POSITION PAPER BY PORTUGAL TELECOM ON THE DRAFT COMMUNITY GUIDELINES FOR THE APPLICATION OF STATE AID RULES IN RELATION TO RAPID DEPLOYMENT OF

Transcription

1 POSITION PAPER BY PORTUGAL TELECOM ON THE DRAFT COMMUNITY GUIDELINES FOR THE APPLICATION OF STATE AID RULES IN RELATION TO RAPID DEPLOYMENT OF BROADBAND NETWORKS

2 Introduction Portugal Telecom (PT) welcomes the opportunity to express its views on the draft Community Guidelines for the application of state aid rules in relation to rapid deployment of broadband networks and shares the Commission s view as to the benefits of increased legal certainty and transparency in this field as regards the Community decision-making process. The recent development of fixed and mobile broadband networks in the EU has changed both the business environment as well as the private landscape of European citizens. This accomplishment has been achieved by means of a balanced combination of significant risktaking and investment by the private sector and the growing awareness by the public sector of the importance of information and communication technologies for the economy and society in general. The development of next generation access networks (NGA) will enable EU countries to take another leap forward into a full-fledged information society. PT acknowledges the support given by the Commission to the widespread availability of broadband services for all European citizens as laid down in the Lisbon Strategy and is fully committed to play an active role in the European Economic Recovery Plan of November 2008 as a means to drive Europe s recovery from the financial and economic crisis. PT is aware that the Portuguese Government envisages supporting investment in NGA and has willingly signed a Protocol entered into by the Government and a number of communications operators for the development of NGA in Portugal. Furthermore, PT has started on its own initiative a massive investment in optical fibre throughout the Portuguese territory and will have deployed km of fibre and connected 1 million homes by the end of PT therefore wishes to contribute hereby to the establishment of a State aid policy which helps overcoming the digital divide and is supportive of private investment leading to the creation of jobs, sustainable economic growth and further deployment of broadband in rural and underserved areas. Paramount to this objective is a level playing field in which nondiscrimination, technological neutrality and technical accuracy are put to the service of fair and undistorted competition in the EU. 2 / 16

3 Our comments will cover the following topics: 1. Definition of NGA 2. Technological neutrality 2.1. Discriminatory bundling obligation 2.2. Discriminatory imposition of network architecture 2.3. Exclusion of mobile and wireless technologies 3. Wholesale access 3.1. A priori obligations 3.2. Equivalent burden 3.3. Duration 3.4. Price 4. Timeframe for definition of areas 5. Last mile 6. Claw-back mechanism 7. Regulatory framework Conclusions 3 / 16

4 1. Definition of NGA The definition of NGA provided for at 48 of the Draft Guidelines 1 is rather limited in scope and risks excluding crucial components of a NGA from the benefits of admissible State aid. This is namely the case of investment in core networks, service platforms and customer hardware. It is important to bear in mind that State aid will benefit rural and underserved areas which will be granted access to new services and solutions under technical and economic conditions that would otherwise only be available over a much longer period or would never be available at all. Therefore, State aid should take into consideration the need to upgrade the core network with fibre capacity in order to guarantee full quality and high level performance of the overall network. State aid targeted at supporting the development of service platforms that respond certain needs of the population and developing services of common interest should also be allowed. Finally, under some circumstances customer hardware should be subsidized, especially in those regions where the economic capacity of consumers is far below the country average standard. On the other hand, by focusing on laying and connecting with fibre and upgrading cable networks the definition of NGA excludes any mobile or wireless technology under the justification 2 that at this stage of technological and market development, neither satellite nor mobile network technologies appear to be capable of providing very high speed symmetrical broadband services although in the future the situation may change especially with regard to mobile services. The acknowledgement of the possibility of change merely recognizes what is currently a fact as regards technical evolution and is in line with the approach taken by the Commission in respect of technological neutrality applied to traditional broadband networks as 1 A NGA network is further defined as involving: (i) laying fibre to existing street cabinets offering the prospects of downstream bandwidths of a minimum of 40 Mbps and 15 Mbps upstream (compared with today's downstream speeds of a maximum of 8 and 24 Mbps for ADSL and ADSL2+ access technologies, respectively); (ii) upgrading current cable networks to deliver speeds up to and beyond 50 Mbps against the previous maximum speed of 20 Mbps, using the new 'DOCSIS 3.0' cable modem standard, or (iii) connecting newly built homes and offices with fibre connections offering services up to 100 Mbps and beyond. 2 At footnote / 16

5 expressed at 45(d) of the Draft Guidelines, where explicit reference to wireless (Wi-Fi, WiMAX), satellite and mobile technologies is made. Given the extraordinarily dynamic character of information and communication technologies and in order to avoid the risk of early obsolescence, the definition of NGA should include mobile and wireless technologies and consider them relevant as from the moment when they are capable of delivering the same technical output (or one adjusted to the needs to be covered). 2. Technological neutrality Technological neutrality is a well-established principle in Community policy as regards the electronic communications sector and one that the Commission reinstates in the Draft Guidelines as a condition of admissibility for State aid measures granted to NGA, namely at 45(d) and 66. It is therefore worth noticing that the unequivocal preference expressed throughout the Guidelines in favour of optical fibre somehow seems to be at odds with technological neutrality. PT concurs with the Commission s view according to which fibre currently appears to be the most capable and promising technology (at least in the short term), this being the reason underlying the large-scale investment project in the deployment of optical fibre that PT has already undertaken in Portugal on its own initiative. However, such prominence should under no circumstance translate into either imposing on fibre obligations which do not fall upon other apparently less-favoured technologies (but which end up being effectively favoured) or extending the detail of fibre regulation to a point where, within fibre technologies, some solutions are favoured in detriment of others Discriminatory unbundling obligation At 74, third bullet, of the Draft Guidelines it is stated that fibre-based networks benefiting from State aid should support an architecture that can be unbundled such as to conduct to longterm sustainable competition. It should be noted that according to 48, upgraded cable networks also fall within the definition of NGA. However, no equivalent obligation is imposed on cable-based networks benefiting from State aid. 5 / 16

6 The proposed approach seems to strongly contradict the principle of technological neutrality in that the treatment afforded to cable-based networks is unquestionably more favourable. It might be the case that the exemption of cable-based networks from the obligation of supporting an architecture allowing for unbundling results from recognized technical impossibility. However, this technical constraint does not seem to prevent the Commission from admitting State aid granted to upgraded cable networks, even where unbundling is not feasible, and nowhere in the Draft Guidelines seems to exist an obligation for cable-based NGA operators benefiting from State aid to provide network architectures allowing for unbundling. The Commission seems to fail to recognize that fibre-based networks may be deployed according to different architectures and that not all of them allow for unbundling, as illustrated further below. Due to economic or technical constraints, a given architecture might indeed prove to be the most appropriate for a given region. Thus imposing in all circumstances on fiber-based operators a specific architecture allowing for unbundling might have the effect of either creating a disincentive for investment and/or actually preventing an operator from investing or creating more cumbersome conditions for the operation of the NGA when compared with a cable-based network, thereby unjustifiably distorting competition between technologies. PT therefore submits that, in order for the technology neutrality principle to be upheld as regards the admissibility of State aid granted to the development of NGA: either the where feasible requirement, as applied to the deployment of point-to-point architecture and multifibre solutions provided for at 74, third bullet, should take into consideration the assessment made by NGA fibre-based operators as regards economic, commercial and technical constraints preventing the deployment of such architecture; or no specific architecture, namely point-to-point architecture and multifibre solutions, should be imposed on fibre-based NGA and the choice of the architecture should be left to the operator, without the result of such choice having a decisive impact on the admissibility of possible State aid. 6 / 16

7 Any other solution, as now provided in the Draft Guidelines, will unjustifiably favour cable-based networks in detriment of fibre-based networks thereby unjustifiably distorting competition between technologies Discriminatory imposition of network architecture At 74, third bullet, the Draft Guidelines state that in order to benefit from admissible State aid, a fibre-based NGA should support point-to-point architecture and multifibre solutions, such as to allow for unbundling. Not only does the Commission seem to fail to recognize that other fiber-based architectures might be more appropriate for a given region, but furthermore reaches a level of technical detail that seems to be absent from the compatibility assessment to be made of non-fibre NGA. This approach imposes a double burden on fibre-based NGA that might hinder desirable investment in certain underserved regions and unjustifiably favours non-fibre operators. First, it should be recognized that other solutions besides point-to-point architecture and multifibre solutions are technically valid and eligible for State aid, even if they do not allow for full-fledged unbundling. As operators take the risk of investment, the decision to determine viability of business models, including network architecture, should remain within their sphere. Point-to-point is knowingly the most expensive of all fibre-based solutions. Limiting the admissibility of State aid to this architecture and excluding other more economical - but equally sound - fibre-based solutions, such as GPON (Gigabit Passive Optical Network), may be tantamount to simply excluding the viability of investment being made in some regions and to failing the stated objective of accelerating the deployment of NGA in rural or underserved areas. This outcome does not appear as acceptable from a policy point of view. A (G)PON configuration reduces the amount of fiber and central office equipment required, compared with point-to-point architectures, optimizing fibre consumption at the access point. It requires less room in ducts and allows for the use of smaller optical distribution frames (ODFs), GPON provides normalized hardware with an optimum price/user ratio. All these features lead to lesser requirements of room and lower consumption of energy. 7 / 16

8 PT therefore submits that GPON is clearly the most adequate technology from a technical and economic point of view on which to base the deployment of NGA. Due to their multiple requirements, point-to-point solutions should be adopted only in very specific cases. Secondly, if a detailed technical solution is imposed on a given technology, such as fibre, in view of a policy objective, such as unbundling, there seems to be no valid reason not to impose an equally detailed technical solution on non-fibre technologies in view of the same policy objective if the technological neutrality principle is to be upheld. At the end of the day, overcoming technical constraints is a function of the required investment. If such effort is being required from fibre-based operators, not requiring an equivalent effort from non-fibre based operators is tantamount to blatant negative discrimination of fibre. This does not seem to be acceptable, neither from a technology neutrality policy approach nor from a more general competition policy perspective. Moreover, it might lead to the paradoxical situation where the technology which the Commission clearly seems to favour (and rightly so given its potential) throughout the Draft Guidelines - i.e. fibre - ends up being the least deployed in rural and underserved areas due to the impossibility of recouping the required investment. This outcome would frontally contradict the stated objectives enshrined in the Draft Guidelines. In the light of the above, PT submits that in order for the technological neutrality principle to be upheld and for competition not to be distorted: either other fibre-based solutions besides point-to-point architecture and multifibre solutions, such as GPON, are considered to be eligible for State aid where the fibrebased operator judges them to be the most appropriate for a given region, in spite of full-fledged unbundling not being technically possible; or detailed technical solutions should be imposed in respect of all non-fibre NGA, in the same way point-to-point is imposed on fibre-based NGA, in order for State aid to be considered admissible; or the amount of admissible State aid granted to fibre-based operators which deploy NGA supporting point-to-point architecture and multifibre solutions should fully cover the 8 / 16

9 accrued cost of such solution, when compared with other equally sound solutions, as well as a premium meant to cover the risk implied by the corresponding investment Exclusion of mobile and wireless technologies As stated before 3, by focusing on laying and connecting with fibre and upgrading cable networks the definition of NGA provided for in the Draft Guidelines excludes any mobile or wireless technology. Nonetheless, it is currently acknowledged, including by the Commission itself, that broadband capacity of mobile and wireless solutions will soon increase in significant terms. In order to avoid the risk of early obsolescence, the definition of NGA should therefore include mobile and wireless technologies as from the moment when they are capable of delivering the same technical output (or one adjusted to the needs to be covered). By not doing so, and given the extraordinarily dynamic character of information and communication technologies, the Draft Guidelines seem to inexplicably sideline two major and promising technologies and thereby appear to undermine the very technological neutrality principle which the document allegedly upholds. 3. Wholesale access According to 74 of the Draft Guidelines, in exchange for receiving state support, the beneficiary is required to provide third parties with wholesale access A priori obligations Utmost care should be taken as regards the imposition of a general obligation of wholesale access as well as the definition of general conditions for wholesale access. In fact, the requirement of access to the network should result from a careful balance between the need of network development, on the one hand, and costs associated with the deployment of the network, on the other. As regards the latter, the deployment of a NGA based, for example, on GPON, generally implies lower costs than point-to-point architecture and multifibre solutions. The decision to use one or the other depends on a number of factors, which can only be 3 See above Section 1 on Definition of NGA. 9 / 16

10 properly assessed on a case-by-case basis. Factors such as geography, orography, demography, resulting costs and expectations are not compatible with inflexible a priori rules. Again, imposing a priori wholesale access obligations in all circumstances, without taking into consideration the specifics of concrete cases, may significantly increase the cost of NGA deployment and might simply prevent any investment at all from being made in rural or underserved areas even if State aid measures are contemplated Equivalent burden At 74, first bullet, of the Draft Guidelines it is stated that the access obligation imposed should also include the right to use ducts and/or street cabinets in order to allow third parties to have access to passive and not only active infrastructure. It is not clear whether this approach takes into due consideration that a NGA may be set up and submitted for the award of State aid not only by operators who have chosen to be at the top of the ladder of investment and have built a complete infrastructure but also by operators further down the ladder of investment who set up their NGA with recourse to components of the former through access to e.g. ducts, bitstream, dark fiber or resale. It s important to bear in mind that the latter will also be active players in the market and will display their own offers. If wholesale access is to be imposed on operators who have borne the investment risk underlying a complete infrastructure, no less should be imposed on operators which have chosen to stay at a lower level of the ladder of investment. Any different approach will penalize investment and will provide a strong disincentive for operators to further climb the ladder of investment, thus contradicting the stated objective of upholding effective competition. PT therefore submits that, if wholesale access is to be imposed in general terms, then all operators which apply for State aid should be subject to the access obligation in respect of their offers and their components, independently from their degree of autonomy in respect of the underlying infrastructure Duration Imposing wholesale access for at least 7 years is out of touch with the technological reality underlying the information and communications sector. Technology changes very fast and may 10 / 16

11 dramatically alter operational assumptions. Different solutions lead to mutations in the market place, investment cycles tend to shrink and competition is more and more based on alternative infrastructures. It comes as no surprise that the Draft Guidelines provide in their last paragraph for a review after 3 years. Had someone suggested to the Commission to review the Guidelines only after 7 years and the likely response would have pointed to the need to keep pace with the markets. For market players which invest billions of euros in the deployment of NGA the need to keep pace with markets is not a matter of policy, is a matter of survival. 7 years is therefore an excessively long period of time. Imposing a priori wholesale access for 7 years in all circumstances, without taking into consideration the specifics of concrete cases, may seriously undermine the operation of NGA and may actually prevent any investment at all from being carried out in rural or underserved areas, no matter whether State aid measures are contemplated or not. PT therefore submits that, in order to keep pace with the fast development of technology and consequent changes in markets: either no minimal period of time at all is imposed as a condition for benefiting from State aid and it is left for the Member State in question to determine the duration of such obligation in the light of the specifics of the case; or the minimum period of time provided for should not exceed the one envisaged by the Commission to review the Guidelines, i.e. 3 years Price Benchmarking, as suggested at 45 (g) of the Draft Guidelines, for the definition of wholesale access prices is certainly a useful tool. However, care should be taken when using wholesale prices prevailing in different areas in order to avoid unrealistic charges and crowd-out possible investment in rural and underserved regions. Costs in building an infrastructure vary immensely across different regions due to inter alia distance, orography and dispersion. Necessarily, prices charged for wholesale access shall very accordingly. 11 / 16

12 In order to ensure that benchmarking is used correctly it is not sufficient to blindly apply wholesale prices that prevail in more competitive areas. Any benchmarking to be used must be based on wholesale prices that prevail in comparable more competitive areas. PT fully supports the principle according to which benchmarking criteria should be clearly indicated in the tender documents. If such criteria are unrealistic or misguided, an operator should be in a position to choose not to commit to a significant investment. 4. Timeframe for definition of areas At 63 and 69 the Draft Guidelines indicate 5 years as the reference for the definition of white and grey areas on the basis of the likelihood of a NGA being deployed. Given the extraordinarily dynamic character of communication and information industries and the pace of technological evolution, predefining a rigid timeframe as a crucial criterion for the definition of different degrees of assessment in respect of the admissibility of State aid measures is risky and might prove inaccurate. PT submits that: either a certain degree of flexibility is left for the Commission to use the 5 year criterion, according to the evolution of the market place; or if the Commission considers that a precise timeframe should be defined, the same timeframe indicated by the Commission to review the Guidelines should be used, i.e. 3 years. 5. Last mile At 74, last bullet, of the Draft Guidelines is stated that the aid should not cover the last mile access segment since this may also enable existing broadband operators to benefit from such infrastructure. This approach is not comprehensible from a technical point of view and seems to be inconsistent with the strong support of fibre-based point-to-point architecture shown by the Commission. 12 / 16

13 From a technical point of view, there seems to be no valid reason to exclude the last mile from the benefit of State aid, as the last mile is the relevant element allowing customers to have access to upgraded services. On the other hand, the Commission reveals an explicit inclination towards fibre-based pointto-point architecture at 74, third bullet, of the Draft Guidelines. It should be recalled that point-to-point architecture implies by definition the last mile. Thus excluding the last mile from the benefit of State aid is inconsistent with the support of point-to-point architecture. 6. Claw-back mechanism At 45 and 66 of the Draft Guidelines it is stated that Member States should include a reverse payment mechanism into a contract with the successful bidder in order to avoid possible overcompensation. This approach draws from the framework used for traditional broadband networks. However, it is worth noticing that whilst for traditional broadband networks a consolidated market experience is in place, no such knowledge exists for NGA. Triple-play solutions and HDTV are expected to face strong demand but apart from them a whole array of other services exists for which retail demand is currently impossible to estimate. The same problem exists for the wholesale markets, the size of which is of yet impossible to establish. It should not be forgotten that the provision of wholesale access implies significant costs and that an operator might actually find itself without viable demand due to lack of interest by other operators. Thus if the risk of over-compensation is theoretically possible, the opposite risk of subcompensation is no less, except in one respect: it might dictate the close of business for an operator. PT therefore submits that where mechanisms to avoid over-compensation are provided for, equivalent mechanisms for sub-compensation should also be established, such as to allow a Member State to fully compensate an operator from an unexpectedly low demand. 13 / 16

14 7. Regulatory framework Last but not least, the importance of an adequate regulatory framework in the electronic communications sector, that does not undermine the effort of both the Commission and the Member States in the field of NGA, should not be forgotten. At 2 of the Draft Guidelines the Commission states that the broadband strategy is an important part of the Recovery Plan and that the aim of the latter is to boost EU investment in defined strategic sectors, such as broadband, that can help support the economy in the short run and over the longer term create essential infrastructures for sustainable economic growth. With such purpose in view, Member States have earmarked millions of euros in State aid. Bearing in mind that taxpayers money is being used to promote entry into the new markets afforded by NGA, regulation of the electronic communications sector should not erect itself as an entry barrier in those very markets. If this happens to be the case, then State aid granted by Member States will not be more than a costly incarnation of the Myth of Sisyphus. 14 / 16

15 Conclusions 1. PT is supportive of the initiative undertaken by the Commission and shares the Commission s view as to the benefits of increased legal certainty and transparency in the Community decision-making process in the field of State aid rules applicable to NGA. 2. Given that PT has started on its own initiative a massive investment in optical fibre throughout the Portuguese territory, its views and concerns may contribute to shape a State aid policy which helps overcoming the digital divide and is supportive of private investment leading to further deployment of broadband in rural and underserved areas. 3. In order to avoid the risk of early obsolescence, the definition of NGA should include as from now mobile and wireless technologies and activate them as from the moment when they are capable of delivering a technical output equivalent the one provided by fibre and cable (or one adjusted to the needs to be covered). In order not to miss critical components of NGA, it should also include investment in core networks, service platforms and customer hardware. 4. In order to uphold the technological neutrality principle, the prominence afforded to optical fibre should not translate into either imposing on fibre obligations which do not fall upon other apparently less-favoured technologies (but which end up being effectively favoured by being subject to less constraints) or extending the detail of fibre regulation to a point where, within fibre technologies, some solutions are singled out in detriment of others. 5. Such would be the case if discriminatory unbundling obligations and discriminatory detailed architectures were imposed on fibre-based NGA only. 6. GPON is the most adequate technology from a technical and economic point of view on which to base the deployment of NGA. Due to their multiple requirements, point-topoint solutions should be adopted only in very specific cases. 15 / 16

16 7. Imposing a priori wholesale access obligations in all circumstances, without taking into consideration the specifics of concrete cases, may significantly increase the cost of NGA deployment and might prevent investment from being made in rural or underserved areas even if State aid measures are contemplated. 8. If wholesale access is to be imposed in general terms, then all operators which apply for State aid should be subject to the access obligation in respect of their offers and their components, independently from their degree of autonomy in respect of the underlying infrastructure. 9. Imposing wholesale access for at least 7 years is out of touch with the technological dynamics underlying the information and communications sector. Either no minimal period of time at all is imposed as a condition for benefiting from State aid and it is left for the Member State in question to determine the duration of such obligation in the light of the specifics of the case, or the minimum period of time provided for should not exceed 3 years. 10. Excluding the last mile from the benefit of State aid finds no technically valid reason and is inconsistent with the support expressed by the Commission of point-to-point architecture. 11. Where mechanisms to avoid over-compensation are provided for, equivalent mechanisms for sub-compensation should also be established, such as to allow a Member State to fully compensate an operator from unexpectedly low demand. 12. Since taxpayers money is being used under the form of State aid to promote entry in the new markets afforded by NGA, regulation of the electronic communications sector should not erect itself as an entry barrier in those very markets. If this happens to be the case, then State aid granted by Member States will not be more than a costly incarnation of the Myth of Sisyphus. 16 / 16