COMMISSION DECISION of 27 May 1998 relating to a proceeding pursuant to Council Regulation (EEC) No 4064/89 (Case No IV/M.1027 - Deutsche Telekom/BetaResearch) (notified under document number C(1998) 1441) (Only the German text is authentic) (Text with EEA relevance) (1999/154/EC)

THE COMMISSION OF THE EUROPEAN UNION,

Having regard to the Treaty establishing the European Community,

Having regard to the Agreement on the European Economic Area, and in particular Article 57(2)(a) thereof,

Having regard to Council Regulation (EEC) No 4064/89 of 21 December 1989 on the control of concentrations between undertakings (1), as last amended by Regulation (EC) No 1310/97 (2), and in particular Article 8(3) thereof,

Having regard to the Commission Decision of 29 January 1998 to initiate proceedings in this case,

Having given the undertakings concerned the opportunity to make known their views on the objections raised by the Commission,

Having regard to the opinion of the Advisory Committee on Concentrations (3),

Whereas:

(1) On 8 December 1997 the Commission received notification pursuant to Article 4 of Council Regulation (EEC) No 4064/89 hereinafter: the Merger Regulation of a proposed concentration whereby Deutsche Telekom AG (Telekom) and BetaTechnik AG (BetaTechnik) would acquire joint control, within the meaning of Article 3(1) of the Merger Regulation, of BetaResearch Gesellschaft für Entwicklung und Vermarktung digitaler Infrastrukturen mbH (BetaResearch). The proposed concentration was to be effected through the purchase of shares.

(2) On 23 December 1997 the Commission decided to suspend implementation of the proposed concentration pursuant to Articles 7(2) and 18(2) of the Merger Regulation pending adoption of a final decision.

(3) By letter dated 22 December 1997 Germany informed the Commission pursuant to Article 9(2) of the Merger Regulation that that concentration, and also the concentration between Bertelsmann, Kirch and Premiere, threatened to create or strengthen a dominant position as a result of which effective competition would be significantly impeded on six markets in Germany, each of which constituted a separate geographic market within the meaning of Article 9(7) of the Merger Regulation.

(4) On 29 January 1998 the Commission decided pursuant to Article 6(1)(c) of the Merger Regulation and Article 57 of the EEA Agreement to initiate proceedings in this case.

(5) The Advisory Committee discussed the draft of this Decision on 6 May and 25 May 1998.

I. THE PARTIES

(6) Telekom is the public telecommunications operator in Germany. It is active, either directly or through subsidiaries, in all areas of telecommunications services, and is the owner and operator of nearly all the German cable-television networks at level 3.

(7) CLT-UFA is a joint venture between Bertelsmann AG (Bertelsmann) and Audiofina SA, into which the parent companies have merged their European television interest. These include the shareholding in Premiere Medien GmbH & Co. KG (Premiere).

(8) BetaTechnik belongs to the Kirch group (Kirch) and does mainly film post-production work such as dubbing. Kirch is the leading German supplier of feature films and television programming and is also active in commercial television.

II. THE PROPOSED OPERATION

(9) BetaResearch which at the present time is still a wholly-owned Kirch subsidiary, is the holder of exclusive, open-ended licences - granted for Germany, Austria and the German-speaking part of Switzerland - for Beta encryption technology for the encryption of programmes on the basis of the d-box decoder. The licenser of the access technology is DigCo BV, in which Kirch and Irdeto BV, owned by the South African group MIH, each have a 50 % stake. BetaResearch is active in the development of decoder software and will continue to develop the encryption and operating software for the d-box technology. It will licence this technology to pay-TV operators, providers of technical services for digital television and decoder manufacturers. It will itself manufacture conditional access modules (CA modules) and smart cards for supply to d-box manufacturers, and will carry out contract and project work for third parties; it has already received orders for such work from Stinnes and MediaGate (Thyssen Telecom).

(10) First of all CLT-UFA will acquire a 50 % interest both in BetaResearch and in BetaDigital Gesellschaft für digitale Fernsehdienste mbH ('BetaDigital`), likewise a wholly-owned subsidiary of Kirch. These transactions are part of the intended restructuring of the German pay-TV operator Premiere, which is to become a digital pay-TV programme and marketing platform using the d-box technology made available to it by Kirch. This operation forms the subject-matter of proceeding IV/M993 - Bertelsmann/Kirch/Premiere.

(11) Telekom has become a party to the arrangement between Kirch and Bertelsmann regarding the introduction of Beta encryption and decryption (Beta access) technology based on the d-box decoder. Thanks to this technology Telekom will have a technical platform for the digital distribution of pay-TV programmes via its cable network and be able to provide technical services for programme transmission. Telekom aim in taking part in the notified transaction is to ensure that it has the necessary rights in the Beta access technology. The operation is part of a complex set of agreements between CLT-UFA and Kirch on the one hand and CLT-UFA, Kirch and Telekom on the other.

III. THE CONCENTRATION

(12) Telekom, in conjunction with CLT-UFA and Kirch, will exercise joint control over BetaResearch within the meaning of Article 3(1)(b) of the Merger Regulation. In the 'Agreement on the restructuring of BetaResearch` ('the restructuring agreement`) it is stipulated that important business decisions, such as [. . .] (4), are to be taken by the proprietors' meeting acting unanimously.

(13) BetaResearch will perform on a permanent basis all the functions of an autonomous economic entity and should not lead to coordination of the competitive behaviour of Telekom, CLT-UFA and Kirch. It will not only license d-box technology to firms - such as Telekom and BetaDigital - which seek to offer technical services for digital televisions, but it will also grant licences to programme suppliers who wish to provide the necessary technical services themselves. It will grant licences to interested decoder manufacturers and supply them with CA modules, and develop decoder software for third parties. Besides its business dealings with its parent companies and with companies associated with them, therefore, it can be assumed that BetaResearch will also do business to a considerable extent with third parties and will hence have its own access to the market. BetaResearch is accordingly to be regarded as a full-function enterprise. Since only CLT-UFA is active in the field of digital encryption technology for pay-TV, through Bertelsmann's stake in Seca SA there is no risk of coordination of competitive behaviour between Telekom, CLT-UFA and Kirch.

IV. COMMUNITY DIMENSION

(14) Telekom, Bertelsmann and Kirch together have an aggregate world-wide turnover of more than ECU 5 billion. Each of them has an aggregate Community-wide turnover of more than ECU 250 million. Only Telekom and Kirch realise more than two-thirds of their aggregate Community-wide turnover within one and the same Member State, namely Germany. The concentration therefore has a Community dimension and does not constitute a case of cooperation under the EEA Agreement.

V. ASSESSMENT UNDER ARTICLE 2 OF THE MERGER REGULATION

(15) The proposed concentration affects mainly the markets for:

- technical services for pay-TV,

- cable networks.

A. RELEVANT PRODUCT MARKETS

1. Technical services for pay-TV

(16) The operation of pay-TV requires a special technical infrastructure which makes it possible to encrypt the televisions signals and to decrypt them for the authorised viewer. This is done by a decoder which is installed in the home of every pay-TV subscriber. Decoders for receiving digital pay-TV do not only decrypt the signals encrypted in the access system, but they also convert them to enable digital television signals to be received on analogue TV sets. Different decoders are used for cable and satellite TV.

(17) In addition to a decoder base, pay-TV requires a system of conditional access. This system includes the transmission of encrypted data, which contain information on the programmes or packages of programmes subscribed to and on the entitlement of the pay-TV subscribers to receive the programmes, together with the television signal, and possibly smart cards which are made available to the viewer and are able to decipher the encrypted authorisation data and transfer them to the decoder.

(18) The infrastructure described above forms the basis for the services relating to the operation of pay-TV. These involve primarily the handling of conditional access and the marketing of decoders and smart cards. Technical services for pay-TV require specific technology for the transmission of digital television signals via satellite and via cable. In the case of satellite transmission, the television signal is processed by the playout centre and sent to the respective satellite transponder, from where it can be beamed down and received directly by each satellite TV subscriber using a decoder designed for satellite reception. In the case of cable distribution, on the other hand, the processed television signals are first beamed down from the satellite transponder to a cable head-end, where they are converted for cable transmission and then fed into the cable network. Despite the separate transmission technology for satellite and cable transmission, a corresponding subdivision of the market for technical services for pay-TV into two independent sub-markets does not seem imperative. Both satellite and cable transmission require the same technical services for the operation of pay-TV. Ultimately the question can be left open, however, as it does not affect the outcome of the assessment under the Merger Regulation.

2. Cable networks

(19) Television signals can be broadcast by terrestrial transmitter, satellite or cable. In its Decision on MSG Media Service (5) (the MSG case) the Commission took the view that there were separate markets for cable TV, direct broadcasting by satellite and terrestrial transmission. It established in that case that there were considerable differences between the technical conditions and financing of the three means of transmission.

(20) From the point of view of the TV supplier the transmission of his programmes by cable is not - in Germany at least - interchangeable with satellite transmission, if for no other reason than the completely different reaches of the two means of transmission. Whereas approximately 18,5 million German households are connected to cable networks, only some 7-9 million receive satellite television. A TV supplier who broadcasts his programmes exclusively via satellite direct to the home would therefore fail to reach two-thirds of all households not receiving television solely from land-based transmitters. In particular, households in large apartment blocks cannot as a rule be reached on a direct-to-home basis as there is usually a clause in the lease restricting the use of satellite dishes.

(21) For these two means of transmission there are also considerable differences between the costs borne by a TV supplier per viewing household. The cost of analogue satellite transmission comes to approximately DEM [5-25] (6) million per transponder per year. With 7 million satellite households, this gives a cost of DEM [1-2] (7) per household. With 9 million satellite households, the corresponding figure is still DEM [0,50-2] (8). According to data supplied by the firm o.tel.o communications GmbH & Co., the cost of cable transmission comes to approximately DEM 10 000 per analogue channel per month for 500 000 households reached. This is equivalent to DEM 0,02 per month or DEM 0,24 per year per household reached. Since in the case of cable transmission the transponder costs are spread over some 26 million households (18,5 cable households and 7-9 million satellite households), the annual cost per household in this instance comes to DEM 0,38. The total annual cost per cable household accordingly comes to DEM 0,62. In the case of German viewers, cable transmission is therefore much more profitable to a TV supplier. There are therefore strong grounds for believing that, in Germany at least, the operation of cable networks still constitutes a separate relevant product market.

(22) The question can, however, be left open in the last analysis, as the competition assessment of the concentration would be the same even on the assumption of a uniform market for cable and satellite.

B. RELEVANT GEOGRAPHIC MARKETS

1. Technical services for pay-TV

(23) Technical services for pay-TV are closely bound up with the supply of pay-TV. In the MSG case (9) the Commission assumed that the relevant geographic market was confined to Germany, but at the same time it stated that, to the extent that German providers of digital pay-TV also acquired subscribers in other German-speaking regions, MSG's service market will probably also spread to such areas. In the present case, the market for technical services for pay-TV may, if there is assumed to be an overall market for satellite and cable transmission, be deemed to encompass the entire German-speaking area. In so far as technical services for pay-TV in the cable sphere constitute a separate market, however, the relevant geographic market would be confined to Germany. Since even on the assumption of the uniform market for cable and satellite the competition assessment of the concentration would be the same, the question of the relevant geographic market can be left open.

2. Cable networks

(24) The structure of the cable markets in most Member States is subject to different conditions in terms of geography, marketing and legislation (10). In the MSG case MSG Media Service (11) the Commission established that, as regards the operation of cable television networks, there was already a national German market resulting from Telekom's statutory monopoly on laying and operating cable networks in public roads. Even if the network monopoly has since been abolished and private network operators are now more active at least at level 4, Telekom still has an advantage in that it owns almost all the level 3 cable TV networks in Germany. The conditions of competition in Germany are therefore still substantially different from those in other countries. The Commission considers as a result that in the present case the relevant geographic market for cable networks is to be confined to Germany. This holds true even if the relevant product market were to include cable networks and direct-to home satellite transmission.

C. EFFECTS OF THE CONCENTRATION

1. Technical services for pay-TV

(a) Telekom's monopoly of the provisions of technical services for the transmission of pay-TV over cable networks

(25) It is to be expected that the concentration will result in Telekom's having a monopoly of the management of access control in the cable network market. Following the parallel concentration between Bertelsmann, Kirch and Premiere, BetaDigital will secure a lasting monopoly in technical services on the satellite television market.

(26) In Germany a total of 18,5 million households, or over half of all TV households, are connected to cable. By contrast, satellite television is received in Germany by only 7-9 million households. Because of the historical development of the structure of cable distribution, there are several levels of supply network in Germany. For the supply of cable television, network levels 3 and 4 are of particular importance. Level 3 is the distribution network from the cable head-end to the boundary of a given plot of land. Level 4, on the other hand, is the network infrastructure between that boundary and the junction boxes of TV households. Network 3 is operated in Germany almost exclusively by Telekom: there are 16,5 million households with cable access whose cable signals are carried over a Telekom network level 3. Telekom is also by far the biggest provider of home link-ups: for 6 million households Telekom also operates the cable network at level 4. In comparison, 2 million households are supplied via private level 3 networks, and 12,5 million are connected at level 4 via networks run by private operators. The largest private cable operator has hooked up some 2 million households, the next largest 1,2 million and 0,9 million respectively. All the other private cable operators, whose networks are dotted across Germany, each supply fewer than 0,25 million households. These figures show that Telekom has the preponderant share of the cable network market.

(27) For the time being Telekom is the only provider of technical services for digital signal transmission over cable networks. CLT-UFA, Kirch and Telekom agreed in the restructuring agreement that, when providing technical services for the transmission of digital TV programmes over its cable networks, Telekom would use exclusively Beta access technology on the basis of the d-box decoder. They also stipulated in that agreement and in the 'understanding on a cable platform for digital television neutral in its effect on programme suppliers` that Telekom was to receive the exclusive right to use Beta access technology to the extent that it was to be granted that right in respect of the cable networks. It is only to other content suppliers that this exclusivity does not apply. This in itself rules out the possibility of other cable operators being granted a licence for BetaResearch's Beta access technology and thereby being able to compete with Telekom in the area of Telekom's cable networks.

(b) The lasting nature of the dominant position

(28) Telekom will remain the sole provider of technical services for the transmission of pay-TV over cable networks in Germany in the short term, and that it will also secure a monopoly position in this area on a lasting basis. This assessment is based on the following considerations.

(i) Premiere will be the only programme platform in Germany in the long term.

(29) Access to programme rights is even more important in the case of pay-TV than it is in the case of free TV. In order to entice the consumer to subscribe, or to take particular productions on a pay-per-view basis, certain specific types of content are required. Experience with pay-TV in Europe so far shows that in order to be sufficiently attractive a pay-TV package must include a combination of premium rights for the first broadcasts of films produced at the major Hollywood studios and for popular sporting events. This is especially true in Germany, where every household with television can already receive more than 30 free-access channels - unlike, for example, - France or the UK. The importance of premium films and sporting events for the organisation of pay-TV is borne out by the programme costs estimated in Premiere's business plan. About [50-80] % (12) of all licensing costs are to be for premium films (about [20-50] % (13)) and sports rights (more than [10-40] % (14)). An alternative programme platform can be built up only by a person with access to premium content so as to be able to offer a promising pay-TV package.

(30) But programme resources are in short supply since, as a general rule, the broadcasting rights for premium content of this kind are given on the basis of longer-term exclusive contracts. In the German-speaking area CLT-UFA and Kirch hold [. . .] (15) of pay-TV broadcasting rights for premium films, as a results of output deals with the Hollywood majors, and [. . .] (16) of sports rights too. As CLT-UFA and Kirch are to transfer their pay-TV broadcasting rights to Premiere, the transaction will initially make Premiere the only operator to dispose of the programme content for a premium channel. For the duration of the existing contracts no other supplier will have access to Premiere's programme resources. Thus potential competitors will not be in a position to create an alternative programme platform for the German-speaking area. They might be able to show particular programmes in specific pay-TV niches, but to do so they would be dependent on Premiere's programme and marketing platform.

(31) Nor is it likely that the situation will change in any way once the current contracts for pay-TV rights have expired. In theory, pay-TV services could potentially be supplied in Germany and the German-speaking area by other European and non-European television operators, such as BskyB or Canal+, or even the major Hollywood studios. But in order to establish an alternative programme platform it is not enough to conclude just a few individual output deals. The output deals concluded by Kirch and CLT-UFA run for different terms. They cover periods of [1-10] (17) years. Potential competitors would have to succeed in acquiring rights under several output deals over a number of years before they could offer an attractive pay-TV package.

(32) Given the market position Premiere will by then have established, competitors' prospects of securing any substantial volume of attractive pay-TV broadcasting rights will be small. This is because, to have any hope of acquiring broadcasting rights, it is vital to have access to viewers in the form of an established subscriber base since rightholders usually want to see their product distributed widely. This is true of premium films, and even more so of sporting events. In addition, the prices of pay-TV rights are usually determined by reference to the number of subscribers, subject to a stated minimum. This is certainly what is done in output deals. As Premiere's subscriber base can be expected to be large in the next few years, it follows that a sale of rights to Premiere should secure a considerably higher price than sale to a newcomer. And a newcomer would run a considerable financial risk by concluding output deals since he would have to guarantee a minimum subscriber base, without knowing whether he could achieve the guaranteed figure. In view of the advantages which Premiere will consequently enjoy over any potential competitor in the negotiation of contracts for premium content, it is unlikely that outsiders will be able to secure adequate access to content of that kind.

(ii) Following the transaction, d-box technology will in practice become the digital standard in the German-speaking area.

(33) Pay-TV requires special technical infrastructure. This can be proved either by pay-TV operators or by other parties, mainly cable operators. Because of the structure of cable networks in Germany, private cable operators are by themselves unable to provide the technical infrastructure for the transmission of pay-TV. Their 'cable islands` are mostly too small to justify the investment necessary to acquire their own conditional access and alternative decoder system for pay-TV. Moreover, private cable operators control only parts of the level 3 and 4 networks involved in the distribution of cable televisions. At network level 3 - that is, from the cable head-end, where the digital programme signal is received and fed into the cable network, to the boundary of the individual house - the great majority of the networks belong to Telekom. Private cable operators are at present dependent on upstream level 3 services supplied by Telekom. This makes it impossible for private cable operators to develop alternative facilities for the transmission of pay-TV without involving Telekom. The private cable operators' islands would have to be linked by level 3 supply networks before they could set up an alternative infrastructure.

(34) After the transaction, Premiere will distribute digital television through d-box technology and d-box decoders. Telekom will supply services for pay-TV by cable on the basis of Beta access technology and d-box decoders. Thus those companies which would have been qualified to set up an infrastructure for digital television and to provide the corresponding services have committed themselves to Beta access technology based on the d-box decoder. It must therefore be accepted that, in the German-speaking area, there will be no alternative technical platform for digital television in the foreseeable future.

(iii) The monopoly position of Premiere's programme platform, and the fact that Telekom is to feed it into the cable networks using the d-box technology, forms a long-term obstacle to the introduction of any other technology.

(35) The installation of an alternative technical infrastructure for the transmission of pay-TV would require a major investment. However, other potential suppliers would be prepared to make that investment only if there were corresponding opportunities for market penetration. That would be so only if it were possible for a second pay-TV operator to set up in Germany. Such an operator could either create his own technical platform, on the basis of alternative access technology, or offer another party an opportunity of developing a technical infrastructure. But as has already been indicated, it is improbable that a second pay-TV operator will enter the market, given Premiere's established position, with its subscriber base and, especially, its programme resources.

(36) Moreover, alternative access and decoder technology could be developed only for satellite transmission, since CLT-UFA, Kirch and Telekom have stipulated in the restructuring agreement that Telekom, in providing technical services for digital transmission of TV programmes on its broadband cable network, will rely exclusively on Beta access technology on the basis of the d-box decoder. This means that potential pay-TV suppliers and other potential providers of conditional access services would in any case be obliged to use Beta access technology and the d-box decoder as far as the Telekom cable network is concerned.

(iv) All potential conditional access operators will be dependent on BetaResearch's licensing policy

(37) BetaResearch is the licenser of the conditional access system applied in the d-box. The parties have stated that a proprietary conditional access system was developed for security reasons. Other pay-TV operators, such as Canal+ and BSkyB, who already have experience in the pay-TV market, also prefer to rely on a proprietary access control system for better protection against breaches of the security of data transmission. Under the DVB standard, there are two processes available to prevent a viewer who receives pay-TV subscriptions with different CA systems from having to use several decoders, Simulcrypt and common interface ('CI`).

(38) With CI, widely varying conditional access systems can be used in the same decoder, enabling all other pay-TV operators and service providers to operate access controls using the existing decoder base. It is not necessary with CI, as it is with Simulcrypt, to conclude an agreement with the person who installed the decoder base on the linking of one's own access-validation system with the proprietary access system of the decoder. If a proprietary conditional access system is to be used, there must be discrimination-free access to the system. In the Commission's view, this requires that the licenser of the decoder technology be able to take his business decisions without being influenced by a programme supplier. In the present case the licenser is not independent, since BetaResearch is controlled mainly by enterprises which have their own interests as programme suppliers.

(39) If all existing and potential suppliers of digital pay-TV, and Telekom as provider of technical services for cable, have committed themselves to Beta access technology and the d-box decoder, it is impossible for an alternative decoder base to be established in Germany. Any other potential provider of conditional access services will have to use the d-box decoder. The d-box uses a proprietary coding system developed by BetaResearch. Potential providers of technical services for pay-TV would therefore have to seek a licence for Beta access technology from BetaResearch. This is the case with both satellite and cable transmission. In managing conditional access for satellite and cable, however, BetaResearch would have no interest in exposing BetaDigital or its joint parent Telekom to competition on the market in technical services for pay-TV. BetaResearch might therefore use its licensing policy to hamper other service providers' access to the market.

(v) Given the present structure of German cable networks, private cable operators would not be able to provide technical services for pay-TV in their networks.

(40) The parties argue that private cable operators would be able to build up and operate a technical platform in competition with Telekom in their networks. This cannot be accepted. The private operators normally handle small level 4 'cable islands` which are often unconnected between themselves and are supplied by a 'backbone` level 3 network belonging to Telekom. This means that private cable operators are usually dependent on upstream level 3 services supplied by Telekom. The same applies to the provision of technical services for pay-TV. In Telekom's cable networks, the signals broadcast by programme suppliers are received centrally at the Telekom playout centre in Usingen, and after processing are distributed from there to the head-ends of the Telekom cable networks. The private cable operator's islands would have to be linked by level 3 supply networks before they could set up an alternative infrastructure. As a rule the construction of a network head-end will not be an alternative, at least for private cable operators who operate only at level 4. Given the fragmented nature of the networks, the investment would not be economically justified.

(41) The possibility that private cable operators might be able to offer technical services for pay-TV in their networks is therefore purely theoretical. If they did want to offer pay-TV technical services in their networks, however, the planned introduction of Beta access technology as the digital standard would mean that they would have to approach BetaResearch for licences to use the Beta technology. It has been explained already, however, that in managing conditional access in the cable field BetaResearch would have not interest in exposing its joint parent Telekom to competition on the market in technical serviced for pay-TV. BetaResearch would have no interest in exposing its joint parent Telekom to competition on the market in technical services for pay-TV. BetaResearch might therefore use its licensing policy to hamper other service providers' access to the market.

(vi) The merger should also be assessed positively in the light of Telekom's neutralising effect on BetaResearch

(42) Telekom has claimed that, through its holding, it will be able to exert a neutralising influence on the competitive conduct of BetaResearch, which would otherwise be controlled solely by the pay-TV operators CLT-UFA and Kirch. Against this claim, it should first be stated that CLT-UFA and Kirch will in any case have a combined two-thirds holding in BetaResearch and that satellite TV is largely outside Telekom's sphere of influence. As long as Telekom is able to exert specific rights of influence in respect of the further development of the d-box, it will have an interest in influencing that development in such a way as to maintain its position as the sole operator of the technical platform for cable TV. The same applies to BetaResearch's licensing policy.

(43) Moreover, Telekom's claim that its acquisition of a holding in BetaResearch actually has a positive impact on competition is conditional on Commission's approval of the merger of the digital activities of Bertelsmann and Kirch, which is the subject of the Bertelsmann/Kirch/Premiere case. However, together with this Decision, the Commission has also declared the proposed parallel Bertelsmann/Kirch/Premiere merger to be incompatible with the common market. In so doing, it has also prohibited the initial 50 % holding by CLT-UFA in BetaResearch covered by the notification in the Bertelsmann/Kirch/Premiere case. Approving this merger proposal involving the acquisition of joint control of BetaResearch by CLT-UFA, Kirch and Telekom would mean that CLT-UFA would also be able to obtain partial control of BetaResearch. It would therefore have the consequence that the only two pay-TV operators on the market would together control the decoder technology, including the CA system. For that reason alone, it is not possible to claim that this merger has a positive impact on competition.

(c) Conclusion

(44) On the foregoing grounds, it is to be expected that the proposed transaction will give Telekom a dominant position on a lasting basis on the German market in technical services for satellite pay-TV in the cable field. If it is considered that there is one market in technical services for both satellite and cable TV, the transaction, taken together with the parallel concentration between Bertelsmann, Kirch and Premiere, will produce a dominant duopoly on this broader market in the German-speaking area since, given the number of television viewers there, Germany is by far the largest component in the German-speaking area. Neither Telekom nor BetaDigital are exposed to any significant competition on this market. Moreover, in view of their common technology and of their legal links with BetaResearch, there are no grounds to expect that there will be competition between Telekom and BetaDigital.

2. The market in cable networks

(a) Telekom's dominant position in broadband cable networks

(45) As has been explained in more detail in paragraph 26, the bulk of the level 3 and 4 cable networks used for the supply of cable television belong to Telekom. Even if Telekom's monopoly over the networks has now come to an end and private network operators are more active, at least on level 4, it can be presumed that Telekom continues to hold a dominant position.

(46) Telekom would still hold a dominant position if in the future it were to be considered that cable and satellite transmission formed one market. As we have seen, there are 18,5 million households connected to the cable network in Germany. At level 3 Telekom controls 16,5 million of these connections. By contrast, only 7 to 9 million households receive television direct by satellite. If a television channel is to be successful in Germany, therefore, it has to be fed into Telekom's cable network; that is the only way to reach a sufficient number of viewers. This is borne out by the example of DF 1. Until November 1997 DF 1 was not carried on the cable network. This is generally considered to be one of the reasons why the number of subscribers to DF 1 fell far below expectations.

(47) On 24 November 1997 the management board of Telekom decided that the entire broadband cable business, including the broadband distribution network, should be made organisationally separate from the rest of the company. The objective of this restructuring operation was ultimately to hive off Telekom's networks to one or more new companies to be set up in partnership with other operators. The final plan, still to be defined, would be implemented with effect from the beginning of 1999. The details of the intended restructuring operation and the time-scale for implementation are not yet certain. In particular, it is not clear whether and, if so, to what extent outsiders may be able to participate in Telekom's cable networks.

(b) Strengthening of Telekom's dominant position

(48) The proposed concentration between Telekom and BetaResearch will strengthen Telekom's dominant position in cable television networks at the expense of the private cable operators.

(49) The cable operators approached by the Commission have stated almost without exception that they intend to use their cable networks for as wide a variety of new services as possible. The main such uses are digital television, internet access, multimedia services, and telecommunications services including voice telephony. The Commission's inquiries have shown that the realisation of these plans is threatened by the proposed transaction.

(50) To set up the technical infrastructure needed to be able to use the cable network for these services requires substantial investment. At network level 4 alone, the investment needed is expected to amount to between DEM 150 and DEM 250 per household. In addition, private cable operators could reasonably develop level 4 only if level 3 is equipped correspondingly. Telekom has so far shown no inclination to undertake any such operation. If the private cable operators want to increase the value added in their cable networks by providing additional services, they will have to expand their activities to level 3 in competition with Telekom. They are now legally permitted to do so, but this too would require substantial investment.

(51) The private cable operators would have to finance these investments in the first place form their core business, television. But to do so they would have to be in a position to take part in the added value chain represented by digital television. The technical platform for digital pay-TV already rules this out: only Telekom is to operate the conditional access system on the cable network. If private cable operators did want to offer pay-TV technical serviced in their networks, however, the planned introduction of Beta access technology as the digital standard would mean that they would have to approach BetaResearch for licences to use the Beta technology. In managing conditional access in the cable field, BetaResearch would have no interest in exposing its joint parent Telekom to competition on the market in technical services for pay-TV. Even more important is the fact that the distribution structure for digital pay-TV which is the objective of the proposed transaction will make it impossible for private cable operators to market pay-TV. In connection with the proposed transaction Telekom has adopted a transparent transmission model for digital pay-TV, as it had already done for analogue pay-TV. This means that Telekom has given up the possibility of marketing Premiere's digital programmes itself. The consequence is that private cable operators will not be in a position to introduce a marketing model in which they act as suppliers of pay-TV to the ultimate consumer. This might have been possible in cooperation with Telekom even with the present structure of the cable networks, if Telekom had not confined itself to a purely carrying function in the pay-TV field. And the private cable operators could market pay-TV only if they were to expand their activities to level 3. Marketing by cable operators would also have been possible once Telekom had implemented the restructuring plan described above, if after the hive-off its cable networks were operated with the participation of private cable operators. But Telekom's opting for a transport model excludes all these possibilities.

(52) In particular, it removes the possibility of an expansion into level 3 on the part of the private cable operators, because if they are confined to transparent transmission they will not be able to finance the necessary investment. Once Telekom has decided in favour of this model, the private cable operators have no prospect of securing Premiere's agreement to a marketing model of their own. Telekom controls by far the largest proportion of level 3, so that Premiere is in a position to refuse to allow the private cable operators to market its programmes without having to fear any serious loss as a result. There is a theoretical possibility that the private operators might impede transmission at network level 4, but this does not provide any real counterweight. They would not be able to justify filtering out Premiere programmes to their customers for any length of time. This will apply with even greater force if digital pay-TV develops strongly in future.

(53) It follows that Telekom's opting for the transport model has made it impossible for the private cable operators to finance an expansion of their activities, and that no such expansion will take place in the foreseeable future. This eliminates the potential competition of the private operators at network level 3, and protects and strengthens Telekom's dominant position. It also prevents the development of the cable networks needed if new services are to be introduced, and thus the introduction of such services in the foreseeable future. The scope for competition on the part of the private operators is further limited, even at network level 4, and Telekom's dominant position is strengthened.

(c) Conclusion

(54) On the foregoing grounds, it is to be expected that Telekom's acquisition of a holding in BetaResearch, and the agreements with Bertelsmann and Kirch which Telekom has concluded in connection with that acquisition, will significantly restrict the scope for competition on the part of the private cable operators, and eliminate competition at network level 3. The transaction will also establish a distribution structure in the German cable network under which there is little prospect that in the event that Telekom's cable activities are hived off, private cable operators would be able and willing to participate in Telekom's cable networks. The transaction is consequently likely to strengthen Telekom's dominant position on the cable network in Germany on a lasting basis.

VI. UNDERTAKINGS PROPOSED BY THE PARTIES

1. Undertakings

(55) By letters dated 28 April and 4 May 1998 the parties suggested a series of undertakings intended to remove the doubts about the proposed Bertelsmann/Kirch/Premiere and Deutsche Telekom/BetaResearch concentrations. These consist basically of the following points:

(a) Cable operators

(56) The parties to the merger are prepared to cooperate with the cable operators on sales. As well as advertising for new customers, this includes marketing information and support. Customer relations will still be Premiere's responsibility, however. Cable operators will receive appropriate financial compensation for their efforts.

(b) BetaResearch

(57) Telekom will set up a technical council of experts, which will be accessible to all firms involved in digital television. The council's recommendations will be implemented in accordance with the already existing agreement between Bertelsmann, Kirch, Telekom, ARD and ZDF. The shareholders of BetaResearch are, moreover, prepared to allocate 25 % of the company's capital to third firms via an intermediate holding company. No veto rights will be associated with this, however. The present veto rights of Bertelsmann, Kirch and Telekom will continue. The leading role of Kirch, in particular as regards management occupancy, is again emphasised by the parties.

(58) BetaResearch will on request immediately grant every customer who intends to perform encryption services for itself or for third parties a conditional access licence (compulsory licence) on the basis of a generally accessible standard contract. If no agreement is reached on the conditions, in particular the licence fees, an arbitration body will take a final decision as to their appropriateness.

(59) BetaResearch undertakes to disclose the API interface of the d-box network (native API) by the end of 1998 and to defer to an arbitration body in the event of licensing disputes. BetaResearch also undertakes to supplement the native API with the interface to be standardised by the DVB, as soon as the standard. BetaResearch, moreover, will adopt all future DVB standards.

(60) BetaResearch will grant manufacturing licences to any interested manufacturer on standardised conditions and defer in the event of disputes about those conditions to an arbitration body. A 'technical verification test` by BetaResearch is the condition on which the decoders may be supplied on the market. Manufacturers may be certified after two to three years by independent third parties.

2. Assessment

(a) Cable operators

(61) The parties have offered cooperation with the cable operators only in the area of sales. Cable operators are thus prevented from marketing Premiere on their own and thereby from developing customer relations and unbundling and repackaging programmes. For the cable operators to develop an alternative programme platform, other than by cooperating with Telekom given the existing network structure, they would require at least to be able to obtain the individual programme packages offered by Premiere, in particular the premium film and sports channels, bundle them with third-party channels and offer them to their cable customers.

(b) BetaResearch

(62) Telekom's offer of setting up a technical council of experts is simply a question of opening up the existing council of experts with ARD and ZDF to third parties. It should be noted that, under the agreement with ARD and ZDF, this council of experts is able to discuss issues relating to technological development but cannot require operators to comply with its recommendations.

(63) The possibility for third parties to participate in BetaResearch, as proposed by the parties, might give rise to some transparency in respect of decisions concerning future technological development. However, since this participation is to be limited to 25 % and the intermediate holding company is not to be granted any veto rights, Bertelsmann and Kirch, as the dominant pay-TV suppliers, and Telekom, as the dominant cable operator, retain joint control of BetaResearch and, consequently, of technological development.

(64) The introduction of a compulsory conditional access licence and an arbitration procedure for settling disputes concerning conditions, in particular the level of licence fees, may indeed go some way to ensuring that third parties are not subject to discrimination where licensing is concerned. However, this undertaking again does not alter the fact that the technological development for which licences are granted will be controlled by Bertelsmann, Kirch and Telekom. The undertakings that BetaResearch will disclose the API interface of the d-box network by the end of 1998, that, in the event of licensing disputes, it will defer to an arbitration tribunal, and that it will supplement the native API with the interface to be standardised by the DVB as soon as the latter has been approved are essentially confirmation of intentions which the parties had in any case stated. Furthermore, although BetaResearch's undertaking that it will grant manufacturing licences to any interested decoder manufacturer on standard conditions and defer to an arbitration body in the event of disputes concerning those conditions might well contribute to creating competition with regard to the manufacture of decoders, it does not alter the fact that Bertelsmann and Kirch, as the dominant pay-TV suppliers, and Telekom, as the dominant cable operator, control the development of decoder technology.

(c) Overall assessment

(65) On the basis of the proposed undertakings, the parties do indeed guarantee a degree of technological transparency and access to d-box technology. However, the parties are not willing to surrender their absolute control of this technology, and in particular of its further development. Nor will the proposed undertakings give any alternative programme and marketing platform a realistic chance. At the same time, without the chance of an alternative programme platform, the undertakings with regard to technology, or at least to licensing for the purposes of controlling access, are rendered even more meaningless by the fact that, without a second programme platform, no alternative technological platform can be expected to be developed.

3. Further undertakings

(66) On 13 May 1998, the parties supplemented and widened the proposals they had made, as follows:

(a) Cable operators

(67) In addition to the cooperation on sales already offered, Premiere will, on certain conditions, also allow cable operators to market its programmes and thus develop their own customer relations. Free-TV, pay-TV and pay-per-view programme suppliers are in principle excluded from marketing. Marketing will not be exclusive, i. e. Premiere and the cable operators will market Premiere in competition with each other. Generally speaking, each party will pursue customer relations with those subscribers acquired by it. The data of customers acquired by the cable operators would, however, have to be made available to Premiere's subscriber management system to that Premiere could provide the customers in question with additional services (such as pay-per-view) and general customer information. Cable operators will not be allowed to market Premiere's pay-per-view services.

(68) The cable operators will have to offer the same combination of packages as Premiere. They will not be allowed to unbundle programmes within individual packages or Premier's package structure. However, cable operators will be able to offer their own services together with the Premiere package and market the packages and individual programmes of third-party suppliers. Because Premiere no longer insists on linking its basic package with its premium package, the cable operators' customers will have the opportunity of subscribing to a basic package offered by the cable operator alongside the Premiere package combination. However, subscribing to such a basic package must not be made a condition for subscribing to Premiere programmes. The price payable by cable operators for the Premiere package will be calculated on the basis of the transfer price of Premiere on the cable market minus the costs which Premiere will have saved on marketing. Cable operators will, like Premiere, have to offer the d-box decoder for rental and, while supplies last, make use of decoders from the Nokia quota (1 million units).

(69) Cable operators must not abuse the freedom they have been offered in order to limit their marketing operations in capacity terms with regard to Premiere programmes, i. e. they must agree to market the Premiere package in toto. Otherwise Premiere will not be obliged to make its programmes available for marketing. In the event of capacity shortages, Premiere and the cable operators will either be bound by the instructions of the Land Media Institutes (Landesmedienanstalten) or reach agreement on another objective capacity-allocation procedure.

(b) Cable networks

(70) Telekom also stated on 20 May 1998 that it will keep, until 31 December 1999 at the latest, two digital channels on the hyperband open for use by a potential third-party programme supplier. Telekom is also willing, together with level-4 operators, to expand cable capacities on band IV, but only where refinancing is possible for cable operators (levels 3 and 4).

(c) BetaResearch

(71) As far as the offer to allow third-party firms to acquire a holding in BetaResearch via an intermediate holding company (see recital 57), all shareholders, i. e. Bertelsmann, Kirch, Telekom and the holding company, will have the same rights. For the purposes of obtaining a majority in the event of voting, the statutory rules will apply (no specific quorum). The special rights previously granted to individual shareholders no longer apply.

4. Assessment

(a) Cable operators

(72) The new undertakings proposed by the parties do not create the conditions for the development of an alternative programme and marketing platform or an alternative technical platform for cable operators. The parties themselves point out in their accompanying letter that 'the partners for the establishment of these marketing forms on an economically plausible basis do not at present exist on the German market or at least have not in the past shown any readiness to carry out the necessary investments to develop the corresponding infrastructures`. With the cable network structure in its present form, it is indeed not possible for private cable operators to set up a programme and marketing platform and technical platform of their own.

(73) As indicated in recital 26, private cable operators are predominantly active at level 4 (home link-ups, cable islands). The level-3 networks (cable head-end to the boundary of a plot of land) are almost entirely owned by Telekom. Private operators are thus generally dependent on obtaining upstream level-3 services from Telekom. Without operating at level 3, however, it is not possible for private cable operators to create an alternative programme platform. Given the fragmentation of the German cable-TV market and Telekom's position as virtually the sole operator of level-3 networks, the private cable operators are not at present able to offer an alternative programme platform which would compete with Premiere.

(74) In the course of this proceeding, Telekom has not shown any willingness to cooperate with private cable operators in creating an alternative cable platform or at least allow them access to level-3 services. It is true that, when asked on 22 May 1998 how the separation between levels 3 and 4 could be overcome, Telekom stated that, for the purposes of making available its services relating to the digital cable platform and digital marketing platform, contracts would be concluded between Telekom and interested operators (programme suppliers, level-4 operators). If level-4 operators so wished, the technical and operational separation of levels 3 and 4 could also be overcome by combining the services of level-3 and level-4 operators. However, these statements are so vague and general that they do not provide adequate reason to expect that the separation between levels 3 and 4 can be overcome for the purposes of enabling level-4 operators to establish a programme and marketing platform. In particular, as far as combining the services of level-3 and level-4 operators is concerned, it is not explained which services could be provided by Telekom and which by the level-4 operators, nor how they could indeed be combined in technical and economic terms. It is still completely unclear whether level-4 operators will have their own access to level 3 and whether, in particular, they could also operate conditional access and how this could be made technically possible. On the basis of Telekom's statements, it must therefore continue to be assumed that Telekom ultimately wants to be the sole provider of the necessary technical services, and that, consequently, level-4 operators will be dependent on it. Above all, this means that there will be no possibility of an alternative platform emerging because only Telekom will operate the conditional-access system within the cable network. Overcoming the separation of levels 3 and 4 would, in all probability, only be possible after a restructuring and privatisation of Telekom's cable networks, whereby they would be operated by regional companies in which private cable operators would have a stake. Telekom has not, however, made any binding statements about when and how such restructuring might take place. It will therefore probably not occur until such time as Premiere is so ahead of the field in terms of its subscriber base that it will be difficult for alternative platforms to gain a foothold.

(75) In addition, cable operators will not, on the basis of the proposed undertakings, be in a position to offer pay-TV programmes on the same competitive terms as Premiere. This is because they will not be allowed to supply programmes for pay-TV and, for example, offer regional programmes produced by themselves. Rather, they will be limited, in their ability to offer additional services, to marketing third-party channels. Moreover, they will not be allowed to market Premiere's pay-per-view services nor offer any pay-per-view services of their own. However, given the lack of exclusivity on pay-per-view rights, allowing them to offer pay-per-view services would be the easiest way of giving them access to the pay-TV market. Cable operators must also offer the Premiere package in the same combination as Premiere itself. This means that they will be completely dependent on Premiere as far as the essential parameter of bundling is concerned.

(76) Cable operators are further required to make their customer data available to Premiere without any corresponding obligations on Premiere's part. However, the supply of such data would be an entirely uncommon practice in this sector, and would provide Premiere with a considerable competitive advantage. Pricing on the basis of the Premiere transfer price minus the costs saved on marketing gives rise to the risk that cable operators will not be able to market Premiere programmes economically. It remains entirely unclear who will determine these cost savings and what criteria will be used in so doing.

(b) Cable networks

(77) The reservation of two digital channels in the hyperband for use by potential third-party programme suppliers might theoretically help make it possible for competing programmes to be supplied given the additional capacity thus made available, but it is not in itself likely to open up an opportunity for the emergence of an alternative programme platform for cable. This is true solely by dint of the fact that the reservation is limited in time. It would be extremely difficult in only one and a half years to set up an alternative programme platform. Since, moreover, private cable operators are not, if they wish to market Premiere, allowed to supply pay-TV themselves, they will be restricted to offering third-party channels and thus be considerably handicapped in their ability to set up a programme platform which could use the two channels. On the other side of the coin, a programme supplier who wished to set up a programme platform on the basis of the 25 % of pay-TV rights could not make use of those rights until 2000. Moreover, in order to use the two channels, a programme supplier would have to sign a contract for ten years even though he has access to the 25 % of pay-TV rights from the output deals, as offered by the parties, only until the end of 2002. The creation, as proposed by Telekom, of additional capacities by extending band IV has not been described in any more concrete terms and should therefore be considered merely a general statement of intent.

(c) BetaResearch

(78) The undertaking that all shareholders in BetaResearch, including the intermediate holding company for third parties, will be granted the same rights might in principle solve the problem of the control of technology by programme suppliers. If CLT-UFA and Kirch no longer have the right of veto and all shareholders have the same rights, shifting majorities are, from a formal point of view, possible within the bodies of the company. However, it should be borne in mind that BetaResearch will only have four shareholders since third parties will be granted a share solely via an intermediate holding company. Majority decision could therefore be taken only with the approval of at least three of the four shareholders. This means that CLT-UFA and Kirch, acting together, would always be able to block decisions and thus continue to ensure that their interests win the day by preventing any decision which goes against those interests from being adopted. The intermediate holding company would, by contrast, only be able to enforce its interest against CLT-UFA and Kirch in conjunction with Telekom if both of them voted against CLT-UFA and Kirch and thereby prevented a majority from being reached. It is obvious, however, that Telekom, given that it will supply technical services for pay-TV on the basis of the d-box technology, is more likely to have interests which converge with those of CLT-UFA and Kirch than with those of the intermediate holding company for third-party firms. This means that the possibilities of CLT-UFA and Kirch, as dominant pay-TV suppliers, and Telekom, as dominant cable operator, of jointly controlling the technology and its further development may be limited from a formal point of view but might nevertheless remain in practice.

(d) Overall assessment

(79) It follows from the above that the proposed undertakings, even in their expanded version, are inadequate to solve the existing competition problems. They are unlikely to give private operators a realistic chance of establishing an alternative programme and marketing platform. This also means, however, that an alternative technical platform will fail to emerge in the cable network because only Telekom will operate the conditional-access system in that network. It is true that opening up the possibility for third parties to take a holding in BetaResearch and the abandonment of the veto and special rights of the existing shareholders is an important concession since it will enable the structurally safeguarded control of the technology and its further development to be lifted. However, since no alternative technical platform will emerge without the chance of an alternative programme platform, the undertaking regarding BetaResearch is inadequate, even in connection with the further undertakings proposed by the parties concerning the CA licence, the API and the manufacturing licence, to prevent Telekom form, in the long term, dominating the market for the supply of technical services for cable pay-TV and strengthening its dominant position in the cable network to the detriment of private cable operators.

VII. CONCLUSION

(80) For the above reasons, the Commission has reached the initial conclusion that the proposed transaction would create or strengthen dominant positions as a result of which effective competition would be significantly impeded in a substantial part of the Community. The merger should consequently be declared incompatible with the common market under Article 8(3) of the Merger Regulation,

HAS ADOPTED THIS DECISION:

Article 1

The notified concentration consisting of the acquisition of joint control by Deutsche Telekom AG, CLT-UFA SA and BetaTechnik GmbH of BetaResearch Gesellschaft für Entwicklung und Vermarktung digitaler Infrastrukturen mbH is declared incompatible with the common market and with the functioning of the EEA Agreement.

Article 2

This Decision is addressed to:

Deutsche Telekom AG

Postfach 2000

D-53105 Bonn

CLT-UFA SA

Boulevard Pierre Frieden, 45

L-2850 Luxemburg

BetaTechnik GmbH

Betastraße 1

D-85774 Unterföhring.

Done at Brussels, 27 May 1998.

For the Commission

Karel VAN MIERT

Member of the Commission

(1) OJ L 395, 30.12.1989, p. 1. Corrigendum (OJ L 257, 21.9.19990, p.13).

(2) OJ L 180, 9.7.1997, p. 1.

(3) OJ C 57, 27.2.1999.

(4) This version of the Decision has been edited to ensure that confidential information has not been disclosed.

(5) OJ L 364, 31.12.1994, p. 1.

(6) Loc. cit. (footnote 5), paragraphs 52, 53 and 54.

(7) Nordic Satellite Distribution (OJ L 53, 2.3.1996, p. 20).

(8) Loc. cit (footnote 5), paragraph 54.