Contribution of. Deutsche Telekom AG. Friedrich-Ebert-Allee 140 D-53113, Bonn, Germany Register ID To the

Transcription

1 Contribution of Deutsche Telekom AG Friedrich-Ebert-Allee 140 D-53113, Bonn, Germany Register ID To the EU Consultation on the Revision of the Community Guidelines for the Application of State Aid Rules in Relation to the Rapid Deployment of Broadband Networks 31 August 2012 Page 1

2 Deutsche Telekom AG Bonn, Contribution to the European Commission Consultation on the EU Guidelines for the Application of State Aid Rules in Relation to the Rapid Deployment of Broadband Networks Deutsche Telekom welcomes the opportunity to comment on the Commission s draft of the revised EU Guidelines for the application of State aid rules in relation to the rapid deployment of broadband networks. General Remarks Deutsche Telekom shares the Commission s goal for the revision of these Guidelines to ensure that State aid measures will result in higher level or faster rate of broadband coverage and penetration. In this context, the Commission makes reference to the Europe 2020 Strategy and one of its flagship initiatives, the Digital Agenda for Europe (DAE). Deutsche Telekom believes that the ambitious next generation access (NGA) targets of the DAE can still be met provided that framework conditions are set in the right way. With her public announcement on July 12, Vice President and Commissioner Neelie Kroes made a bold step in the right direction, intending to support a higher level of investment in high-speed broadband infrastructures. The new policy direction for our sector is very much appreciated by European network operators and the investors community alike. Core elements of this new policy approach are inter alia non-discrimination through equivalent access as the most important guarantee of sustainable competition, lightened regulatory intervention to provide for more flexibility for incumbent operators and others to explore new business models and technology neutrality, i.e. not picking winners in terms of best technological solutions for NGA network rollout. Several proposals for revision of theses Guidelines are not effectively supporting the DAE targets and are not compatible with the new policy approach announced by Commissioner Kroes. These are in particular proposals to give preference to point-to-point fibre to the home (FTTH) networks, proposals to discriminate against investments by operators with significant market power (SMP) and proposals to increase the regulatory burdens on SMP operators where they benefit from State aid. Page 2

3 Guidelines should strictly respect the principle of technology neutrality As a matter of fact, the profitability of NGA investments varies significantly depending on the density of population in different geographic areas. Accordingly, a substantial part of NGA network rollout will depend on State aid. As financial resources are limited, public money should be used as efficiently as possible to maximize the effect of scarce public resources in achieving the DAE targets. Several proposals in the consultation document are clearly biased in favour of point-to-point FTTH access networks, by far the most expensive NGA network architecture, thereby discriminating against much more cost and time efficient technologies like fibre to the curb (FTTC) Vectoring, equally able to deliver on the DAE targets. With reference to the Commission Communication on State Aid Modernisation (SAM) of May 2012, the consultation document itself emphasises that State aid policy should focus on facilitating well-designed aid targeted at market failures and objectives of common European interest. Hence, aid measures should focus on helping Europe reach its policy goals. Favouring technologies and topologies that are costly would only hinder Europe from reaching the DAE targets, thereby negatively impacting EU competitiveness, growth and social development. While the consultation paper acknowledges that the gradual upgrade of existing infrastructures (for instance extending fibre connectivity to the nodes) may still greatly contribute to achieving of the speeds objectives of the DAE, it then differentiates between interim and future proof NGA networks. In the consultation paper s view, interim infrastructures are more adequate for disadvantaged areas and future proof networks for urban areas. The latter networks are, according to the paper under review, point-to-point FTTH networks providing symmetric speeds and supporting infrastructure-based competition via full unbundling (paragraphs 54 and footnote 88) 1. Preference for these so called future proof (point-to-point FTTH) networks can be observed throughout the document, and it strongly impacts the eligibility criteria. For example, in areas that are white in terms of both broadband and NGA, tender rules may award more points to networks supporting point-to-point topologies with full unbundling capabilities (paragraph 67d and footnote 88). In white or grey NGA areas that are grey or black in terms of basic broadband, point-to-point topologies supporting full unbundling even becomes an eligibility requirement (paragraph 76b and c and footnotes 103, 108). And in well served grey and black 1 References made in brackets in the text are referring to the consultation document. Page 3

4 NGA areas, the draft document allows for the usage of public money to finance so called future proof networks that would crowd existing networks which are not point-to-point FTTH networks providing symmetric speeds and supporting full unbundling. All in all, these proposals will not only result in a misallocation of public money to the detriment of the DAE targets; they also create a high risk of crowding out and set strong disincentives to invest in networks that are not considered future proof by the text under review. The consultation paper justifies the discrimination of other, more cost efficient network topologies inter alia by assuming that at this stage of market development, a point to point topology is more conducive for long term competition in comparison with point to multipoint topology, while the deployment costs are comparable especially in urban areas (footnote 88). This assessment is inaccurate as regards the investment costs as well as the operative costs. Studies show that e.g. for Germany, FTTC costs range from around 100-2,000 per homes passed while FTTH costs range from around 600 to 4,500 per homes passed. These estimates assume a 50% take-up and do not include in-house wiring and customer premises equipment. 2 Point-to-point FTTH does not only imply significant higher upfront investment in urban as well as in rural areas, but also higher operative costs. These high operative costs are driven by the higher number of active network elements, also resulting in much higher energy consumption involved in point-to-point FTTH. This, in turn, is counterproductive to the fulfilment of the energy efficiency goals and the EU 2020 emission reduction targets. Furthermore, the consultation paper wrongly suggests that symmetry or asymmetry of speeds is determined by the network topology, such as point-to-point FTTH vs. GPON 3 as well as FTTH, FTTB 4 or FTTC. The upload/download split is determined by the active network equipment, i.e. routers and modems, not the network topology. Even though demand for upload will increase over time, we expect demand for broadband to continue to be asymmetric. Hence, a symmetrical split would reserve excessive capacities for upload, which would remain unused while missing for higher demand in download capacities. Symmetry would entail, for the time being and the average user, an inefficient resource allocation. As demand for further upload capacity increases, operators will take due account of this development and adjust the split to provide more upload capacity in NGA networks. 2 WIK, 2010: Broadband deployment and its limits. 3 Gigabit passive optical networks. 4 Fibre to the building. Page 4

5 Specific access obligations should not be a condition for State aid where they would indirectly result in a violation of the principle of technology neutrality. For example, the full unbundling requirement introduced by the Commission (paragraph 76b and footnote 103), would result in the preclusion to offer higher broadband speeds on the basis of FTTC Vectoring to the detriment of consumers. Furthermore, any access obligation imposed should not exclude the owner of an infrastructure from its usage as such an approach goes against the core principles of ownership law and would create disincentives to invest. In sum, the rationale put forward by the consultation paper is extremely concerning as it picks winners without knowing how markets and technologies will develop. And this is exactly what the well established principle of technology neutrality, for good reasons, intends to preclude. 5 Infringements to the principle of technology neutrality (as regards both network design and access obligations) would result in inefficiencies, thereby hindering the achievement of the DAE targets. Guidelines should not discriminate against SMP operators Further inefficiencies arise when the beneficiary aid is chosen in a discriminatory manner. Deutsche Telekom has serious concerns regarding the potential for discrimination in tenders against firms that hold SMP in broadly defined product and geographical markets. The main purpose of tender procedures is to ensure that the minimal amount of aid is granted in a non discriminatory manner. Tenders should be designed in order to maximize efficiency considerations, including those based on economies of scale, and keep administrative costs at a minimum. Award criteria should further ensure that the beneficiary is chosen fairly and that no taxpayers money is wasted. The consultation paper s suggestion that tenders be designed in order to distribute State funds among a large number of operators (paragraph 67a and footnote 79), would go against the very purpose of the procedure, being both discriminatory and inefficient. Guidelines should respect the principle of proportionality It would also be inefficient to burden the beneficiary with disproportionate, unnecessary obligations, not suitable to achieve the objective of the aid measures. Indeed, these 5 As stated by Commissioner Kroes on July 12, institutions should be wary of picking winners. Technology neutrality is just another way of saying that we cannot predict with any certainty what the best technological solutions will be, nor how they will compete and interact. Incremental solutions may Page 5

6 obligations would only increase costs and consequently the amount of aid needed. The text under review seems to be based on the rationale that access and non discrimination obligations imposed on aid beneficiaries with SMP should go beyond those already set under sector specific rules, just because the network has been partly financed with taxpayers money (paragraph 67g). The consultation document further proposes that access prices be based on the pricing principles set by the regulator and on benchmarks, and that process should take into account the aid received by the network operator (paragraph 67h). Deutsche Telekom shares the view that publicly funded access networks should provide open access for third parties. Access should be offered under terms and conditions that enable competition on the retail market. But there is no reason to go beyond the standards set by the regulator for open access to networks of SMP operators. Commissioner Kroes in her policy statement of July 2012 recalled that too much intervention constrains flexibility, which in turn reduces the range and quality of services that can be offered to different consumers, and that we should be aware of both direct and indirect effects of regulation. Consequently, regulation should be more focussed, allowing for a lightened overall regulatory burden. This is not less valid in the context of State aid as all regulatory obligations imposed on the beneficiary will be reflected in its business case. This rise in costs results in an increase of the profitability gap and therefore of the amount of aid needed. In sum, disproportionate obligations will inevitably result in less efficient aid measures and consequently in a loss of taxpayers money. In cases where aid is granted to operators not designated as having SMP, and therefore not being subject to open access requirements, equivalent open access needs to be ensured by the aid measures. Furthermore, symmetrical and equivalent access obligations for all access network operators would be in the interest of all competing service providers as it would reduce complexity and make the offer of retail services on the basis or third party wholesale easier to handle. Guidelines should provide for a more objective definition of white and grey areas The profitability of specific areas differs strongly based on geographical and demographical considerations. Due to lower costs per household, densely populated areas with a high rate of urbanization are typically more profitable than low density areas with scattered population. As costs and profitability are driven by these economic factors, these objective criteria should be the main drivers to classify areas. help to address weak demand in the short term for example, new technology combining fibre and Page 6

7 The Commission has chosen an approach that relies on short term operators commitments to invest in a given area. The rationale seems to be that if an area is profitable, operators will invest in it in the near future. While it is true that investment will take place in profitable areas, it is inaccurate that all network investments can and will be made within a rather short period, i.e. in the next few years. As a matter of fact, due to limited resources (e.g. planning and building capacity) no operator can start the roll out of NGA networks in all profitable areas simultaneously or within the proposed three years time period. Even the ambitious DAE targets only provide for the NGA goals to be achieved by 2020 and not within three years. We need to strike a balance between the speed of deployment and the cost efficient provision of NGA to the highest number of households, bridging of the digital divide. If aid was granted to deploy broadband in profitable areas only because undertakings were unable to start rolling out their networks within three years, public funds would have been used where they were not needed and will be missing elsewhere. Within this context, Deutsche Telekom supports the proposal of the consultation paper allowing for the time period considered to classify an area as white or grey to be extended beyond three years. Unfortunately, the proposal only intends to accept a longer time horizon if this is in line with national broadband strategies and the investment perspective of the granting authority for the development of the subsidised infrastructure (paragraph 59), not taking into account specific factors that may affect the investment decision of private operators. We would therefore go one step further than the current proposal and encourage the Commission to replace the three year period as a rule by a longer timeframe which takes into account the complexity of the totality of rollout projects planned by each undertaking in order to timely achieve the DAE targets. Guidelines should expressly mention the eligibility of FTTC Vectoring for State aid The consultation paper suggests not to grant State aid to operators subject to certain universal services or licence obligations (paragraph 44 and footnote 57) or when the investment in new infrastructures is regarded as small (paragraph 48). Deutsche Telekom agrees that operators with specific coverage targets under their LTE licence conditions or designated with a universal service obligation should not be granted additional aid to fulfil the specific obligation in question. However, this should not exclude the operator from becoming eligible for State aid when making investments that go beyond the scope of its obligations. In the same vein, the copper, or upgrading TV cable, can be very cost-effective in delivering higher download capacity. Page 7

8 incentive effect can be considered low when investments are marginal, as it is the case of an upgrade from FTTC to FTTC Vectoring (footnote 58). To the contrary, an upgrade from copper to FTTC or FTTC Vectoring is investment intensive and should therefore be eligible for State aid, in particular as this is for a large part of the EU the only realistic option to achieve the DAE targets. It is therefore crucial that the eligibility of FTTC Vectoring be expressly clarified in the Guidelines. State aid should not per se be limited to passive infrastructures in areas that are grey or black in terms of basic broadband The consultation document proposes that in white or grey NGA areas that are grey or black in terms of basic broadband, aid should be limited to passive infrastructures only (paragraph 76a). Deutsche Telekom believes that this limitation would render many projects economically unfeasible, as costs associated with the active components of the network alone can result in a profitability gap, independently from the costs of the passive elements. Only where this is not the case should aid be limited to passive infrastructure. Safeguards against discriminatory behaviour should be put in place if the beneficiary is not already subject to SMP regulation. Where a non discrimination obligation has already been imposed under sector specific rules, additional safeguards are unnecessary and therefore disproportionate. In any event, the broadly accepted term non discrimination should be preferred to broad and unclear language such as any conflict of interest or hidden indirect advantages. Requirements for commitments to invest should be more pragmatic Deutsche Telekom shares the view expressed in the consultation document that a commitment to invest should be supported by some evidence that goes beyond a mere expression of interest to invest in a given area. At the same time, requirements set within this context should allow for the protection of business secrets and respect the applicable confidentiality rules. For instance, if the aid granting authority were to request the delivery of a business plan, confidentiality must be ensured. Furthermore, standards set to deem a commitment reliable should remain realistic. For example, bank loans are not normally requested years before the investment takes place, and they are often not associated with a concrete rollout project. On these grounds, the reference to bank loan agreements and detailed scheduling (footnote 71) should be deleted. Page 8

9 Finally, it should be noted that, currently, market surveys followed by undertakings commitments agreed to with the relevant authorities are proving to be sufficient. Therefore, there seems to be no need for both parties to enter into a contract which would only increase the administrative burden. Consequently, the reference to possible contracts between private operators and the granting authority and to inherent obligations (paragraph 60) should be removed. Conditions for the use of synergies should be enhanced Deutsche Telekom welcomes the approach to foster the use of synergies between existing infrastructures not only of the telecommunications sector but also of other industries. Centralized inventories of planned and existing infrastructures can become key enablers of synergy use, and Deutsche Telekom welcomes their introduction where they are not already in place. We would however like to point out that information on planned works should not be integrated into an inventory of existing infrastructure. Such information would, if combined with information on existing infrastructures, give competitors hints on the investors rollout strategy. Containing sensitive information from a competition point of view, inventories of planned and existing civil works should be kept separate. In order to benefit from synergies arising from planned civil works to be used, telecommunications operators need to receive information on these works at least one and preferably two years in advance. Shorter periods are unsuitable with regards to the complex internal and external planning processes required for NGA rollout. Moreover, the legal framework for infrastructure sharing and coordination of civil works should be designed in a technology neutral manner, encompassing more than just fibre connections (paragraph 29). Page 9