It can take a while depending on the size of the document..please wait
Draft BEREC Common Position on Mobile Infrastructure Sharing
0 days left (ends 18 Jan)
Public Consultation for the Common Position on Mobile Infrastructure Sharing
In June 2018, BEREC adopted a report on infrastructure sharing which depicts a picture of different regulation and legal frameworks applicable in European countries, of the existing sharing arrangements, of the benefits and challenges related to such arrangements and of possible evolutions with regards to 5G.
BEREC expended its work on identifying common positions on the subject. The common position consists of:
1) a background section which describes the applicable legal framework and the benefits and drawbacks related to sharing agreements ;
2) a common position section which provides:
a) common definitions for some types of sharing agreement;
b) the main objectives to be pursued when considering network sharing agreements;
c) the parameters to consider when assessing network sharing agreements;
3) an indicative analysis of different types of network sharing, according to the objectives and parameters.
The response to the public consultation will serve as inputs for the finalization of a BEREC common position on mobile infrastructure sharing.
This public consultation will run from 12 December 2018 to 18 January 2019, 17:00 CET.
Enquiries about the consultation, including registration problems with the online platform, should be sent to the following email address: Mobileinfrastructure-Consultation@berec.europa.eu
LEVEL OF AGREEMENT
MOST DISCUSSED PARAGRAPHS
MOST ACTIVE USERS
2.2.2 Potential drawbacks of infrastructure sharing
Potential drawbacks associated with infrastructure sharing were identified by NRAs in a smaller number of markets compared with potential benefits. This does not necessarily demonstrate that drawbacks associated with infrastructure sharing are less common compared with benefits; rather, it demonstrates simply that conclusive evidence demonstrating the negative impacts of infrastructure sharing agreements is less well developed (partly because of difficulties properly defining the counterfactual – i.e. whether the lack of an infrastructure sharing agreement would result in no network deployment whatsoever, in deployment of two or more competing networks by participants, or even a mixture of the two in different geographic areas).
The 2011 report and the June 2018 report both described in simple terms the common drawbacks associated with infrastructure sharing agreements that were identified by NRAs. This position paper elaborates this analysis by grouping potential drawbacks into three different categories relating to competition, coordination and network resilience.
Drawback 1: Reduced incentives to invest/ability to compete
Sharing agreements can negatively impact incentives for participants to invest in their own infrastructure, as any gains in service offering (relating, for example, to coverage, network quality etc.) resulting from a new investment are likely to be shared with other parties involved. The degree to which other parties benefit from this will necessarily depend on the sharing agreement type, but will be particularly pronounced for investments in new technologies and network upgrades.
At the same time, network operators participating in sharing agreement are likely to have fewer opportunities to differentiate their service offerings from those made by competitors also engaged in the related agreement. This is because network architecture will exactly or closely match that of other participating parties, meaning that factors such as coverage and quality of service are likely to be closely aligned. This potential drawback is likely to be particularly pronounced in active sharing agreements, as these further limit the ability of service providers to differentiate their services compared with passive sharing agreements (where providers still retain substantially more control over service delivery).
This reduction in incentives and ability to compete for those parties involved in a network sharing agreement means that end user choice – both between different infrastructure providers and between different service offerings – might be reduced. The range of problems relating to investment incentives/ability to compete were identified by both the 2011 and June 2018 reports, and remain a concern in some European markets currently. The degree to which these concerns impact dynamics in a given market will depend on context. It is possible, for example, for an infrastructure sharing agreement to provide a greater incentive for investment in a network, as they can reduce the costs to operators of offering coverage in a wide area (compared with when an operator is required to rollout an entire network on its own).
Drawback 2: Requirement for increased coordination between participants
Sharing agreements will necessarily require greater coordination between participants which will need to share at least some information to collaborate on network deployment. This presents an obvious risk relating to tacit collusion as well as potential breaches of competition law (dealt with in section 2.1 of this report) which must be addressed by participating parties.
More broadly, sharing agreements might lead to delays in deployment, as joint-decision making processes can add a layer of bureaucracy to the already complex process and potentially reduced incentives of network deployment. Extensive planning coordination can lead to delays at both the ‘strategic’ level (relating to network design and network evolution) and the ‘operational level’ (relating to the actual deployment of the network). Loading a host network site with active equipment from different network operators (which is typical of some passive sharing agreements), for example, might load a site in such a way that installation of new equipment modules related to the introduction of new technologies might be more difficult. The extent and impact of this drawback type is likely to depend on sharing type and network design.
Linked to this, consolidating two previously distinct networks following an infrastructure sharing agreement is likely to involve costs to network operators (relating, for example, to site removal, dismantling of superfluous equipment etc.) Often, this will be offset by efficiency gains resulting from the new infrastructure sharing agreement, but it does present a potential risk to parties with their own networks entering into such agreements.
Drawback 3: Reduced network resilience due to increased demand on host networks/sites
Shared infrastructure might reduce the overall resilience of mobile networks in a given geographic location. This is because fewer independent mobile networks will reduce the ability for end users to switch to alternative network operators when their own host network is unavailable (for example, when needing to contact emergency services). Similarly, network problems (e.g. RAN SW errors) can have a higher impact (affecting a greater number of end users over a wider areas) in situations where the RAN is shared. Again, this drawback needs to be balanced against the risk of no network whatsoever being deployed in the absence of a network sharing agreement.
3 Common Position
3.1 Common position (CP1) on the typology of infrastructure sharing types
The term ‘mobile (or wireless) network infrastructure sharing’ is used in Europe to denote different types of arrangement whereby two or more operators share some network or infrastructure elements to deliver services. Currently, there are no established criteria used to denote different types of infrastructure sharing (although a number of stakeholders, including the 3GPP, the GSMA and the OECD have provided a description of the different sharing types which can be seen on the market).
For the purposes of promoting common understanding between NRAs on sharing, BEREC is of the common position that NRAs should use the following definitions for inter-European discussions.
These definitions are primarily retrospective and do not preclude the emergence of any new types of sharing agreements not described here which might emerge in future.
3.1.1 Passive sharing
Passive sharing is the common use by two or more operators of passive elements of their respective networks. Passive elements are those which are not able to process or convert telecommunication signals in any way and which are not integrated parts of the system dedicated specifically to the conveyance of signals. Passive elements are sometimes referred to as ‘unpowered components’ as these elements usually do not require a power supply. This is however not always the case. For instance, air conditioning for cooling equipment might be considered a passive element, but usually requires an external power supply. Passive sharing can encompass the sharing of passive backhaul elements.
Co-location is a form of passive sharing where the operators share the same location (such as compound, base station sites, rooftops, etc.) for the construction of the base stations. It could be limited to a common access to the location. It could also include the use of common masts and other mounting/supporting constructions or cabinets including related installations (such as air conditioning, power supply etc.).
Site sharing is a form of co-location where two or more operators agree to deploy their masts or other supporting constructions in the same location. Typically, each operator provides own mast, backhaul, cabinets and active equipment.
Mast sharing is a form of co-location where two or more operators agree to use the same mast or other supporting construction. Generally, each operator provides own backhaul, cabinets and active equipment.
3.1.2 Active sharing
Active sharing is the common use by two or more operators of active elements of their respective networks. Active elements are those which are able to generate, process, amplify and control signals. Examples of active elements are very diverse and include many different types of electronic equipment (hardware and software) capable of various functions (transmitters, receivers, amplifiers, decoders etc.). While antennas have been traditionally classified as passive elements, technology advance has led to a paradigm shift to active antenna systems (AAS), which are considered a key enabler for 5G networks. Such antennas (or antenna arrays) can also be considered as active when equipped with radio frequency units such as amplifiers and signal processing elements. Furthermore, with transition to 5G, mobile networks would become more granular and virtualised so that operators are able to share complementary (hardware or software) components to build common network slices tailored to specific services.
RAN sharing is a form of active sharing where two or more operators agree to use the same access network equipment, including base station active equipment and possibly the antenna. Each operator uses its own core network. This type of active sharing itself can typically be split into two types.
Multi-Operator Radio Access Network (MORAN) sharing is a form of RAN sharing where only equipment is shared (i.e. not spectrum). The end-users of each operator access the services of their respective MNO with the frequencies of their respective MNO.
Multi Operator Core Network (MOCN) sharing is a form of RAN sharing where all elements of the radio access network, including potentially spectrum, are shared. The end-users of each operator can access the services of their respective MNO with the frequencies of one or several operators that are part of the sharing.
Frequency (or spectrum) sharing is a form of MOCN sharing where the frequencies of several operators are used. It consists in a common exploitation of frequencies among several operators: the end users of these operators can access the services of their respective MNO through all the frequencies that are shared in the access network.
National/local roaming is a form of active sharing where one operator uses the mobile service of another operator within the same country for the purpose of providing services to its end users.
3.1.3 Other sharing types
Core Network sharing is a form of sharing where operators agree to share elements of their core network, either on a standalone basis or in addition to sharing elements of their access network(s). Core network sharing can be limited to data transmission ring which connects the core network components and can extend to components themselves (such as switching centre with HLR, billing platforms and value-added services (VAS)).
Backhaul sharing is a form of sharing where one or more operators share backhaul elements. It is a form of passive sharing when the shared elements are passive, for example ducts and poles. It is a form of active sharing when it is the common use of network components for data transmission.
3.2 Important objectives and factors to consider when assessing mobile network infrastructure sharing agreement
Depending on the country, NRAs may have to assess sharing agreements between operators. Such situations may occur when:
1. the competent authority is preparing spectrum awards. In particular, when granting rights of use of spectrum, competent authorities/NRAs may include rules in the conditions attached to the rights of use. Those rules may in particular either aim at securing the effective and efficient use of spectrum, ensure the local provision of services or specifically address competition issues. This could correspond to the case where a competent authority assesses the possibility to introduce transitory roaming obligation in order to host a new entrant operator. This could also be to introduce sharing obligations associated with ambitious coverage objectives. This list is not exhaustive. Competent authorities will clearly establish such conditions attached when granting rights of use of spectrum and inform and consult interested parties;
2. An entity (e.g. competition authority, ministry, operators, other) asks the NRA to do so;
3. An NRA has to settle a dispute;
4. The national law provides the NRA with the power to assess sharing agreement.
For such situations, BEREC is of the common position that NRAs should, having regard to the particular circumstances of each case, support future network sharing agreements that comply with the objectives of NRAs while remaining in any case subject to competition law. It is of the common position that the below objectives and factors are likely to be relevant to NRA consideration of infrastructure sharing agreements.
3.2.1 Common position (CP2) on the main objectives to be pursued when considering network sharing agreements
When NRAs are to assess sharing agreements, BEREC is of the common position that the NRAs should aim at achieving/maintaining at least the following regulatory objectives as provided in the legal framework:
1. Effective competition: this general objective could be achieved in different ways according to the situation.
a) Infrastructure-based competition: the European electronic communications code explicitly mentions infrastructure-based competition as an objective. Infrastructure-based competition ensures the independent ability and incentive to invest for individual operators. It involves stand-alone network roll-out but may also involve passive infrastructure sharing. The prospect of protecting or increasing profitable sales by offering customers greater benefit (e.g. by improved coverage) or a better-value offer due to efficiency enhancements drives investments and constitutes a central incentive for the build-up and expansion of mobile communications infrastructure. The larger an operator’s market share is, the smaller is the incentive to make such investments in order to win additional customers from other competitors. And the lower the competitive pressure of others, the smaller is the incentive to retain the existing customers through investments and better offers.
b) Service-based competition: This relates to relatively direct competition in defining a specific offer with regard to price, quality of services, bandwidth, data volume and similar parameters. In fact, infrastructure-based competition is not always feasible. For example, a new entrant operator will initially not be able to compete solely based on its own infrastructure. Therefore, an entrant might rely on a temporary national roaming agreement, allowing market entry in technical and economic conditions that permit effective competition with existing operators. Another example is a project bringing additional coverage that might not be economically profitable and thus feasible for an individual operator. Thus, infrastructure sharing and realizing economies of scale in a network sharing agreement might be indispensable to realize such a project at all.
Based on these considerations, NRAs should ensure effective competition between actors for the end user’s benefit, when assessing sharing agreements.
2. Better connectivity
a) service improvements in terms of coverage (digital land development) or quality of service (throughput, service continuity or other mission-critical performance parameters such as low latency and reliability, e.g. needed for connected and automated driving along highways)
b) allow the development of IoT, machine type communication, network slicing for the next generation networks, management of legacy technology or services with a long lifecycle (such as GSM-based machine-type communications, including access to e-Call for cars), etc.
c) reduction of cost of deployment for passive infrastructure of high speed electronic communications network (in line with the broadband cost reduction directive)
3. Efficient use of spectrum
Spectrum rights of use are a scarce and essential input into the concentrated market for the provision of mobile services. Access to spectrum rights of use may be a barrier to entry and also to network expansion by MNOs. Many NRAs are legally required to ensure the effective and efficient use of spectrum, and this would necessarily apply when assessing sharing agreements between operators.
The assessment of the positive impact on effective competition, better connectivity and efficient use of spectrum should generally be considered based on a ‘counterfactual analysis’, i.e. in comparison with a situation where rollout is made without the proposed sharing, in order to determine the incremental positive impact that could not be achieved without sharing. This counterfactual analysis may be different if the NRA has imposed an obligation of passive sharing or may identify that one or more of the sharing partners would not be able to roll out on a standalone basis.
3.2.2 Common position (CP3) on the parameters to consider when assessing network sharing agreements in order to achieve/maintain the above mentioned objectives
When assessing network sharing projects in order to achieve or maintain the above mentioned objectives, NRAs have to consider a set of parameters. BEREC does not define an exhaustive list of parameters relevant for the assessment of all existing network sharing agreements. However, BEREC is of the common position that, inter alia, the following parameters are to be considered when assessing network sharing agreements in order to achieve/maintain the above mentioned objectives:
1. Competitive market forces evolution (including for the non-sharing parties and the market in general):
Sharing agreement is likely to have an impact on competition. This potential impact on competition requires careful examination. Several sub-parameters (market shares, number of operators involved in the sharing, the technologies involved, the geographic scope, the time frame…) are relevant for the assessment of the impact on competition. NRAs should at least refer to the below parameters when assessing the competitive market forces evolution:
a) Market shares/competitive forces: Sharing agreements can change the market structure, especially if one of the sharing parties is already benefiting from a strong position in the market. A sharing agreement should not lead to a situation where any or all sharing party/parties is/are in a position to impose its/their commercial strategy on the market, independently of its competitors, and ultimately the market. Effective competition between MNOs needs to be ensured. The NRA should thus consider whether a proposed agreement negatively affects rivalry between the parties to the agreement and thus competition and/or creates entry or expansion barriers for other competitors on the market;
b) Number of operators involved in the sharing: the above point implies that the number of operators involved in the sharing is an important parameter. For example, a sharing agreement of three operators out of four in the market should be cautiously assessed, because a large part of the market could jointly decide on infrastructure decisions, without an external force that is sufficiently strong to challenge/disrupt and to competitively constrain the operators on the shared infrastucture;
c) The technologies involved should also be considered (3G, 4G…), and whether all or only some of the technologies are included. The more a sharing agreement involves competitive technologies (4G, 5G…) that still require substantial investment, the more this sharing is critical;
d) The geographic scope: BEREC is of the common position that the concerned area is likely to have a great impact on competition (detailed below as a specific point);
e) The time frame: The impact on competition should be assessed in different time frames (i.e. short term and long term). For example, whether it is a temporary sharing agreement that aims at assisting entry of a new operator into the market, or a permanent sharing between the involved actors.
2. Feasible level of competition (in particular related to geographic and demographic factors):
Whether and to what extent infrastructure-based competition is feasible depends in many countries on geographical circumstances and thus on the areas concerned. Consequently, with respect to the geographic scope, the specificities of individual member states have to be taken into account. For example, infrastructure based competition might be feasible for the total area of some countries since the whole area is densely or moderately populated.
a) Areas, where full infrastructure-based competition is reasonably feasible (in particular, densely-populated areas): Deployments of own infrastructure tend to be profitable for operators in these areas since large part of the demand can be addressed with a deployment over a limited area (i.e. the economies associated with population densities are high). In this context, the promotion of infrastructure-based competition is strongly relevant because it encourages investment, innovation, and in the end effective competition for the benefit of customers;
b) Areas, where the feasibility of infrastructure-based competition is not pre-determined, requires a case-specific assessment (in particular, moderately populated areas): In such areas, a level of infrastructure sharing could bring benefits with regard to the regulatory objectives, but a lack of infrastructure-based competition leads to reduced incentives to invest and less innovation, and reduces the autonomy of actors, hampering effective competition between operators, both in the retail and wholesale markets;
c) Areas where infrastructure-based competition is not reasonably feasible (in particular, least densely-populated areas): These areas are the most expensive areas in which to build and expand networks. Sharing enables deployment cost reduction in such a way that it is indispensable to bring a minimum level of service quality to customers. In these areas, such a minimum level of service quality is infeasible with stand-alone deployments and thus infrastructure-based competition. With regard to coverage objectives, network sharing might be of particular relevance in isolated territories, where a careful consideration should be given to the sharing conditions in order to enable inclusion of other operators.
However, within all areas, further consideration should be given to non-replicable sites or deployments. In that case, operators are confronted with such a scarcity of available sites or limited space or other essential inputs (such as power connection availability, power monitoring possibility, backhaul links, etc.) such that they cannot individually deploy their parallel networks in order to supply the demand. Examples might be indoor deployments or significant deployments of small cells in specific situations. In these situations, infrastructure sharing could be objectively necessary for competition among MNOs, and competent authorities might – in those specific cases – even mandate sharing.
3. Type of sharing:
Different types of infrastructure sharing have a different impact. In general, solely passive sharing is the most preferred solution. Active sharing is seen as already substantially reducing infrastructure competition and virtual access (or national/local roaming in the context of mobile networks) is the least preferred form of infrastructure sharing.
4. Shared information between the sharing parties and its impact on their ability to compete:
Shared information between operators should generally be strictly limited to the level that is indispensable for the agreement itself and restricted to the persons necessary to the proper functioning of the shared network. The information exchange should not limit the sharing parties’ ability and incentive to compete and invest. For example, if joint partners are immediately informed of independent upgrade plans, one MNO's investment could immediately be matched by another MNO's investment. This reduces the incentive to invest and could potentially stop such upgrade plans, as there is not even a temporary competitive advantage and thus no customer gain from the partner MNO.
4. Reversibility and contractual implementation:
Sharing agreements are intrinsically inflexible, since they typically involve a commitment from the sharing parties in order to produce the expected benefits from sharing. But this rigidity should be kept to the indispensable level for the overall sharing agreement. The reversibility of the agreement as well as the structure of the co-operation between the parties (a separate infrastructure company jointly owned, exchange of assets between the parties, wholesale arrangements, etc.) have strong implications on the rigidity of the agreement. Sharing agreements operate under the framework provided by contract law.
4. Indicative analysis of different types of network sharing, according to the above mentioned objectives and parameters
This section describes the assessment of some types of network sharing (cf. the typology of network sharing established in subsection 3.1) based on the above parameters. It does not prejudice future analysis to be conducted by NRA on future network sharing agreements.