 |
|
|
|
|
|
|
|
|||
 |
 |
 |
|
|||||||
| Bundling in the pay television market | |
|||||||||
 |
 |
 |
|
|||||||
 |
|
|
|
 |
|
|||||
|
|
|
||||||||
|
Submission by the Director General of Telecommunications to the Independent Television Commission December 1997 Contents Chapter 2 Overview of the general economic and competition issues associated with bundling Chapter 3 Market definitions and competition analysis Chapter 4 Bundling of television and telephony Chapter 5 Bundling of basic channels Chapter 6 The bundling of basic with premium channels Chapter 7 Carriage obligations SummaryThis document sets out Oftel’s response to the ITC’s recent public consultation into the bundling of pay television services. ’Bundling’ generally means the tying of one service or product to the supply of others. Bundling may be economically beneficial , but it may also have anti-competitive effects. The document provides an overview of the general economic issues raised by bundling before going on to set out Oftel’s market and competition analysis which provides the basis for Oftel’s views on the issues raised by the ITC. Bundling of television and telephony (Chapter 4) A number of cable companies offer ‘access packages’ comprising telephone rental together with a small number of basic subscription channel and the terrestrial free-to-air channels, priced at around the level of the rental for a telephone line from British Telecom. There have been concerns about whether these packages constitute unfair competition against pay television operators who are not also able to offer telephony. Oftel’s view is that it is very unlikely that such packages involve a cross-subsidy. In any case, even if there were cross-subsidy, it would not be anti-competitive unless it could be established that the subsidised price was predatory. Oftel’s view is that this is does not seem plausible, and that there is no case for taking action to prevent bundling of pay television and telephony. The bundling of basic channels (Chapter 5) Subscribers to pay television receive a package of general news and entertainment channels (basic channels). Oftel’s view is that competition in the retail pay television market means that there is scope for competition between rival packages of services and indeed innovative packaging is a key way in which rival retailers can seek to differentiate themselves from their competitors. Oftel believes that bundling of basic channels is unlikely to give rise to competition problems and that any which do occur could be dealt with on a case by case basis. The bundling of basic with premium channels (Chapter 6) BSkyB currently obliges its DTH subscribers to buy its Multichannels package before they can take one or more of its premium (ie sports and movie) channels. This may create a significant barrier to entry to the retail pay television market for a provider of a package of basic channels which does not have access at the wholesale level to BSkyB’s premium programming. The document suggests that one possible remedy might be an obligation on BSkyB to make premium programming available to providers of stand-alone basic channel packages on fair reasonable and non-discriminatory terms. Should this approach prove unfeasible, a possible alternative would be a prohibition on BSkyB imposing a buy-through requirement on its retail customers. Minimum carriage obligations (Chapter 7) Many contracts between channel providers and cable operators guarantee that channels will be made available to a minimum percentage of the subscriber base. These guarantees are a way to share the risks faced by channel providers who have high fixed costs. While there are arguments for these guarantees they also have disadvantages. Oftel’s view is that there could be good reasons why a move away from minimum carriage obligations might be desirable, but such alternative arrangements could and should be the result of commercial negotiation and not imposed by regulatory intervention. Deep discounting (Chapter 8) The price for the second and third premium channel is currently substantially less than for the first premium channel. Oftel’s view is that this may have a secondary anti-competitive effect in making entry by stand-alone premium channels more difficult. However, the most important barriers to entry into the market for the supply of premium programming derive from the exclusive rights held by pay-television operators. It will be some time before the key sports rights come up for renewal. Oftel’s view is that this is not an urgent candidate for regulatory action. There does however, appear to be a good case for undertaking further investigation of this issue at a later date. ‘Bonus’ and ‘sidecar’ channels (Chapter 9) The document considers the issues raised by ‘bonus’ channels (ie channels given away free to subscribers to other premium services) and ‘sidecar’ or overspill channels. Concerns have been expressed as to whether these are either anti-competitive in their effect or require consumers to take an unnecessarily large combination of services. Oftel’s view is that there is no inherent reason to assume that the channel as traditionally defined should be the minimum or maximum element of any bundle of television services. The channel itself is a bundle of programmes and is simply one way in which television services may be packaged. Oftel believes that any action should be on a case by case basis.
Chapter 1Introduction1.1 The ITC has recently conducted a public consultation into bundling of pay television services. Oftel hopes that this document will be of use to the ITC in its deliberations. In line with its normal practice, Oftel is placing this submission on the public record. 1.2 This document contains Oftel’s analysis of a number of competition issues arising from bundling of television services, either with other television services or with other types of communication services. While the Director General of Telecommunications has no statutory responsibility for competition issues arising in the supply of television services per se the issues are nevertheless of considerable relevance to his statutory responsibilities for the following reasons:
Structure of this document 1.3 This submission begins with a brief discussion of the general economic and competition issues associated with bundling and ‘full line forcing’. It then goes on to a market analysis which includes a discussion of the definition of the relevant market, the nature of barriers to entry and an analysis of the state of competition and the market positions of the various players. The submission then moves on to a discussion of the individual issues. The term ‘bundling’ has been used in this context to cover a number of separate issues. The submission discusses the issues under the following headings:
Chapter 2Overview of the general economic and competition issues associated with bundlingIntroduction 2.1 ‘Bundling’ generally means the tying of one service or product to the supply of others. It may however also include situations where the supply of services are linked through the use of discounts. ‘Full line forcing’ is a form of bundling where, in order to obtain an individual product or service, the full range of products or services must be purchased even though there might be demand for only one product or service within the bundle. This is relevant to the buy-through requirements by which customers wanting to take premium channels first have to purchase basic channels. 2.2 Bundling can be beneficial in economic terms. Where marginal costs are low in relation to average costs, willingness to pay for different services varies among consumers (ie customers do not have the same preferences) and where conventional price discrimination is not possible, bundling may result in the combined sales of the products or services being greater than they would have been if sold separately (see glossary under Bundling and price discrimination). 2.3 Bundling may also have anti-competitive effects where:
2.3 In considering whether a particular bundled offering is anti-competitive it would therefore be necessary to consider the economic benefits and the technical and economic feasibility of unbundling together with any exclusionary effect. The welfare benefits from bundling need to be set against the benefits which may accrue in the longer term from greater dynamic efficiency due to greater competition (or the threat of competition).
Chapter 3Market definitions and competition analysisIntroduction 3.1 Oftel’s view is that any regulatory intervention on the issues under investigation should be founded on an analysis of competition. This is essential for a number of reasons:
Market power and market definitions 3.2 A firm is said to have market power when it has the ability to set its prices independently of its competitors or customers. (In a competitive market a firm’s pricing behaviour will be constrained by competition and it is said to be a ‘price-taker’). 3.3 The simple fact of a large market share does not on its own indicate that a firm has market power. The threat of entry into the market may act as a significant constraint on its pricing behaviour. The existence of market power also therefore requires the existence of significant barriers to entry. 3.4 The purpose of market definition is to support the analysis of competition. The definition of the relevant market is therefore an essential first step in determining whether a firm has market power. The standard test used in market definition is substitutability. This is defined in terms of whether an increase in the price of one product would lead consumers to switch to other competing products or services (demand-side substitutability) or lead producers to switch rapidly into the supply of the good in question (supply-side substitutability). 3.5 Oftel’s view (which it shares with that taken by the Office of Fair Trading in the report of its Director General’s review of the wholesale pay television market published in December 1996) is that pay television (ie television services predominantly paid for by the viewer through subscription or other payment for service) is a distinct market from free-to-air television (that is television which is funded predominantly by advertising or licence fee). 3.6 This does not mean that the subscription and free-to-air television are unrelated. Clearly subscription television providers are competing in some cases for viewers against free-to-air channels, they may also be competing in ‘upstream’ markets for advertising and programming rights. 3.7 However, the UK pay-television market measured in terms of the numbers of multichannel homes has been growing at 27% per year over the last five years and now accounts for over a quarter of television homes. It is also the case (as the OFT report notes) that satellite Direct to Home (‘DTH’) prices have risen sharply since 1990 which suggests that BSkyB has considerable freedom in setting its DTH retail prices. This suggests that the availability of free-to-air television is not a significant constraint on the price of pay television with the result that it is appropriate to regard the two as separate markets. (It should be noted that the observation on the pricing of pay television services does not imply any view on the profitability of pay television services. It is the case that while retail prices for pay television have increased, so have the underlying rights costs. The number of available services has also increased). 3.8 Pay television differs from free-to-air television in that in the former the consumer’s valuation of the programme services can be expressed in the price paid to the service provider. This means that pay television is able to utilise the willingness of consumers to pay for programming they value. This underpins the ability of pay television channels with a relatively low share of the total audience to bid successfully for exclusive programme rights in competition with free-to-air terrestrial channels with much larger audiences. 3.9 Within the overall pay television market definition it is helpful to identify further distinctions between:
Premium and basic channels 3.10 One of the issues raised in the request for comments is the appropriate definition of premium programming. The document suggests that the most appropriate definition is channels which have no close substitutes for which viewers have a relatively high willingness to pay. 3.11 There seem to be three characteristics which are relevant to the definition of premium programming:
3.12 Premium programming may be seen as covering a spectrum. At the extreme there is Premier League football where there is a high degree of time-criticality, limited supply and high willingness to pay. Other premium programming may have these characteristics to a lesser degree. 3.13 In this discussion we have used the term ‘premium programming’ as distinct from ‘premium channels’. Premium channels will normally contain a good deal of material which is not in itself premium (eg library films, highlights and archive material). However, the key point seems to be not the proportions of premium and non-premium material shown, but what the consumer is paying for – this may be for programming which is only a small part of the schedule (eg the Sunday afternoon football match, or a couple of ‘blockbuster’ first-run films each month). 3.14 The boundary line between premium and basic programming may in practice be somewhat blurred at the edges. It is possible that new premium channels will emerge as pay television operators seek to find ways to tap into greater than average willingness to pay for particular types of programming among particular groups of consumers. For example, it may also be the case that viewers will be willing to pay a premium for the first run of popular entertainment shows such as soaps and comedy and in the future pay channels may increasingly compete for the rights to offer these programmes. It remains to be seen whether this becomes a generalised phenomenon leading to new premium channels – this will depend in part on the amount that premium viewers are prepared to pay. 3.15 From the point of view of competition analysis the key point however is the general importance of limitations of supply which in turn underpin the premium value of such programming and the channels which derive from it. These limitations on supply and the lack of substitutes constitute significant barriers to entry. 3.16 The presence of these barriers to entry contrasts with the current position for general news and entertainment channels (’basic’ channels) where there are lower barriers to entry, the supply of programming is less limited, and there is a wider range of substitutes. 3.17 Premium programming and premium channels are regarded as the key driver of subscriber take-up of pay-television. To the extent that setting up an effective distribution system for pay television has substantial fixed, and/or upfront costs it may only be possible to create such a system through access to (and exploitation of) premium channels. Given this limited supply, monopoly control of such programming material may, in the absence of measures to ensure its availability on fair, reasonable and non-discriminatory terms, effectively foreclose the creation of alternative distribution systems or restrict their development. Retail and wholesale markets 3.18 It is also important to distinguish between retail and wholesale markets. Retail distribution is the sale and delivery of pay television services to the final customer. The wholesale market involves those products and services needed by the broadcaster or pay television operator for it to enter the retail market ie principally, transmission, conditional access, subscriber management (ie control) and programming. 3.19 The retail-wholesale distinction is of particular importance where there is ‘vertical integration’ and a vertically integrated operator is supplying wholesale services to competitors of its retail arm. Barriers to entry and the implications of the transition to digital 3.20 In considering whether a firm has market power or whether there is effective competition in a market, it is essential to take into account the nature and extent of barriers to entry. The nature of the barriers to entry into the pay television market is considered below. (i) Transmission 3.21 Until relatively recently all television services were broadcast using terrestrial transmission and were free-to-air. Spectrum scarcity meant that there were very significant barriers to entry – so much so that terrestrial television has always been characterised by a regime of ongoing sector-specific regulation. 3.22 The construction of cable systems and the launch of direct-to-home (DTH) broadcast satellites created extra capacity which has made possible the launch of pay television. 3.23 This is not to say that spectrum has ceased to be of significance as a barrier to entry. As the number of new satellite television services has increased the availability of transponders at the key orbital positions has emerged as a key constraint. Similarly the availability of capacity on analogue cable systems has become a significant issue in a number of countries. 3.24 The transition to digital technology will significantly increase the number of services which can be carried in a given amount of spectrum. This has the potential to reduce still further the significance of capacity constraints although it is not necessarily the case that the transition to digital of itself eliminates capacity as an issue. In digital terrestrial transmission it is likely to remain of significance at least until additional spectrum can be made available. 3.25 Although the transition to digital will ease capacity constraints, it is not necessarily the case that it will eliminate or reduce other barriers to entry, or lead inevitably to an increase in competition. Existing market practices, including agreements between network operators and programme suppliers, might also form barriers to entry which could result in the potential for increased competition resulting from increased capacity not being fulfilled. 3.26 While the availability of spectrum is in the process of diminishing as an issue, the ability to control access to services has become of much greater importance. This has two dimensions. The first is access to the transmission system: owners of transmission capacity may be in the position to control access to the network and exclude services even where there is available capacity. 3.27 The second aspect is that pay television requires both transmission and control ie the ability to restrict access to those who have paid for services. It is possible for new advertising-supported free-to-air channels to be launched, and a number exist. However experience suggests that the majority of new channels will require a combination of both advertising revenue and subscription. This in turn requires broadcasters to have the use of a conditional access system and subscriber management services. 3.28 The importance of conditional access as a gateway is reflected in the EC directives and UK legislation putting in place a framework for the regulation of conditional access for digital television, and proposals for extending this framework to other digital services. The issue of conditional access for analogue television is covered by undertakings agreed between BSkyB and the OFT, as well as an ITC Code of Conduct. 3.29 Programming production and supply is characterised by relatively very high fixed costs and relatively very low or zero marginal costs per subscriber. In this context if a supplier of programming is unable to gain access to a significant proportion of the potential customer base this is likely to have a substantial impact on its viability. 3.30 Due to the fixed cost nature of television services, channel providers without access to a major part of the pay television subscriber base channels may face a major disadvantage which may make them unviable. Access to the majority DTH subscriber base has been seen as a critical factor in launching new channels in the UK pay television market. In other countries (notably Germany and the United States) the attention has been on access to cable systems. 3.31 It is possible that a combination of the cable companies’ increasing share of retail pay television and further consolidation within the industry will put the issue of carriage on cable networks on the regulatory agenda in the UK. (ii) Content 3.32 It is of course the case that the key content rights come up for auction at periodic intervals. It is possible for new entrants to bid for such rights and, if successful to launch new premium services. However, there are a number of significant hurdles which such a new entrant would need to overcome:
3.33 The advent of pay-per-view may change this picture somewhat, but this is by no means inevitable. The announcement of recent distribution deals between certain Hollywood studios and a consortium of cable companies has illustrated the potential for pay-per-view to change the nature of distribution agreements Similarly pay-per-view may create the conditions for rights holders to seek to ‘cut out the middle man’ and seek a direct relationship with subscribers. However, it is by no means certain that pay-per-view will supplant either traditional premium film channels or traditional movie channels. Competition analysis 3.34 This analysis looks at the market position of players in the UK market from three perspectives. These are the ability to act exploit market power in:
The retail market 3.35 There is increasing competition between cable and satellite in the retail market. At June 1997 there were 2,068,000 homes receiving pay television via cable, accounting for 37% of subscribers, an increase of 5 percentage points since June 1996. The corresponding satellite figures were 3,532,000 homes representing 63% of subscribers (down by 5 percentage points from the previous year – although absolute numbers of subscribers had increased). This competition can be expected to intensify as cable companies’ emphasis shifts from network construction to increasing market penetration. 3.36 The arrival of digital terrestrial television in 1998 offers the potential for further competition, although capacity constraints will mean that digital terrestrial television will be constrained in the number of services it can offer, but it has other compensating advantages such as near ubiquity and the fact that it does not require either the installation of a fixed connection or a satellite dish. By 1999 therefore much of the UK population will have two potential delivery mechanisms for pay television – and a substantial proportion will have three potential sources. 3.37 There is also the potential that retail competition might be further increased by the entry of competing pay-television operators on satellite DTH. The supply of programming in the wholesale market 3.38 The number of cable and satellite channels has increased from 6 in 1988 to 66 in 1997 – of which the majority are basic channels. The transition to digital offers still more scope. The respective shares of viewing of basic channels were as follows:
3.39 The existence of a number of firms with comparable market shares is not in itself sufficient evidence that a market is competitive but suggests that the supply of basic programming is more competitive than for premium programming where , with the exception of certain niche services, BSkyB is the only wholesale supplier for premium sports and movie channels. If it is the case that such channels are a key driver of the pay television market this means that, put in stark terms, companies wishing to compete with BSkyB in the retail market must obtain such programming from BSkyB’s wholesale programming supply operation which is currently the only source. This market is subject to the substantial barriers to entry discussed above. There appear therefore to be good grounds for deeming BSkyB to be dominant in the wholesale premium programming market. 3.40 BskyB’s dominant position in the wholesale supply of premium programming has been, and remains, the central competition issue in the development of pay television. It was this issue which, in Oftel’s view, lay behind the OFT investigations which led to the March 1995 undertakings to the Director General of Fair Trading which were subsequently revised in December 1996 following the Director General of Fair Trading’s review of the wholesale pay television market. This was also a key issue in the competition for the award of the digital terrestrial multiplexes and to subsequent developments. In Oftel’s view it is also a central issue in this investigation of bundling issues.
Chapter 4Bundling of television and telephonyIntroduction 4.1 In recent months a number of cable companies have started to offer ‘access packages’ comprising telephone rental together with a small number of basic subscription channel and the terrestrial free-to-air channels. These packages are usually priced at around the level of the rental for a telephone line from British Telecom. This has apparently led to sharp increases in the penetration rates for multi-channel television in the areas concerned. 4.2 The request for comment sought views on whether the incremental price of the television element of a bundled package (which in some cases is as low as 1p) may be predatory and detrimental to fair and effective competition. 4.3 There are three sets of issues to be considered:
General principles 4.4 A key aspect of cable companies’ businesses is that they seek to exploit the economy of scope in providing telephony and entertainment. Economies of scope are any cost savings which occur when two or more products are produced together. The exploitation of economies of scope is economically efficient as it allows the minimisation of the resources consumed in producing goods and services. It follows from this that prohibiting the exploitation of cost advantages deriving from economies of scope would result in the loss of economic benefits. 4.5 The existence of substantial economies of scope means that the incremental costs of providing television services over a cable network, given that telephone services are already being provided, are relatively low. The ability of high bandwidth networks to exploit the economies of scope between telephony and television has been of central importance to the investment in the UK and the development of competition in telephony and television. Cable companies have made commitments to invest on the basis that they would have this ability and without it there would probably not be a viable UK cable industry. 4.6 Oftel recognises equally the importance of the economy of scope between digital television and interactive services. These mean that the incremental costs of providing these services may be relatively small (eg limited to providing additional facilities in the consumer equipment). Allegations of cross-subsidy 4.7 The standard test used in such circumstances to establish whether prices involve cross-subsidy is to compare the prices with the incremental and stand alone costs of providing the services. Prices should not be below incremental cost or above the stand alone cost if they are to be subsidy free. 4.8 In carrying out such a test, it is important to draw a distinction between the marginal price and the incremental revenue of a given service within a bundle. The marginal price of a given service is simply the difference between the price of the bundle including that service and the price of the bundle (which may consist of only one service) without the service concerned. However, when products are sold as a bundle the marginal price will not be equal to the incremental revenue attributable to the service in question, and cannot therefore be used directly in an assessment of whether a service within a bundle is being cross-subsidised, or whether it is priced at a predatory level. 4.9 Attempting to attribute prices to the different elements of a bundled price is an arbitrary exercise and likely to result in numbers which are economically meaningless. A low marginal price, even one below marginal cost, for one or more individual services is not a proper indicator that the service or group of services is being cross-subsidised. All that can be safely asserted is that the incremental revenue per customer will be at least equal to the marginal price of the service. At the other extreme, the incremental revenue might be as high as the stand alone price for the service taken on its own. A more detailed discussion of this issue is contained in Chapter 6. 4.10 Oftel’s view is therefore that the only economically meaningful test to apply to combined television/telephony packages is whether the incremental income to the cable company exceeds the incremental costs of the additional subscriber. Given the high level of fixed costs in cable networks it seems extremely likely that incremental customers won by the cable companies through such packages are covering their incremental costs. 4.11 The evidence available does not support assertions made as to the presence of cross-subsidy. However, even if there were cross-subsidy this is not necessarily anti-competitive behaviour. For cross-subsidy to be anti-competitive it would be necessary to establish that the subsidised price was predatory. Allegations of predatory pricing 4.12 The standard definition of predatory behaviour involves a firm foregoing short-run profits in order to weaken competitors or drive them out of the market enabling it to earn monopoly rents in the longer term by sustaining prices (and therefore profits) at a level higher than would otherwise obtain in a competitive market. This test therefore has three elements which must be fulfilled:
4.13 Even if it were possible to demonstrate that the incremental revenues from bundled television services failed to cover the marginal costs this would not necessarily equate to predatory behaviour. For this to be true there would need to be a realistic prospect of cable companies driving competitors out of the retail market in order to be able to subsequently maintain higher prices. It is difficult to see this as a plausible scenario in the context of the current UK pay television market. 4.14 In the absence of the ability to drive competitors out of the market it is difficult to see how cable companies using television as a loss leader would have an anti-competitive effect in the pay television market. Indeed, if it is the case that such promotional offers have had the result of increasing penetration rates and expanding the overall market for pay television, then the impact on competition may well be a positive one. BSkyB and telephony services 4.15 For the avoidance of doubt, it should be noted that BSkyB (or for that matter BDB) is not precluded from offering ‘affinity deals’ involving television and telephony. Oftel has published guidelines (Guidelines for Assessing Affinity Deals July 1997) which set out the criteria Oftel would use in considering such arrangements. These state that Oftel would scrutinise particularly closely arrangements where one or both parties had market power in a related market in order to ensure that the arrangements did not result in a restriction, distortion or prevention of competition in a relevant or adjacent market.
Chapter 5Bundling of basic channels5.1 As argued earlier, bundling may result in an improvement in overall economic welfare (ie the sum of consumer and producer surpluses). However, in certain circumstances bundling may have harmful effects on competition in putting established channels which are in the bouquet at an advantage against newcomers offering similar channels. 5.2 Oftel’s view is that competition in the retail market means that there is scope for competition between rival packages of services, at least, as at present on different delivery mechanisms. In a fully competitive market there are likely to be an array of different packages aimed at different segments of the markets. 5.3 Recent experiments by cable companies seeking to move away from the so-called ‘big basic’ approach are the first examples of such competition becoming a reality. Innovative packaging is a key way in which rival retailers can seek to differentiate themselves from their competitors – as it has been in mobile telephony. 5.4 The launch of digital terrestrial television services and the possibility of the emergence of rival retail service providers on the Astra digital platform offers the potential for a further enhancement of competition and innovation. 5.5 Oftel’s view therefore is that such packaging has the potential to be an important dimension of retail competition and, provided the retail market is competitive, bundling of basic services in the retail market offers consumers benefits. 5.6 This does not preclude the possibility that bundling may give rise to competition issues in relation to specific services or groups of services. Oftel’s view is that these are best dealt with on a case-by-case basis rather than by attempting in advance to specify generalised prescriptive rules on bundling.
Chapter 6The bundling of basic with premium channelsIntroduction 6.1 This chapter discusses two issues which are raised by the selling of premium channels bundled with a package of basic channels:
The issue of the pricing of second and subsequent premium channels (’deep discounting’) is discussed in Chapter 8. The marginal price of the first premium channel 6.2 The request for comment sough views on the issue of whether the bundling of basic and premium channels via buy-through involves a subsidy of premium channels from the price of the basic package. This line of argument takes as its starting point the marginal price of the premium channels and compares this with the imputed costs of the programming. This is based upon a the premise that the marginal retail price is equal to the incremental revenue attributable to the first premium channel when a customer buys the basic-plus-premium bundle. Oftel’s view is that this premise is faulty and the marginal price of the premium channels is not a reliable starting point in determining whether there is a cross-subsidy between the basic package and the premium channels. 6.3 Given the existence of the buy-through (ie the premium channel is not available on its own), the marginal price of the first premium channel cannot be regarded as the incremental revenue from the first premium channel. Without knowledge of customers’ willingness to pay, the correct split of the price of the bundle between basic and premium channels cannot be determined. In fact, the marginal retail price is the minimum possible value of the incremental revenue of the premium channel; and the maximum incremental revenue is the total price of the package, which is currently £20.99 including VAT (rounded to £21 for convenience in this discussion). Minimum and maximum implicit retail price of the premium channel 6.4 Subscribers will only buy the basic plus premium bundle if their willingness to pay for the premium channel is at least the marginal retail price. Subscribers whose willingness to pay for the premium channel is less than the marginal retail price will opt to take the basic package on its own. 6.5 Where subscribers have a willingness to pay for the basic package of zero, but a relatively high willingness to pay for the premium channel they would want to buy the premium channel on its own. Given the buy-through restrictions the premium channel is not available on its own, so they will buy the basic plus premium bundle. However, the incremental revenue attributable to the basic package would be zero and the incremental revenue attributable to the premium channel would be the total price of £21. 6.6 There are of course other possibilities between these two extremes. For example, some subscribers might have a willingness to pay for the basic package roughly equal to their willingness to pay for the premium channel – these subscribers would buy the basic plus premium bundle. In this case, the incremental revenue attributable to of the premium channel might be considered to be 50 per cent of the total price of the package (though this would not be the only justifiable division of revenues). 6.7 But it is probably more likely that incremental revenue attributable to the premium channel is nearer the total price of the basic plus premium bundle (the maximum) than it is to the marginal price of the first premium channel. It seems plausible to suggest that most subscribers are likely to have a relatively low willingness to pay for the basic package and a higher willingness to pay for the premium channel. Such subscribers want primarily to purchase premium channels and, because of the buy-through, they also all buy the basic package. Hence, BSkyB sells the basic package to all of its subscribers, but, by offering the basic package on its own at a high price, extracts the high willingness to pay for basic of a small proportion of subscribers (in 1996 only 4% of BSkyB’s DTH subscribers purchased only the basic package). 6.8 In summary, in the absence of knowledge of subscribers’ relative willingness to pay for the basic package and the first premium channel, it is not possible to identify incremental revenue attributable to the first premium channel, though the minimum (£9) and maximum (£21) amounts are known. The marginal price of the premium channels is not a reliable starting point in determining whether there is a cross-subsidy between the basic package and the premium channels. Implications for wholesale pricing and supply 6.9 This retail pricing structure has implications for BSkyB’s pricing structure for the wholesale supply of premium programming. The wholesale prices are set as a percentage of the retail price of the basic package plus the relevant premium channel or channels. This gives wholesale prices for the first premium channel which is higher than the marginal retail price. As argued above this is not necessarily discriminatory because the marginal retail price may not be a meaningful figure. In the absence of information about consumers’ willingness to pay it is impossible to determine what is incremental revenue attributable to these services. In these circumstances the appropriate test adopted by the competition authorities in examining these issues has been the size of the margin between the wholesale price to cable companies and the BSkyB retail price. Buy-through 6.10 The term ‘buy-through’ refers to the restriction of the sale of premium channels to subscribers who have already purchased a basic tier of services. Buy-through restrictions may either refer to restrictions applied by the wholesaler to other retailers governing how premium channels may be resold, and to restrictions applied by retailers governing how customers may purchase these channels. Retail and wholesale markets 6.11 In the foregoing analysis, it has been argued that BSkyB has market power in the wholesale supply of premium programming. Three examples of courses of conduct which might have the effect, whether intentional or otherwise, of extending market power from the wholesale supply of premium programming to related markets are:
6.12 We consider the use of buy-through restrictions at retail level before turning to their use by wholesalers. Retail 6.13 BSkyB currently obliges its DTH subscribers to buy its Multichannels package before they can take one or more of its premium channels. This buy-through requirement at the retail level may have some anti-competitive effects. 6.14 Under the current ‘buy-through’ arrangements, a BSkyB basic channel package is bundled with any premium channels that are bought. Given the popularity of BSkyB premium channels among pay television subscribers, the retail buy-through requirement may create a significant barrier to entry to the retail pay television market for a provider of a package of basic channels which does not have access at the wholesale level to BSkyB’s premium programming. 96 per cent of DTH pay television subscribers take at least one BSkyB premium channel. It could therefore be very difficult for a distributor of a substitute ‘stand-alone’ basic channel package to compete for the custom of such subscribers when they already receive a package of basic channels bundled with BSkyB’s premium channels. Customers who take at least one BSkyB premium channel face a marginal price of zero for BSkyB’s basic channel package since this is supplied to them for no additional outlay (they are obliged to purchase it in order to be able to take a premium channel). It cannot be inferred that the zero marginal price for BSkyB’s basic channel package within the basic-premium bundle represents cross subsidisation, or predatory pricing because an evaluation of these issues would require knowledge of incremental revenues from each service and these are not known. However, the zero marginal price may have an exclusionary effect on competing providers of basic packages. This creates the potential for a dominant position in the wholesale market for premium programming to be leveraged into the retail market. 6.15 Potentially it is possible that a provider of a freestanding basic package could enter the market as a complement to the BSkyB package or that it could win customers among that segment of the market which only wanted basic services. It could be argued therefore that buy-through arrangements do not constitute a barrier to entry. As has been noted, over a third of cable subscribers take only a basic package and this proportion has been growing (although many of these are new subscribers to low cost access packages and such customers account for only about one in twelve of the total UK pay television subscriber base. 6.16 To determine whether or not this argument is true it would be necessary to establish whether a new entrant offering a basic-only package of services could achieve a share of the satellite DTH subscriber base large enough to sustain a viable business (either alone or in combination with cable carriage). In this respect it is not clear whether or not the difference between the proportions of basic-only customers on cable and satellite reflect inherent differences between cable and satellite DTH customers. 6.17 It should be emphasised that this would not alleviate the barrier to entry to providers of basic channel packages wishing market to customers who also buy premium channels. 6.18 We discuss later in this chapter the remedies which could be appropriate if it were demonstrated that buy-through had an exclusionary effect. Wholesale 6.19 It is important to be clear that the buy-through restrictions applying to BSkyB’s premium services do not require the retailer to take any specific channel as part of its basic package. If, at some point in the future, this were to occur then competition issues would be raised. However, Oftel’s understanding is that this is not an issue here. 6.20 The existence of buy-through requirements does not appear to have restricted the ability of cable companies to decide upon the composition of their basic packages and to offer low-cost mini-basic packages. We would expect that BDB would have the same freedom. 6.21 The requirement may however hamper the ability of other retailers to compete with BSkyB at the retail level by removing the option of retailing BSkyB premium channels separately from basic channels. Possible remedies 6.22 If it were demonstrated that making premium channels available only as part of a bundled package with basic channels was a barrier to entry to a competing provider of a package of basic channels, one possible remedy might be an obligation on dominant operators (at this stage BSkyB) to make premium programming available to providers of stand-alone basic channel packages on fair reasonable and non-discriminatory terms. A basic channel provider would then be able to compete with BSkyB’s and cable companies’ basic-premium bundle, by bundling its own basic channels with BSkyB’s premium channels. BSkyB’s basic-premium bundling would then no longer represent a barrier to entry to the retail market. 6.23 This begs the question of what is meant by ‘fair, reasonable and non-discriminatory’. As a general rule, a minimum requirement should be that the wholesale prices of BSkyB premium channels are:
6.24 The non-discrimination condition should ensure that all competitors in the downstream market face the same price for a key input, ensuring that a high retail market share reflects genuinely higher efficiency, rather than simply an extension of market power from the upstream market. The prohibition on cross-subsidy of BSkyB’s DTH arm should ensure that the wholesale price is not set at so high a level that a retail competitor of BSkyB’s efficiency would not be profitable given BSkyB’s retail prices (which would be equivalent to refusal to supply). 6.25 It is recognised that competition authorities have generally been reluctant to impose obligations to supply in relation to intellectual property rights unless this is the most effective way of solving an identified competition problem. However there are precedents: in the United States the Programme Access Rules under the 1992 Cable Act have the a similar effect and have been important in providing the basis for the entry of digital satellite pay television operators which compete both with the cable companies and with one another. 6.26 An obligation to supply premium programming on fair reasonable and non-discriminatory terms would have the effect of enabling an alternative provider of a package of basic channels to compete with BSkyB both for customers who were interested only in basic programming and those interested in premium programming. Should this approach prove infeasible, a possible alternative might be a prohibition on BSkyB imposing a buy-through requirement on its retail customers. This would have the effect of enabling market entry without the need to replicate the bundling of premium and basic. 6.27 There are two possible forms a prohibition on buy-through might take:
6.28 A difficulty with the first approach is that it could prove ineffective, unless further controls were placed upon BSkyB to ensure that it does not achieve a similar exclusionary effect through its retail tariff structure, eg by charging a very high stand alone price for premium channels. Such controls would therefore require ongoing intervention in the retail market. The second approach however would risk losing any economic benefits resulting from the bundling of premium and basic channels – although it is not clear how substantial these are. 6.29 A prohibition on buy-through could have the effect of bringing about a ‘rebalancing’ in the relative prices of premium channels and basic packages. It would be expected that the price of basic channel packages would fall while the price of premium channels would rise. The effects of this are uncertain. On the one hand substantial numbers of consumers might opt to bypass the basic tier, it is possible that this could have the effect of reducing the number of basic channels available. On the other hand, if the prices for basic subscription packages were reduced this could make these packages more attractive as a stand alone proposition. Together with the reduction of barriers to entry by competing packages of basic channels this could have a beneficial impact on the development of the market and encourage entry by new basic channels. Impact on wholesale terms and conditions 6.30 Such a rebalancing of BSkyB’s retail tariffs could also have consequent implications for its wholesale terms and conditions. It might make it possible to move away from the present rate card which is based on the combined retail prices of BSkyB’s Multichannel package and premium services. While not in itself a strong argument for a prohibition on buy-through, this could make possible simpler and more transparent rate card arrangements. There would of course remain the issue of ensuring that the wholesale prices charged to the cable operators and other retailers were non-discriminatory. Should cable operators also be subject to a prohibition on buy-through? 6.31 The application of controls on BSkyB in the retail market would raise the issue of whether similar obligations should apply to other retailers. There are two points to be made on this. 6.32 The first point is that BSkyB’s subscriber base accounts for two-thirds of the UK retail subscriber base – and an even higher proportion of those customers who take premium and basic services. BSkyB has three times as many subscribers as the nearest rival cable company. This means that any exclusionary effect of its retail pricing policies would be of greater potential significance than that of any individual cable company acting alone. 6.33 The second point is that if retail buy-through on DTH were prohibited, it would be logical for wholesale imposed requirements for buy-through on other platforms also to be prohibited. (Otherwise, BSkyB would be able to insist, as a wholesaler, that other retailers did something that it, as a retailer, was barred from doing). In this circumstance, it is relevant to consider whether cable companies would voluntarily impose buy-through. It seems quite possible that they would not. Cable operators have incurred, and continue to incur, large costs which are fixed in relation to the number of customers on their networks, and would wish to attract as many as possible in order to cover their fixed costs. It is probable that they would therefore wish to offer packages which were attractive to many different types of customers, including those with an interest in taking telephony and premium services, but not basic. A buy-through requirement might not be compatible with this. Duration of obligation to supply or restrictions on buy-through requirements 6.34 There are a number of developments which have the potential to change the competitive position in the UK pay television market. These include:
6.35 These developments suggest that BSkyB’s market power in wholesale premium programming may not necessarily persist in the longer term. This implies that the appropriateness of obligations for the wholesale supply of premium programming (or restrictions on buy-through) would need to be reviewed periodically. It should be stressed that the possibility that a dominant position may not be sustainable does not necessarily invalidate the case for specific regulatory rules at this stage of the development of the market – indeed regulatory intervention at this stage may reduce the need for ongoing intervention as the market matures.
Chapter 7Carriage obligations7.1 Obligations to carry channels to a minimum percentage of a cable operator’s subscriber base (minimum carriage obligations) are a form of risk-sharing between broadcaster and cable operator reflecting the need on the part of the channel provider to recover a fixed level of costs. Normally these are set at between 80 and 100 per cent of subscribers. This discussion examines minimum carriage obligations from four perspectives:
The general principle of minimum carriage obligations 7.2 There are four major arguments for minimum carriage obligations
7.3 Minimum carriage obligations are not the only possible risk-sharing arrangements nor are they only way to deal with issues of differing incentives – and indeed they may not be the best way to deal with these issues. They have important disadvantages:
7.4 Recent trials suggest that where cable companies have introduced ‘access packages’ these have had a substantial impact in improving multi-channel television penetration rates. This suggests there may be benefits to both retailers and channel providers from more flexible arrangements. With an increasing number of channels available to cable companies new forms of arrangement are likely to be needed. While some customers may wish to subscribe to all-inclusive packages, it is likely that many others will prefer more a choice of more tailored packages. Varying the level of the minimum carriage obligation 7.5 Generally minimum carriage obligations are set at 80 per cent. One option would be to reduce the level of the carriage obligation to a lower percentage. Programme suppliers’ costs are fixed. In these circumstances, for any given level of penetration, different levels of carriage obligation would be expected to lead to different per subscriber prices. Programme suppliers would be expected to offer more attractive terms depending on the number of subscribers receiving their services or make a higher charge per subscriber where they bore a greater share of the risk. 7.6 If the object of the exercise is for the channel provider to recover a fixed sum of money, then there are a large number of possible combinations of guarantees and per subscriber price which would yield the same total expected revenue. This relationship is complicated if advertising is introduced. Some industry sources suggest that the relationship between advertising revenue and viewing (and thus indirectly, carriage requirements) is not linear, with a large fall off in advertising revenue coming once carriage drops below a critical level. If advertising revenue does respond to carriage requirements in this way, then there could be a justification in some circumstances for a non-linear increase in per subscriber rates as the guaranteed carriage fell below that critical level. Alternatives to minimum carriage obligations 7.7 A move away from minimum carriage obligations would require the cable industry and suppliers of programming to find alternative arrangements which are acceptable to both sides. Such arrangements could involve fixed lump sum payments or other types of guarantee such as guaranteed minimum revenue or guaranteed minimum (absolute) numbers of subscribers. Each involves a different apportionment of risk and provides different incentives. Combinations of these would also of course be possible. 7.7 There may well be good economic arguments in favour of lump sum payments in that these face the retailer with the same cost characteristics as the channel provider – ie costs are all fixed costs and marginal costs are zero. 7.8 In considering whether fixed lump sum payments were a feasible option, one question would be how to divide up the total amount to be recovered among wholesale customers. One possible method would be to base payments by wholesale customers on the potential subscriber base, in terms of homes passed. The drawback with this would be that it would create a disincentive on cable operators to build out their networks. As they passed extra homes, the total cost of acquiring their programming would rise. An alternative which is worthy of further consideration is the total number of households in the franchise. At the point when cable operators have fully built out their networks, this might be an appropriate method on which to base channel prices. The main drawback with this approach would arise where cable operators had built only a proportion of their networks. Regulatory intervention 7.9 Oftel’s view is that there appear to be good arguments why a move away from minimum carriage obligations might be desirable. It does not necessarily follow that this should be achieved by regulatory intervention rather than through normal commercial negotiations. The individual parties are best placed to make a judgement on the most appropriate apportionment of risk. 7.10 There might be an argument for regulatory intervention in cases where carriage requirements resulted from the exercise of market power. Oftel’s view, is that, at this point, general entertainment channels are not subject to the same lack of substitutes and barriers to market entry as premium channels. Even the providers of the most popular basic channels face competition and the threat of competitive entry. Oftel has not seen evidence which would lead it to take the view that there is market power among providers of basic programming. 7.11 Oftel’s view is that at the level of individual cable companies and individual channel providers there is not a convincing case for regulatory intervention. If there were a specific problem of market power (eg in relation to the most popular general entertainment channels) this would not justify a general prohibition on minimum carriage obligations, but rather a case-by-case approach. 7.12 The second potential argument for intervention is that the effect of the network of agreements between cable companies and channel providers is to prevent new channels or channel providers from entering the market. Were this demonstrated then there might well be a case for regulatory intervention. 7.13 The third argument for intervention is that there is a ‘free rider’ problem which can only be resolved by regulatory intervention. It is argued that new more flexible agreements between cable operators and the channel providers would enable cable operators to achieve higher penetration, and in turn higher overall sales of the individual channels, but the normal process of negotiation is not capable of resolving this issue. This is because an individual channel provider agreeing to move away from minimum carriage obligations would suffer losses unless all other channel providers agreed to do so at the same time. It is argued that the effect of this ‘last mover advantage’ together with the fact that there is no common end date to contracts means that this problem cannot be resolved through commercial negotiation, even though such a move would be economically beneficial. 7.14 If there genuinely are economic benefits from a move away from minimum carriage obligations, it is not clear why such a move could not be achieved by negotiation. If the change would be economically beneficial, it must be the case that the gains would outweigh the losses (ie the gainers would gain more than the losers would lose). Therefore, it should in principle be possible to devise arrangements which ensured that no individual channel provider lost out as a result of the transition. Such transitional arrangements would possibly require a degree of coordination by various parties but there appears to be no underlying reason why they should not be achievable. 7.15 It may of course be that some channel providers are unconvinced that they would benefit from a move away from the existing arrangements and believe they would stand to lose from any transition. In these circumstances regulatory intervention would amount to second-guessing the judgement of the channel providers about the extent to which they would gain or lose. Irrespective of the merits of the arguments on the two sides there would, in Oftel’s view, have to be very compelling arguments for regulatory intervention in these circumstances. The evidence and argument so far advanced do not appear to justify such a step.
Chapter 8Deep discountingIntroduction 8.1 Deep discounting’ refers to the retailing of the second and subsequent premium channels at significantly lower prices than the first premium channel. Possible anti-competitive effect 8.2 It has been suggested that the deep discounting pricing strategy could have an anti-competitive effect. It might be exclusionary, because a competing provider of a single premium channel either at the retail or the wholesale level would have to compete against the relatively low prices of second and subsequent channels. This issue would not arise if the competing providers were always or mostly battling against the first premium channel. But at least 80% of DTH subscribers and 50% of cable subscribers take two or more premium channels (figures derived from BSkyB Annual Report, 1996), so new entrants would need either to be providers of multiple premium channels or would generally find themselves competing against the relatively low marginal prices of the second and subsequent channels. Alternatively, the issue would not arise if BSkyB were required to set the prices of its premium channels on an à la carte basis, ie a single price for each channel. 8.3 Anti-competitive behaviour typically involves a sacrifice of profit in the short run in order for the company to earn higher profits in the longer term. For example, a typical exclusionary strategy would involve a set of prices that does not maximise profit today, but by deterring entry would increase the profit earned tomorrow. The first question to be addressed, therefore, is whether it seems likely that the deep discounting price structure would be profit maximising in the short run. If so, the company would have a justification for adopting deep discounting that did not rely upon anti-competitive intent. Profitability of deep discounting at the retail level 8.4 A structure of prices set by the broadcaster that involves deep discounting is very likely to be more profitable than à la carte prices. By reducing the price of second and subsequent channels, deep discounting is likely to increase the number of channels sold to subscribers. For example, a subscriber with a relatively high willingness to pay for one premium channel (say sport), but a significantly lower willingness to pay for a second channel (say movies) would buy probably only buy one channel (sport) under à la carte prices. But under deep discounting the subscriber might buy both channels, in this example, sport as the first channel and movies as the second channel. 8.5 In this way deep discounting can allow subscribers not only to buy the channels for which they have a relatively high willingness to pay, but also channels for which their willingness to pay is much lower, which are bought as the second and subsequent channels. It is possible that the success of this pricing strategy can be seen in the high proportion of DTH subscribers that take all premium channels (72% in 1996). | ||||||||||
