UKCTA response to Openreach on Exchange Closure: Key Considerations

1. UKCTA is a trade association promoting the interests of fixed-line telecommunications companies in the residential and business markets. We advocate regulatory outcomes designed to serve consumer interests, particularly through competition. Details of membership of UKCTA can be found at www.ukcta.org.uk.

2. UKCTA is grateful for the opportunity to comment on Openreach’s proposed plan for exchange closure.

3. We have split our comments into five areas:

• Section 1: The need for Transparency and Certainty;
• Section 2: Securing Good Product Outcomes;
• Section 3: Fair approach to Costs recovery;
• Section 4: Putting the Customer first;
• Section 5: Ensuring Competition Can Flourish.

4. Openreach needs to provide as much certainty and transparency as possible to industry in order to enable an effective migration and closure programme.

5. Openreach should provide as much forward planning information as possible, as early as possible e.g. an upfront plan of all affected exchanges so providers can check their own footprint against it for impacts. Information in this context is unlikely to be commercially confidential (unlike sensitivities over FTTP rollout plans), and therefore Openreach should be able to share more with industry in this programme.

6. The industry should be embedded into the closure planning process, not just the implementation. For example, there should be a mechanism for communications providers (“CPs”) to highlight critical exchanges for them in their own network footprint. Openreach should take account of CPs information into the planning process, beyond their own incentives for exchange closure. We would like to know how Openreach plans to balance these considerations in this programme.

7. CPs will need sufficient advance notification; this is definitely more than the contractual minimum requirement of 12 months. Selected exchanges should be locked in at a certain advance date.

8. We recommend a “point of no return” requirement for exchange closures, so as to ensure that CP investment and resource is not wasted.

9. Currently there is no intra exchange Dark Fibre Access at a number of exchanges where BT are deemed to not have SMP. With the reduced number of exchanges in the proposed footprint it would be anticipated that the level of connectivity into those exchanges from multiple parties would remove any SMP for BT. Would this then impact intra exchange DFA, in effect removing it as a product?

10. Exchange resilience – whilst resilience has been touched on, the question of exchange resilience needs to be raised in light of the significantly wider footprint served by each exchange. Resilient power supply from divergent sources has to be supplied along with multiple separation of duct access to the exchange. There is also the question of failover exchanges in the event of a critical failure such as fire. This is not a new concept, the original 21st Century Network design incorporated the model of automatic failover.

11. Enough availability of space at a suitable exchange that are going to remain is essential (including rack space and power). Openreach will require to carry out upgrades to keep pace with CP footprint growth. Openreach will also need to consider the congested footways around popular exchanges.

12. Sensible steps are required to prevent CPs doing an early “land grab” and sitting on floorspace / power budget for no reason other than to secure competitive advantage over their rivals.

13. Products with a greater distance budget need to evolve as the inter-exchange distances get longer. This is all the more important should a CP be unable to secure rack space in one of the remaining handover sites. We would welcome early clarity around how Openreach would approach this issue commercially.

14. Openreach needs to clarify its approach to the EAD LA variant and knock-on pricing effects of EAD should the local access variant be removed from the market. There is an incentive for Openreach to consider in mitigating its migration challenge and costs between now and 2030s. Where a CP places an EAD LA order to an Exchange slated for closure, Openreach could incentivise the CP to re-route the order to the nearest Parent Exchange (assuming it is not also slated for closure) on LA pricing.

15. Openreach have to put in place processes across the programme to ensure that the cost of all activities are efficient and minimised, with an emphasis on repetition and standardisation;

16. Openreach are to establish a cost recovery framework to allow all fair costs incurred by Communication Providers (Openreach Charges, CP’s own kit and labour costs and end customer management costs) to be recovered within the scope of the programme. Openreach require CPs to undertake significant amounts of work that they have not requested in order to unlock exchange closure savings. It is vital that CPs own costs are fully taken into account.

17. The principle that no CP should be left worse off by the timing of their migration (moving early or late in the program should not have cost implications);

18. The principle that CPs should not be worse off in new co-location set up (supporting similar service volumes) should be made clear from the outset. Our expectation is that costs will be lower in future given the substantial savings Openreach will be making and we would expect that these savings should be passed on to CPs in reduced charges

19. All costs incurred should be scrutinised to ensure environmental aspirations are met, with green technology choices and equipment configurations prioritised.

20. Openreach’s assertion that “We expect that most of the activities required to exit exchanges will not directly impact the end customer experience” (Section 19 of Consultation Document) is not a realistic one. This programme involves moving a vast number of circuits nationally across all customer and product types. There will also need to be a comprehensive communications strategy to inform customers of this activity. To assume that this will all happen seamlessly is at best optimistic and at worst, naïve and risky.

21. Particular attention must be paid throughout this programme to vulnerable customers, the continuity of their services and any essential equipment connected to their lines.

22. Likewise, the continuity of services serving Critical National Infrastructure should be of the highest importance when developing processes to accommodate exchange closures.

23. We could propose an approach of ringfencing circuits serving Vulnerable Customers/CNI (where viable) to follow a more closely managed migration process.

24. Notwithstanding the needs of the vulnerable and CNI, avoidance of any prolonged loss of service should remain a priority for all customers and a key principle throughout the exchange closure programme

25. We should emphasise the need for a robust and fair out-of-hours migrations capability, evenly divided across CPs of all sizes. This is particularly important for business customers who shouldn’t suffer an outage during trading hours, however brief.

26. The proposal to refresh the leased line portfolio as part of this programme should not have a detrimental impact on the choice of products available to customers who currently use these products.

27. Any contractual changes made to accommodate this programme should include an effective SLA/SLG mechanism to give customers the confidence that any failure caused by exchange closure activity will result in compensation and/or a penalty to Openreach.

28. The proposals have impacts on competition both in the downstream market and in the infrastructure market. Despite this, there has to date been limited dialogue through the Openreach consultations regarding the impact of exchange closure (and the copper switch-over more generally) on competition.

29. The proposals have impacts on competition both in the downstream market and in the infrastructure market. Despite this, there has to date been limited dialogue through the Openreach consultations regarding the impact of exchange closure (and the copper switch-over more generally) on competition.

30. At a time when many rival operators are deploying fibre and thereby seeking to compete directly with Openreach, concerns have been raised about the extent to which the exchange closure programme could provide an unfair advantage for Openreach in the infrastructure market and also could impact CPs’ plans to plan ahead and consume Openreach services going forward.

31. Openreach’s Exchanges are national infrastructure of the highest importance to the entire UK telecommunications market. Whilst the timeframe for closure from 2030 onwards falls outside of a current Wholesale Market regulation window, it is fair to note that we are only one regulatory period away from the cycle that will overlap with the beginning of the closure programme. As such, there is an expectation on Openreach as custodians of an irreplaceable Exchange network, which constitutes an enduring economic bottleneck to competition in the telecommunications market, that it anticipates the impact on competition and consumers that Exchange closure will have.

32. Openreach should act in good faith and not attempt to hide from clearly foreseeable regulatory expectations merely because the timeframe is slightly beyond the current regulatory horizon.

33. Exchange space and power, or “accommodation services”, is an essential feature of SMP regulation across markets where Openreach has a dominant position. This finding has been held consistently across wholesale markets and review periods in which Ofcom has recognised that without regulation, Openreach would have the opportunity to abuse its dominant position in the provision of accommodation services with material impacts across the range of downstream regulated markets.

34. This is a feature of the current and forthcoming regulatory frameworks. For instance, the current Business Connectivity Market Review (2019) framework states:

14.24 Accommodation in BT exchanges is an important enabler of competition in leased lines markets. It allows telecoms providers to make use of products such as EAD and EAD Local Access and facilitates competition in other markets.

14.25 Access to space and power in BT’s exchanges can be limited, and in the absence of regulation BT would have the incentive and ability to discriminate in favour of its own needs in allocating such space and providing power. We further note that BT could choose not to supply some or all of these services or charge excessive prices. As telecoms providers must purchase these services to use regulated products, this would have the same effect as refusal to supply, or excessive pricing for the main wholesale products. The absence of requirements in relation to accommodation services could thus undermine the effectiveness of other remedies in the relevant markets.

35. The forthcoming regulatory framework is almost certain to follow the long-standing principle of incorporation of space and power obligations within Openreach’s SMP conditions for regulated wholesale markets.

36. The soon to be finalised Wholesale Fixed Telecommunications Market Review 2021-2026 includes the following proposed approach in its draft (Volume 3, Non-pricing remedies):

3.46 The availability of accommodation services in BT exchanges is an important enabler of competition in the WLA, LL Access and IEC markets as well as the physical infrastructure market. It allows telecoms providers to make use of disaggregated products such as FTTP and EAD Local Access and facilitates competition in downstream markets. Space and power in BT’s exchanges are particularly limited, and in the absence of regulation BT would have the incentive and ability to discriminate in favour of its own needs in allocating such space and power.

37. As is noted in Para 3.46 of the draft WFTMR, Exchange space and power is also an essential feature of the physical infrastructure market. For some operators, deployment of fibre networks utilising the PIA framework is most efficient when planned to route via Openreach Exchanges that are already part of the operator’s existing aggregation network. This generates obvious efficiencies in relation to backhaul supply.

38. The 2019 PIMR set out Openreach’s obligations to provide ancillary services including power and Co-Location at the Exchange to make PIA usable and to remove the possibility of exercise of market power in the supply of those ancillary services. Again, the draft WFTMR (Volume 3: Non-pricing remedies) is consistent to these principles and it is all but certain that these requirements will be incorporated into the forthcoming five year regulatory cycle:

4.44 BT are required to provide PIA ancillary services as may be reasonably necessary for the use of PIA, including as a minimum: power, PIA Co-Location, PIA Co-Mingling (the provision of space and the ability to house equipment in a BT telephone exchange or equivalent), PIA Site Access (access to equipment that the telecoms provider has in a BT telephone exchange or equivalent) and PIA Database Access

4.45 We continue to believe that it is appropriate and proportionate to require BT to provide PIA ancillary services. A requirement to offer access to ancillary services has the purpose of assisting in promoting competition in downstream markets.

39. Complex interdependencies and impacts will doubtless arise as we go through the process of migration of existing PIA capacity in effected Exchanges to new Exchanges. Given the fact that the PIA proposition is still very much evolving and subject to iterative improvement, the closure programme introduces a new and very significant layer of complexity that must not be allowed to overwhelm a fragile (but improving) PIA. PIA complexity is hard to imagine in the abstract; a lot of the operational complexity and interdependency will present itself when we get into the thick of the closure process. What is therefore essential is that close working processes and forums are established between both programmes and that industry has the opportunity to raise complexities with equally informed personnel that are speaking the same language and working towards the same resolution.

40. Some of the foreseeable complexities are likely to be:

• Capacity crunches in meet-me chambers surrounding remaining Exchanges
• A concertina effect on network adjustments in ducts leading to and from remaining Exchanges
• A requirement for additional capacity in ducts leading to and from remaining Exchanges
• Provisioning of new cablelink circuits and resulting capacity crunches for Openreach splicing resource

41. In addition to those engineering challenges, there are economic impacts that need to be worked through as with the case for leased lines. Users of PIA will need to understand what Openreach’s approach to cost recovery will be where network deployments are left stranded as a result of Exchange closure. Additional dig distance between a rollout cluster and nearest extension will drive cost inflation for deployers using PIA; how will Openreach address this to ensure that it meets its requirements to provide accommodation services to support PIA in such as a way that does not impact downstream competition?

42. Whilst we can’t predict future regulatory approach to accommodation (space and power) it is fundamentally intertwined with wholesale and passives regulatory framework and we can reasonably anticipate that Ofcom will continue to view it as a potential constraint to downstream competition in future WFTMRs.

43. Identifiable risks for downstream competition resulting from constraints on space and power:

PIA:
▪ Uncertainty – including near-term uncertainty – raises the risk to Openreach that it will face challenge on whether it is meeting its accommodation requirements for PIA under WFTMR.
▪ Some but not all users of PIA design their rollout plans via Openreach Exchange footprint; various contexts in which this is the optimally efficient approach to rollout utilising passive infrastructure
▪ Moreover, in-flight PIA build, designed around the Openreach Exchange footprint, is potentially impacted by Exchange closure despite the long lead times. Backhaul into Exchanges and Access from Exchanges to Nodes is costly to re-route at a later date. Even the nearest Exchange may generate incremental costs that change the economic viability of a rollout area – and users have no guarantee that the nearest exchange will have capacity for PIA customers.
Ethernet:
▪ Long-term structure of EAD LA product uncertain as a result of the Exchange closure programme
▪ As highlighted above, we would welcome commercial clarity from Openreach (presumably) on its plans for geographic pricing on EAD and the future of local access pricing, in the context of many LA sites closing and circuits needing to be re-arranged as a result. In doing so Openreach should be mindful of competition concerns.
Ancillary charges:
▪ As we get closer to a cease to provide cut-off at a particular Exchange, there will likely be a window in which CPs can consume products for the full remaining time (e.g. 1 year term EADs or other such products) but in order to provision them, will face sunk costs that cannot be recovered over the remaining term of the Exchange (e.g. hefty Excess Construction Costs; Cablelink fees). How will Openreach ensure it is not over-recovering against those charges and how it will ensure that those costs don’t become prohibitive to downstream competition?

44. The ability (under Ofcom’s current proposals for the copper switch-over) for Openreach to be able to selectively switch off individual exchanges and thereby benefit from the relaxation of certain SMP obligations is a concern. The freedom provided by using a single exchange as the geographic unit would grant Openreach the ability to strategically target rivals’ build, in an effort to foreclose that rival deployment.

45. Given Openreach are planning to migrate from some ~5,500 existing exchanges to ~1,000 handover exchanges as the topological structure for its full fibre network, we consider that this would be a much more appropriate geographic unit and would also require Openreach to make a material deployment commitment across a large contiguous area (as opposed to being able to target full fibre deployment at a local level).

46. For downstream CPs, aligning the copper switch-over programme and the exchange closure programme would also make it easier for CPs to manage the substantial impacts of both programmes in parallel and clarify their decision-making in relation to consuming OR fibre or alternative fibre across a contiguous footprint.
Interplay with Outside/In

47. Openreach should also consider the interplay and potential impact on competitiveness on Government’s Outside/In subsidy scheme that may result from the Exchange closure programme. DCMS has acknowledged the centricity of the Openreach Exchange footprint to the emergence of an effective wholesale patchwork of networks participating in the scheme. In areas of no network competition for fibre/Gigabit connections, requiring retail CPs to integrate with a non-Openreach’s supplier’s point of presence could present a material barrier to adoption. Retail CPs may be disincentivised by migration and backhaul costs associated with reaching a non-Openreach network provider’s POPs – costs which are highest in the most rural areas. Where a particular Child Exchange has been closed, it is conceivable that a retail CP might face the perverse incentive of falling back to the nearest OHP and foregoing retail opportunity in areas it previously served with LLU, if the costs of rerouting to a non-Openreach POP are excessive.

48. Openreach is honour-bound to consider the interplay between removal of Exchanges in Final 20%/Area 3 and how that may restrict competitive participation in the Outside/In subsidy scheme. We’d anticipate strong views from Government in identifiable cases where the closure of a particular Exchange has the effect of undermining a non-Openreach application for subsidy (not sure how theoretical this argument is).

49. We believe Openreach should give consideration to the fact that in in areas that are borderline Area 2/Area 3, the closure of the Exchange could be the tipping point that confines an Area into Area 3.

50. Openreach should seek to establish whether such a risk is likely to emerge in practice, in consultation with potential participants in the Outside/In programme and with DCMS and BDUK.