that BT has SMP in those markets.

Vertical integration

4.52 BT is a vertically integrated communications provider, present at both the wholesale level (see the Wholesale Fixed Narrowband Market Review and Wholesale International Services Market Review published today) and at the retail level. Where BT has SMP upstream, this confers the opportunity to leverage market power into downstream markets, thereby adversely affecting competition. In particular, difficulty in obtaining wholesale inputs or in obtaining such inputs at competitive (ie cost-based) prices can serve to accentuate barriers to entry at the retail level.

Barriers to entry – specific to access markets

4.53 BT controls a vast infrastructure of exchange lines (29 million) that cannot easily be duplicated. Not only would the development of an exchange line network require substantial sunk investment (see above for the importance of sunk costs as a barrier to entry), but the extent of BT’s access network affords it scale economies in exchange line support services, even when compared to existing operators. For example, the UK’s two cable operators possess a combined 5.4 million exchange lines.

4.54 The provision of new wholesale broadband access by means of satellite and fixed wireless may be a viable alternative to narrowband exchange lines in the future. However such services are not yet provided commercially to any significant extent and as such do not provide a competitive constraint on BT's behaviour at present and are not expected to do so over the period of this market review.

Barriers to entry – specific to call markets

4.55 Entry may be easier where entrants can benefit from an economy of scope – the cable companies being an example – or where the entrant can make use of the incumbent’s network. The barriers to entry into retail voice calls by IA operators and resellers in relation to all markets in this Review are significantly lower than for operators rolling out networks, because such suppliers do not need to incur the large sunk costs of building direct access networks. Whilst IA operators depend on the regulatory framework which enables them to obtain wholesale call origination and termination from BT at cost oriented charges, they have been given a clear framework for planning future business via the existing Network Charge Control and proposed wholesale ex-ante regulations (see the Wholesale Fixed Narrowband Review and the Review of the Wholesale International Services Markets published today)

4.56 Customer inertia in terms of the likelihood of switching to an entrant away from the incumbent can also act as a barrier to entry and/or expansion. As noted below, although there is fairly strong evidence of customer switching and awareness of other operators, a significant proportion of those who have not switched would appear unlikely to do so at present. Moreover, price has never appeared to be an important reason for not switching. In Oftel’s customer surveys conducted for the purposes of this review, for residential non-switchers to IA only 3% cited a lack of potential savings as a reason for not switching. For non-switchers to CPS this figure was 5%. These figures were slightly higher for SMEs, 11% and 14% respectively. For business non-switchers to another fixed line operator, those not doing so for lack of a price incentive were 5%.Therefore, it would seem likely that an entrant would have to offer substantial discounts or a highly differentiated and valued service in order to gain market share from the incumbent. In practice, the expected revenues from such a strategy may be insufficient to cover the costs of entry and exit (which may be necessary if entry is unsuccessful).

Entry into access in order to better compete in calls – two-bills issue

4.57 Although Oftel’s latest customer research reveals that few customers cite the need to pay two bills as a barrier to taking up IA or CPS, [Only 6% of business customers not using IA cited either complicated billing and/or not wanting to pay two bills as a reason for not using IA. The corresponding figure for residential customers was 5% and for those not using CPS the figure was 2%.] the lack of tariff flexibility arising from not being able to offer a bundle of fixed telephony services comprising direct access and calls may nevertheless serve to deter some potential entrants or restrict the expansion of recent entrants. This applies to all calls markets in this Review. Given the significant sunk costs in establishing a direct access network, the likelihood of new entry by undertakings offering a variety of multi-part tariffs covering access and calls is very low. Entry might be more attractive (and expansion more likely) if non-direct access operators were able to offer similar bundles and varied tariff structures as BT. Indeed, overcoming this barrier is part of the rationale behind Oftel’s proposals for BT to offer a wholesale line rental (WLR) product.

Countervailing buyer power

4.58 A potential constraint on market power of a supplier is the strength of buyers and the structure of the buyers’ side of the market. A supplier’s market power will be offset by a customer’s buyer power if, in the absence of that buyer, prices would have been higher. Such countervailing power is more likely when a customer accounts for a large proportion of the supplier’s total output, is well-informed about alternative sources of supply, is able to switch to other suppliers at little cost to itself and may even be able to self-supply the relevant product.

Residential customers

4.59 The Director considers that the buyer power of residential customers is negligible. Overwhelmingly, the fact that residential customers individually account for a negligible proportion of BT’s total fixed narrowband output, there is insignificant scope for such customers to exert countervailing buyer power. This applies in relation to all residential markets in this Review.

Business customers

4.60 BT has submitted that large customers typically benefit from suppliers offering tailored contracts and that large customers frequently engage in competitive tendering. (The issue of business tariffing, particularly in respect of large business customers is discussed in Chapter 7.) In practice, buyer power is most likely to be exerted by business customers who face a number of potential suppliers able to supply all of the services that they require. In practice, at present this will usually comprise the ability to offer direct access at all the customer’s sites. Buyer power is likely to be greatest in all call markets in this Review (rather than in direct access), given the presence of IA and CPS operators.

4.61 As with residential customers, small and, to some extent, medium-sized businesses, are considerably less likely to exert much effective countervailing buyer power.

Evidence on customer switching

4.62 When barriers to switching suppliers are low, a given supplier’s market power will be reduced. Evidence of significant actual switching behaviour would suggest that barriers to switching are low, that switching is a credible strategy and that supplier market power is less than it otherwise would be.

Residential customers

4.63 A summary of information on residential customers obtained from Oftel’s consumer research is summarised below. (Although this evidence was collected on a UK-wide basis, given the small size of the Hull population, this will not materially affect drawing conclusions for the UK excluding the Hull area).

• Oftel’s residential survey conducted for the purpose of this review indicated that around two thirds of UK adults with a fixed line telephone are aware of indirect access (IA) operators and 16% claim to use them. Awareness and use of carrier pre-selection (CPS) is substantially lower, with 32% of fixed line residential customers being aware of CPS and only 3% using such suppliers. (However, given that CPS – and in particular ‘all calls’ CPS – is a relatively new supply option, awareness and take-up of CPS might be expected to rise in the near future.)
• Around a quarter of consumers with fixed line phones claim to have switched supplier, broadly similar to the results published in Oftel’s quarterly report conducted in August 2002. As with the August report, the proportion of fixed line customers ever having switched increases to 38% when including IA or CPS users who do not consider the use of such suppliers to be a switch.
• The overwhelming reason for not switching to either IA or CPS was satisfaction with the existing supplier.
• Oftel’s August survey reported that the average quarterly fixed telecoms spend, including rental, calls and metered internet usage was £75 (incl. VAT).

4.64 While the extent of switching is consistent with customers exercising alternative supply options, this does not appear to have had a significant impact on BT’s returns in residential call markets. Moreover, as was noted previously, it would seem likely that an entrant would have to offer substantial discounts or a highly differentiated and valued service in order to gain market share from the incumbent. In practice, the expected revenues from such a strategy may be insufficient to cover the costs of entry and exit (which may be necessary if entry is unsuccessful).

Business customers

4.65 Evidence from Oftel’s survey on small and medium sized (SME) business customers conducted for the purposes of this review is summarised below:

• 84% of SMEs are aware of IA operators and 16% use IA for fixed line services. Although awareness of CPS is lower than IA, usage relative to awareness is significantly greater. Currently 11% of SMEs claim to use CPS.
• SME awareness of either IA or CPS is currently 87% with awareness of both at around one-third.
• Around 17% of SMEs claim to have changed fixed line supplier, although incorporating those switching to IA or CPS operators increases overall switching to 40%.
• The main reasons for not using an IA/CPS operator were satisfaction with the existing supplier (although this was less important than in the case of residential customers) and insufficient savings. Of those not using IA or CPS, few did so because of lack of information (7% for IA, 10% for CPS).
• The main reasons for not using an alternative fixed line supplier were the absence of a need to change and satisfaction with the existing supplier. Of those not having ever switched fixed line supplier few did so because of lack of information (4% were not aware of choice in their area and 3% claimed to have insufficient information on alternatives).
• Comparing fixed suppliers was typically easiest in terms of finding out about the different services on offer (68% found this very or fairly easy). Comparing quality of service was most difficult (51% found it either not at all easy or not very easy).

4.66 In summary, awareness and use of alternative sources of supply appears greater among SMEs than residential customers and lack of information does not appear a barrier to switching. However, despite this, BT’s profitability in business markets has typically remained high and in a number of cases has increased over the past three financial years.

Conclusions for BT

4.67 The Director considers BT to have significant market power (SMP) in the following retail markets:

• Residential analogue exchange line services in the UK excluding the Hull area;
• Business analogue exchange line services in the UK excluding the Hull area;
• Business ISDN2 exchange line services in the UK excluding the Hull area;
• Business ISDN30 exchange line services in the UK excluding the Hull area;
• Residential local calls in the UK excluding the Hull area;
• Residential national calls in the UK excluding the Hull area;
• Residential IDD category A calls market in the UK excluding the Hull area;
• Residential IDD Category B call markets in the UK excluding the Hull area;
• Residential calls to mobiles in the UK excluding the Hull area;
• Residential Operator assisted calls in the UK excluding the Hull area;
• Business local calls in the UK excluding the Hull area;
• Business national calls in the UK excluding the Hull area;
• Business calls to mobiles in the UK excluding the Hull area;
• Business Operator assisted calls in the UK excluding the Hull area.

4.68 The Director considers BT to have SMP in the all the above calls markets because of the reasons set out in the Chapter: high market shares, excess profitability, typically low buyer power and significant barriers to entry. The Director considers BT to have SMP in the all the above exchange line markets because of the reasons set out in the Chapter: very high market shares, excess profitability in all but one access market, significant barriers to entry and restricted scope for switching.

4.69 The Director considers that BT does not have SMP in residential 128kbit/s-capable digital access and business IDD markets because its market shares have been eroded to low levels.

Q4.1 Do you agree with the assessment of SMP in respect of BT?

Q4.2 Do you have any comments on future developments that might affect this assessment?

Access and call markets in the Hull area

Fixed narrowband access markets

Residential analogue and 128Kbit/s-capable digital access

4.70 No operators other than Kingston offer analogue or 128kbit/s-capable digital access to residential customers in the Hull area. Therefore, the Director considers Kingston to have a 100% share of these markets in the Hull area and hence there is a presumption that KC is dominant.

Business analogue, ISDN2 and ISDN30

4.71 As for the residential access markets, the Director considers that Kingston has SMP in these markets as he is not aware of any other direct access operator being present in the Hull area.

Fixed narrowband call markets

Residential call markets

4.72 There are currently no CPS or IA operators in the Hull area. However, there is no restriction on access to national IA operators that offer access to their services via a freephone or a local rate number. Oftel’s research of residential consumers in the Hull area to be published in March 2003 identified that around 8% are aware of and use IA accessed via a freephone or local rate number. Of this 8%, about a fifth (2% of all customers) use IA for all their calls. Of the 8%, around 60% (5% of all customers) use IA for more than one type of call. From these figures it can be inferred that Kingston has a market share by customer number of between 95% and 98% across all call types in the Hull area. This leads to a presumption of dominance in all call markets.

Business call markets

4.73 The Director believes that competing operators may be offering services to business customers in the Hull area. This is supported by evidence from Oftel’s survey of business customers in the Hull area, which indicates that business users are to some extent using IA operators. However, the survey evidence indicates that Kingston maintains its large market share and there is no evidence to suggest that Kingston has lost significant share in any call market. Therefore, the Director considers that Kingston has market shares leading to a presumption of dominance

Entry barriers

4.74 There are entry barriers in access and call markets in the Hull area, similar to those in such markets in the UK excluding the Hull area (only on a smaller scale). In particular, sunk costs, economies of scale, economies of scope and vertical integration will all make entry into the access and call markets in the Hull area more difficult.

Barriers to entry – access markets

4.75 As was the case with BT (in the UK excluding the Hull area), direct access competition would again require significant sunk investment in the face of an incumbent who also possesses the advantages of retail customer inertia. Moreover, given the size of the sunk investment required to enter Kingston and the relatively small number of customers who might be won coupled with likely lower post-entry prices, direct access entry seems unlikely.

4.76 The provision of new wholesale broadband access by means of satellite and fixed wireless may be a viable alternative to narrowband exchange lines in the future. However such services are not yet significantly provided commercially to any significant extent and do not provide a competitive constraint on Kingston's behaviour at the present time nor are they likely to over the period of this market review.

Barriers to entry - call markets

4.77 As noted previously, the barriers to entry into retail voice calls by IA operators and resellers are significantly lower than for operators rolling out networks, because such suppliers do not need to incur the large sunk costs of building direct access networks.

4.78 Customer inertia in terms of the likelihood of switching to an entrant away from the incumbent can also act as a barrier to entry and/or expansion. As noted above, only around 8% of residential customers in the Hull area use IA and of this 8%, approximately a fifth (only approximately 2% of all customers) use IA accessed via freephone or local rate telephone access for all their calls. Moreover, as reported earlier, survey evidence from the UK as a whole indicates that price does not appear to be an important reason for not switching. This is similar to the reasons for not being interested in using IA or CPS suppliers in the Hull area, where only 5% of residents not interested in using IA/CPS cited a lack of potential savings as a reason for not being interested. For SME’s in the Hull area the corresponding figure was 2%. Therefore it would seem likely that an entrant would have to offer substantial discounts or a highly differentiated and valued service in order to gain significant market share from the incumbent. In practice the expected revenues from such a strategy may be insufficient to cover the costs of entry and exit (which may be necessary if entry is unsuccessful).

Entry into access in order to better compete in calls – two-bills issue

4.79 As noted in the discussion of entry barriers in the UK excluding the Hull area, although Oftel’s latest customer research reveals that few customers cite the need to pay two bills as a barrier to taking up IA or CPS, the lack of tariff flexibility arising from not being able to offer a bundle of fixed telephony services comprising direct access and calls may nevertheless serve to deter some potential entrants or restrict the expansion of recent entrants. Given the significant sunk costs in establishing a direct access network, the likelihood of entry by competitors offering a variety of multi-part tariffs covering access and calls is non-existent. Entry might be more attractive (and expansion more likely) if non-direct access operators were able to offer similar bundles and varied tariff structures as Kingston. Indeed, overcoming this barrier is part of the rationale behind the proposed obligations on Kingston in respect of offering wholesale access to fixed narrowband exchange line facilities. (See wholesale fixed narrowband review published today.)

Countervailing buyer power

Residential customers

4.80 As was concluded for the UK excluding the Hull area, the Director considers that the buyer power of residential customers is negligible. Overwhelmingly, the fact that residential customers individually account for a negligible proportion of Kingston’s total fixed narrowband output, the scope to exert countervailing buyer power is effectively non-existent by individual residential customers.

Business customers

4.81 In its response to the Director’s information request, Kingston indicated that there may be buyer power in the ISDN 30 Primary rate services as these are being purchased by third parties on longer term contracts and then re-sold. For calls, Kingston considers its ‘volume of calls’ related discounts for business national calls to be an indicator of countervailing buyer power. Notwithstanding these observations, the Director considers it unlikely that any business customer(s) will have sufficient buyer power to restrict Kingston’s power to behave to an appreciable extent independently of competitors and customers in the markets in this Review. This would seem particularly the case for SMEs.

Evidence on customer switching

Residential customers

4.82 Research commissioned by Oftel to be published in March (www.oftel.gov.uk/publications/research/2003/hullr0303 ) indicates:

• that around 45% of customers with a fixed line phone in the Hull area are aware of suppliers that they can access by dialling a fixed rate or freephone or local rate number from their fixed line phone before each call;
• around 8% use IA as an alternative supplier to Kingston;
• the main reasons given by households for not switching supplier was because of satisfaction with the existing service or supplier costs; and
• Oftel’s consumer research indicated that the average residential consumer spend in the Hull area, including rental, calls and metered internet usage, was £68 (including VAT) per quarter.

4.83 In summary, while considerably fewer residential customers use IA in the Hull area compared with the UK as a whole (see above), residential customers in the Hull area estimate their fixed narrowband expenditure to be less than residential customers in the rest of the UK estimate their expenditure on the same services to be.

Business customers

4.84 Oftel also commissioned research of SME customers in the Hull area. [ "Business use of fixed telecoms services and Internet in the Hull area, Oftel business survey, January 2003", Oftel, to be published in March 2003 www.oftel.gov.uk/publications/research/2003/hullb0303] This research indicated:

• around 13% of businesses in the Hull area use an IA operator;
• around 5% of businesses in the Hull area use CPS suppliers;
• around 6% of businesses in the Hull area use an additional supplier whereby they dial a freephone or local rate number to connect to their chosen alternative;
• the main reason for businesses in the Hull area not using an IA operator or CPS supplier was because of satisfaction with the existing service;
• the average monthly spend on fixed telecoms services amongst SMEs in the Hull area for August to November 2002 was £359 (including VAT); and
• the range of monthly spend for medium sized businesses was from £237 to £2,512.

4.85 In summary, SME customers in the Hull area are more likely to use an additional fixed line supplier compared to residential customers. In addition, the level of SME’s who have used an alternative supplier is comparable with the level in the rest of the UK.

Conclusions for Kingston

4.86 The Director therefore considers Kingston to have SMP in the following retail markets:

• Residential analogue exchange line services in the Hull area;
• Residential 128kbit/s-capable digital exchange line services in the Hull area;
• Business analogue exchange line services in the Hull area;
• Business ISDN2 exchange line services in the Hull area;
• Business ISDN30 exchange line services in the Hull area;
• Residential local calls in the Hull area;
• Residential national calls originating from Hull;
• Residential IDD category A calls market from Hull;
• Residential IDD Category B call markets from Hull.
• Residential calls to mobiles originating in Hull;
• Residential operator assisted calls in the Hull area;
• Business local calls in the Hull area;
• Business national calls originating from Hull;
• Business IDD category A calls market from Hull;
• Business IDD Category B calls markets from Hull.
• Business calls to mobiles originating in Hull;
• Business operator assisted calls in the Hull area.

4.87 The Director considers Kingston to have SMP in the all the above markets because of the reasons set out in the Chapter: high market shares and significant barriers to entry and to switching.

Q4.3 Do you agree with the assessment of SMP in respect of Kingston?

Q4.4 Do you have any comments on future developments that might affect this assessment?

Definition of the dominant provider

4.88 In paragraphs 4.67 and 4.86 above, the Director has proposed that BT and Kingston have SMP in the relevant markets.

4.89 The Director considers it appropriate to prevent a person to whom an SMP service condition is applied (i.e. the dominant provider) which is part of a group of companies, exploiting the principle of corporate separation. That is to say, the dominant provider should not use another member of its group to carry out activities or to fail to comply with a condition, which would otherwise render the dominant provider in breach of its obligations.

4.90 Article 16 of the Framework Directive requires that, where a national regulatory authority determines that a relevant market is not effectively competitive, it shall identify "undertakings" with significant market power on that market and shall on such "undertakings" impose appropriate specific regulatory obligations. For the purposes of EC competition law, "undertaking" includes companies within the same corporate group,[ Viho v Commission Case C-73/95 P [1996] ECR I-5447]for example, where a company within that group is not independent in its decision making.

4.91 Accordingly, the Director considers it appropriate that the obligations detailed in this consultation document and notification shall apply to:

• British Telecommunications plc, whose registered company number is 1800000, and any BT subsidiary or holding company, or any subsidiary of that holding company, all as defined by Section 736 of the Companies Act 1985 as amended by the Companies Act 1989.
• Kingston Communications plc whose registered company number is 2150618, and any Kingston subsidiary or holding company, or any subsidiary of that holding company, all as defined by Section 736 of the Companies Act 1985 as amended by the Companies Act 1989

Chapter 5

Approach to remedies

Introduction

5.1 As set out in Chapter 4, the Director proposes that BT has SMP in the following markets (excluding the Hull area):

• Residential analogue exchange line services in the UK excluding the Hull area;
• Business analogue exchange line services in the UK excluding the Hull area;
• Business ISDN2 exchange line services in the UK excluding the Hull area;
• Business ISDN30 exchange line services in the UK excluding the Hull area;
• Residential local calls in the UK excluding the Hull area;
• Residential national calls in the UK excluding the Hull area;
• Residential IDD category A calls market in the UK excluding the Hull area;
• Residential IDD Category B call markets in the UK excluding the Hull area;
• Residential calls to mobiles in the UK excluding the Hull area;
• Residential Operator assisted calls in the UK excluding the Hull area;
• Business local calls in the UK excluding the Hull area;
• Business national calls in the UK excluding the Hull area;
• Business calls to mobiles in the UK excluding the Hull area;
• Business Operator assisted calls in the UK excluding the Hull area.

5.2 He also proposes that Kingston has SMP in the following markets:

• Residential analogue exchange line services in the Hull area;
• Residential 128kbit/s-capable exchange line services in the Hull area;
• Business analogue exchange line services in the Hull area;
• Business ISDN2 exchange line services in the Hull area;
• Business ISDN30 exchange line services in the Hull area;
• Residential local calls in the Hull area;
• Residential national calls originating from Hull;
• Residential IDD category A calls market from Hull;
• Residential IDD Category B call markets from Hull.
• Residential calls to mobiles originating Hull;
• Residential operator assisted calls in the Hull area;
• Business local calls in the Hull area;
• Business national calls originating from Hull;
• Business IDD category A calls market from Hull;
• Business IDD Category B call markets from Hull.
• Business calls to mobiles originating Hull;
• Business operator assisted calls in the Hull area.

5.3 Clause 87 of the Communications Bill provides that where SMP is found, the Director "shall...set such conditions authorised by this [Section]...as they consider appropriate". This implements Article 17 of the Universal Service Directive. The SMP Guidelines state at paragraphs 21 and 114 that this means that Oftel must impose one or more SMP conditions on a dominant provider. Further the European Commission states that the imposition of no SMP conditions on a dominant provider would be inconsistent with the new regime. Oftel must impose at least one SMP condition on a dominant provider.

5.4 However, Recital 27 of the Framework Directive states that ex ante regulation should only be imposed where there is not effective competition and where competition law remedies are not sufficient to address the problem. Oftel’s own guidelines on regulatory option appraisal note that Oftel will consider no formal ex-ante regulation in its option appraisal process.

5.5 In this light, Oftel considers in Chapters 6 and 7 the extent to which ex-ante regulation is appropriate in the markets identified above and which SMP conditions are appropriate in each particular market As part of that assessment Oftel considers whether it would be sufficient to rely on competition law alone to address market failures caused by SMP in these markets.

The need for ex-ante regulation

Introduction

5.6 As a competitive market will produce a more efficient outcome than a regulated market, the promotion of competition is central to Oftel’s goal of securing the best deal for the consumer in terms of quality, choice and value for money.

5.7 Without the imposition of ex-ante regulations to promote actively the development of competition in a non-effectively competitive market, it is unlikely that ex-post general competition law powers will be sufficient to ensure that effective competition becomes established. For example, this is because ex-post powers prohibit abuse of dominance rather than the holding of a dominant position. Ex-ante powers can be utilised to reduce the level of market power in a market and thereby encourage effective competition to become established.

Characteristics of telecoms markets in general

5.8 Generally, the case for ex-ante regulation in telecoms markets is based on the existence of market failures which, by themselves or in combination, mean that competition might not be able to become established if the regulator relied solely on its ex-post competition law powers. Therefore, it is appropriate for ex-ante regulation to be used to address these market failures and entry barriers that might otherwise prevent effective competition from becoming established. By imposing ex-ante regulation that will promote competition, it may be possible to reduce the need for such regulation later, with greater reliance on ex-post competition law.

5.9 The European Commission has also stated that ex-ante regulation is justified: "[…] where the compliance requirements of an intervention to redress a market failure are extensive (eg the need for detailed accounting for regulatory purposes, assessment of costs, monitoring of terms and conditions[...])." This is the case for many markets where lack of competition leads to a risk of a firm setting excessive prices, where a charge control would be justified, or where there is likely to be a need for intervention to set detailed terms and conditions for access to networks.

Legacy monopoly

5.10 Although competition has developed in some of the markets considered in this Review, this is from a position in which most were controlled by a legacy monopoly operator. The increase in competition that has occurred inevitably reflects the imposition of ex-ante regulation to counter the market power of the dominant operator. Despite this, BT and Kingston remain strongly dominant in these markets. Therefore, it may be appropriate to continue to impose ex-ante regulations in these markets in order to ensure that effective competition can become established.

Network externality effects

5.11 Externality effects are present in many telecoms markets. In particular, the

network externality effect, which means that the value of a network increases more than proportionately with the number of subscribers, gives the large incumbent network a great advantage over potential competitors. As a consequence, this would enable it to exclude rivals from the market.

5.12 General ex-post competition law powers may not be sufficient to address the effects of the network externality effect. This is because the network externality effect generally reinforces a dominant position and under general competition law there is no prohibition on holding a position of dominance. Therefore, It may be more appropriate to address the impact of network externality through ex-ante regulations, for example by requiring interconnection with the incumbent’s network, which would have pro-competitive effects in the retail market.

Entry barriers

5.13 Telecoms networks require significant investment and most of this will be sunk costs, in the sense that the costs will not be recoverable if the entrant decides to exit the market. Significant sunk costs create an asymmetry in the market between incumbents and potential entrants that the former could exploit to deter entry, if allowed to. Incumbents could exploit this asymmetry by signalling to a potential entrant that if it were to enter the market prices would be too low to cover sunk costs. Entry might therefore be deterred.

5.14 Also, although entry at the retail level by operators without their own networks is likely to require relatively smaller sunk investments, it is also likely to require regulated supply of wholesale inputs if retail competition is to become established where there is market power at the network level. Therefore, in telecoms markets, especially where there is a requirement for larger sunk investments, ex-ante regulation should remain in place to address this effect.

Options for regulation

5.15 The Communications Bill (Clauses 87(5)) sets out what obligations the Director can impose if he finds that any undertaking has SMP in retail markets. Those obligations "are conditions imposing on the dominant provider such regulatory controls as OFCOM may from time to time direct in relation to the provision of any public electronic communication service to end users of that service."

• Charge control
• no undue discrimination;
• price publication/notification
• accounting separation;

5.16 Any SMP conditions to be imposed must comply with the tests set out in the Communications Bill at Clauses 41-46 and also, in relation to reatil markets, at Clause 87. The Director must also bear in mind the duties set out in clause 4 of the Communications Bill. In particular, each SMP condition must pass the test set out in clause 43 of the Communications Bill, namely that each condition must be:

• objectively justifiable in relation to the networks, services or facilities to which it relates;
• not such as to discriminate unduly against particular persons or a particular description of persons;
• proportionate to what the condition is intended to achieve; and
• in relation to what it is intended to achieve, transparent.

5.18 Chapter 6 discusses whether the proposed SMP obligations set out in Schedules 1 and 2 to the Notification in Annex A to the document are appropriate to the markets within this Review where BT and Kingston have initially been found to possess SMP.

5.19 Chapter 7 discusses the applicability and interpretation of specific remedies relating to the provision of services to business customers within the markets covered by this review where BT and Kingston have initially been found to possess SMP.

5.20 It is the Director’s current view that the conditions proposed in this consultation (see Schedules 1 and 2 to the Notification in Annex A) satisfy the relevant requirements specified in the Communications Bill, as discussed in detail in chapter 6.

Click here to continue reading the document