Numis Report on Cityfibre Reveals New Details of UK FTTP Broadband Build
A new research report from investment bank Numis, which was pieced together by analyst John Karidis, has revealed some useful information about Cityfibre’s £5bn project to deploy a new gigabit-capable Fibre-to-the-Premises (FTTP) broadband ISP network across the UK and their perspective on overbuild and prices vs Openreach (BT) etc.
Just to recap. Cityfibre (CF) currently has enough funding (Numis estimates c.£2.4bn equity and c.£4.9bn debt) to cover up to 8 million premises – across around 285 cities, towns and villages (c.30% of the UK) – by the end of 2025 (here). So far, the operator has already covered 2 million UK premises – with 1.8m Ready For Service (RFS) via a supporting ISP (here). The network is also expected to pass around 800k businesses, 400k local authority sites and 250k 5G access points.
The new analyst note reveals that CF’s build costs remain around £500 per premises and the efficiencies they’ve found in this process are currently “offsetting most inflation pressures“, although the operator does remain a “little exposed to rising interest rates“, consistent with its financing obligation to hedge at least 70% of its debt over 5 year.
Advertisement
Total premises passed with FTTP are predicted to reach c.2.5 million at the end of 2022, including c.2.1 million Ready-for-Service (RFS) – the latter is where ISPs allow you to actually take the broadband service. The operator’s build rate is said to have now exceeded c.1.1 million premises per year and continues to grow.
“Near term, CF is managing its build rate so as not to miss opportunities to win government [e.g. Project Gigabit] subsidies to augment its planned commercial build. Thus, we think CF will deliver its 2022 year-end run rate in 2023, at a minimum,” said the report. CF has previously hinted that they might be able to reach 9 million or more premises if they win enough Project Gigabit contracts, so it’s a serious prospect for them, provided they can outbid Openreach and others on the larger regional contracts.
The operator is currently forecast to have 144,000 connected lines, which is up 60% over the past 6 months or so and they appear to be growing this number at a rate of c. 15,000 per month (6 months ago it was more like 9,000 per month). Put another way, customer penetration across CF’s RFS footprint – where ISPs now allow you to take the service – is climbing at a rate of c.1% a month. In some cities where CF networks are complete, customer take-up is already said to be >20% and rising monthly, although they don’t provide any examples.
Advertisement
Overbuild and Competition
The analyst then switches to examine how much overbuild and competition Cityfibre faces, which reveals some interesting information. Firstly, it’s noted that Vodafone and TalkTalk (two of Cityfibre’s biggest ISPs) have yet to connect to 100% of their committed footprints on the new network.
Currently, Vodafone and TalkTalk sell in 88% and 82% of CF’s available footprint, which we suspect may partly reflect the natural gap between new build completion and ISP onboarding of the new streets / areas etc.
The report goes on to note that TalkTalk has recently become the subject of much speculation, not least with respect to the possibility of Virgin Media (VMO2) scooping up the operator for £3bn. But that proposal appears to have been shelved (here). Nevertheless, the analyst states that TalkTalk’s “long-term contractual commitments to CF will remain in place regardless of any potential future change of ownership,” although it’s often unwise to assume such things – major contracts do sometimes get broken.
Similarly, the analyst states that it’s “only a matter of time” before Sky Broadband starts to wholesale over CF’s network as well, which is something that we’ve been expecting for the past few years (Sky has previously been openly courting alternative networks). But so far the provider hasn’t joined any AltNets and seems to have stuck with Openreach instead.
Advertisement
“The alternative scenario for Sky is that it continues to live with a material (input) cost disadvantage versus meaningful competitors such as [Vodafone] and [TalkTalk],” said the report. But they seem to forget that Sky might also have a strong interest in a third option, which could see them tie a wholesale agreement with Virgin Media and / or VMO2’s new fibre joint venture.
At this point the analyst also touches on the issue of overbuild between rival networks. At present, CF’s network is said to have an overbuild with Virgin Media’s cable infrastructure of 62% (plus another 2% if adding Virgin’s FTTP lines), which drops to around just 25% if only looking at Openreach’s FTTP network.
None of this is surprising, as OR’s fibre coverage is less than Virgin’s cable, but all three operators focus on commercial urban areas. Generally, big urban areas should be able to support four or more gigabit-capable operators. In terms of CF’s overbuild with other alternative networks (often smaller players), the figure is “negligible” at around 1%, but that will change over the next few years.
The report then states this – “we think VMO2 may well become a CF customer to extract more revenue synergies from its nationwide mobile network“, which is something we haven’t ever considered before. We could certainly see VMO2 harnessing some of Cityfibre’s Dark Fibre, but the size of VMO2’s network and future build plans suggests that they’d probably rather build their own solutions. Time will tell.
Finally, the report touches on the contentious issue of Openreach potentially launching a new ‘Equinox 2‘ discount offer on their FTTP products (here), which would further reduce consumer prices. “The risk for AltNets and ever more Britons who benefit from infrastructure-based competition is that, near term, OR competes on even lower prices to halt growth in infrastructure-based competition from AltNets, and then raises prices after it re-establishes its network duopoly with VMO2,” said the note.
Numis Statement
We are not surprised to hear that OR wants to offer further line rental discounts for its less popular higher-speed products, and further discounts to all one-off connection charges. We think:
● First, merely ongoing awareness that OR is working on Equinox 2 hurts AltNets who wholesale their FTTP networks (either exclusively, such as CF, or as well as retailing them, such as Community Fibre) since ISPs must ‘wait and see’ to get clear information to make economic decisions.
● Second, connecting a home to any FTTP network for the first time is a major pain point for customers. They need to be at home and it requires drilling holes in walls. Therefore, we think many FTTP customers are unlikely to connect to an alternative network as well.
Consequently, we think Equinox 2 cannot be allowed to offer even lower connection charges because this will be, effectively, anticompetitive.
We’re still not entirely convinced by the analyst’s arguments against allowing Openreach to price their own products more competitively, particularly in dense urban areas where aggressive competition is natural and to be expected. But it is also easy to understand why CF would be concerned and there remains a question mark over where Ofcom might draw the line with a dominant player, or if they even draw one at all with future discounts.
The analyst makes the point that Ofcom waived through Equinox “in part because it estimated only [a] little FTTP network overlap between OR and AltNets in the short-term (even though the regulator did not ask any AltNets about their rollout plans).” But it points out that CF is already at 25% overbuild with OR – just 12 months after Ofcom approved Equinox 1.
At the very least, we are expecting that Ofcom’s review of Equinox 2, once it’s actually revealed to the public, will need to do a deeper dive and take more account of the impact upon AltNets. But whether that produces a different outcome is currently very open to debate.
Mark is a professional technology writer, IT consultant and computer engineer from Dorset (England), he also founded ISPreview in 1999 and enjoys analysing the latest telecoms and broadband developments. Find me on X (Twitter), Mastodon, Facebook, BlueSky, Threads.net and Linkedin.
« Starlink Broadband Introduce 1000GB Fair Usage Policy in Some Markets
Virgin Media O2 UK to Create 125 New Engineers in Big Cities »
Please spell my surname correctly. It’s Karidis
A nice summary of the state of CityFibre progress, but nothing new the Numis report. Funny that BT say they aren’t seeing any impact from altnets – but CityFibre are gaining 10-15k subs a month and Openreach are now losing 20-30k subs a month…
The final paragraph in the Numis statement doesn’t make a lot of sense to me:
“Consequently, we think Equinox 2 cannot be allowed to offer even lower connection charges because this will be, effectively, anticompetitive.”
Competition is based on providing a product, or service, at the cheapest possible price and still turn a profit. As long as Openreach aren’t offering a service at below cost, it couldn’t possibly be anticompetitive. Openreach appear to have the cheapest build costs and their scale allows them to offer the cheapest wholesale prices should they choose too, again this couldn’t be described as anticompetitive.
On several occasions, Jansen has stated Openreach have enough pricing flexibility to lower prices as required, to remain competitive, the Equinox offerings are likely backing up those statements. Equinox (1) is about encouraging Openreach’s wholesale customers to migrate off FTTC and onto Fibre, Equinox 2 will likely offer more discounts to wholesale customers.
No doubt Jansen is sending a message to prospective Altnet investors, telling them BT/Openreach are going to aggressively fight competition through pricing going forward. This aggressive messaging is quite new and if I was an Altnet investor, i’d be wondering how safe is the return on my investment? The Numis statement answers that question.
Predatory pricing. A company with market power. Competition and markets authority. That’s how
A few quick points, if I may.
First, BT’s OR is cutting prices, I think, to choke off network-based completion, and at the same time, BT’s Consumer is raising prices by CPI+3.9%. How does this combination help consumers?
Second, OR has a massive existing business to defend, whereas AltNets do not have existing businesses to defend. Thus, they, not OR, can afford to price lower. This, also because their build costs remain under control + they incur tiny operating (//running) costs where their networks are complete. After all, everyone, including OR, is using OR’s ducts and poles. Note also that OR’s ‘below cost’/‘above cost’ argument in effect assumes that much of the existing, non-FTTP, business shrinks fast. In other words, this ‘below’/‘above’ cost argument totally ignores OR’s existing business.
Third, Ofcom itself noted that AltNets need a few years (3 perhaps?) of some protection to gain reasonable size. I think waiving through Equinox 1, 2, etc sits at odds with this.
Lastly, if assuming OR will raise prices when it reestablishes its duopoly with VMO2 is speculative, how would you characterise the assumption that it will not do so? (clue: very many years of price regulation by Ofcom)
If you reply, please reply to all these points. Thank you.
“First, BT’s OR is cutting prices, I think, to choke off network-based completion, and at the same time, BT’s Consumer is raising prices by CPI+3.9%. How does this combination help consumers?f you reply, please reply to all these points. Thank you.”
Openreach is effectively a separate company from BT, with its own management and Board of Directors. BT’s consumer division is effectively a customer of Openreach and will receive the same discounts as Openreach’s other wholesale customers; It’s entirely up to BT ‘s consumer division as to whether they pass on price reductions gained on the wholesale prices, same as all the other Openreach wholesale customers. Competition should drive the price the consumer pays, so the various other Communication Providers will drive that competition with their own pricing
“Second, OR has a massive existing business to defend, whereas AltNets do not have existing businesses to defend. Thus, they, not OR, can afford to price lower. This, also because their build costs remain under control + they incur tiny operating (//running) costs where their networks are complete. After all, everyone, including OR, is using OR’s ducts and poles. Note also that OR’s ‘below cost’/‘above cost’ argument in effect assumes that much of the existing, non-FTTP, business shrinks fast. In other words, this ‘below’/‘above’ cost argument totally ignores OR’s existing business.”
Openreach’s legacy business is currently profitable, but shrinking. Openreach don’t own the ducts, Fibres, Cables, etc, they just manage them on behalf of BT Group. Openreach is basically a workforce whose income is gained by providing a service to their wholesale customers, their biggest being BT. BT pays for their own Fibre rollout through income and some debt. Altnets are entirely funded by investors and debt, those investors will expect a return at some point. Running a network isn’t cheap, the Altnets need to maintain a head end, whether it’s in a datacentre or in their own building and they also need engineering and NOC support; Running network operations is around the same cost whether you have 100 customers, or 100,000 in an area. From BT’s perspective, as their Fibre rollout progresses to completion their Capex will drop massively, and as the building closure program progresses the Net Debt will also decrease, since £5.5 Billion of BT’s £19 Billion total Net Debt is Lease Liabilities; Also, because of the PSTN switch off and the move to a converged PON/IP operating model, they’ll have a much more resilient and power efficient network. BT and Openreach will need far less staff than they currently employ, so I’m guessing Peter Jansens forecast of £3 Billion annual cost saving by 2026 will be superseded by more cost savings further on. Because of all this, I don’t see how the Altnets will ever be able to compete on operating costs once BT have finished their rollouts, this all has a 1990’s Deja Vu cable company feel to it.
“Third, Ofcom itself noted that AltNets need a few years (3 perhaps?) of some protection to gain reasonable size. I think waiving through Equinox 1, 2, etc sits at odds with this.”
I’m not sure Ofcom care that much about the Altnets, I suspect the original plan was to turn Openreach into something like Telecom equivalent of the National Grid in the power industry.
“Lastly, if assuming OR will raise prices when it reestablishes its duopoly with VMO2 is speculative, how would you characterise the assumption that it will not do so? (clue: very many years of price regulation by Ofcom)”
Anything could happen with Openreach, the workforce may be spun off with long term contracts supporting their revenue and valuation. Ofcom could regulate Openreach’s wholesale pricing ad-infinitum, its anyone’s guess.
Thank you for replying. A few observations, if I may.
First, OR may be a separate company in theory, but BT owns 100% of it. Also, I think your answer to my second point is at odds with your answer to my first point. On the one hand, OR is an independent company, but on the other hand, OR is just a ServeCo and BT pays for the FTTP rollout and takes the hit from OR’s legacy business shrinking.
Second, in 2016, Ofcom switched from wholesale price regulation to effecting infrastructure-based competition (with a view that the latter will render wholesale price regulation redundant in time). Example: regulated access to OR’s/BT’s ducts and poles. Absent the switch in Ofcom’s focus, private investors/numerous banks/the state’s UK Investment Bank would not have ploughed billions into AltNets.
Third, respectfully, I don’t think your reply to the last point holds water. For one, it contradicts Britain’s experience for the need for wholesale price regulation over many years.
Lastly, and I’m on thin ice here (and probably you’re better placed than I am to know), but I thought, to date, OR staff opex is well in excess of network opex. As for OR savings to come, I think we’re looking for these to start materialising c.5 years from now at the earliest (tiny, statistically insignificant but interesting example: note OR now plans to build FTTP to c.3.2m properties in FY23, vs 3.5m before).
I don’t really understand the points you’re highlighting. In simple terms scale counts in Telecoms, the more scale the lower the operating costs.
The key events leading to BT growing Net Profit are:
1) The move to FTTP
2) Switch off of PSTN and closure of 4,600 Exchanges
3) FTTP/IP/Cloud convergence requiring less operating costs, with a much simpler operating model
4) Less staff required due to points 1, 2 and 3
5) Big reduction in Capex as FTTP and 5G rollouts end
6) Pension contributions reducing in line with the deficit going into surplus
Adding everything up, even if Openreach lose a chunk of customers to Altnets and VMO2, their scale will still give then an operating cost advantage over all the competition, and it remains to be seen if the Altnets will be able to attract customers in significant numbers. Altnets seem to be popping up all over the place, with much of the investment aimed at building to sell to a bigger player, but geography and backhaul costs could make a takeover bid unlikely, especially if the for sale Altnet has failed to build a good customer base. I’d bet on BT/Openreach doing well going forward, I’m not so sure about the majority of the Altnets, and hoping that Ofcom will somehow pin BT down, to allow Altnets to do well, is a wing and a prayer way of doing business.
Some of these posts are huge and I’m not going to even attempt to answer them as I have neither the knowledge or expertise.
One key difference between altnets and Openreach, though, is that for the lowest possible initial CapEx altnets have to consume PIA and Openreach don’t. Add to that the economies of scale Openreach have and their pre-existing workforce and it seems reasonable that the same regulation applied to BT Wholesale to prevent them squashing LLU in the 2000s be applied to Openreach: that they may not price below a level that would undercut a ‘reasonably efficient competitor’.
Of course the number of altnets is unsustainable however it’s reasonable for the majority of the country to have BT + VM + CF + 1, London BT + VM + 2, tailing off to BT + VM + 1, mostly BT + 1 and finally ‘1: whomever’.
* significant market power
Predatory pricing is defined as selling below cost, Openreach don’t and won’t need to do that.
Significant Market Power is more about the ability to raise prices above marginal cost without losing revenue; BT are currently in litigation, based on alleged SMP in relation to raising prices of landline only services. I don’t see how Openreach dropping prices can be described as SMP, since they’re not raising prices above marginal cost and the consumer isn’t damaged by a drop in the price. To say Openreach will raise prices down the road is speculative.
Should Altnet investment be protected by forcing a company to maintain high prices? Probably not, since the investors should factor in any drop in pricing by an incumbant in the face of stiffening competition.
If the Altnets can’t make a profit just because Openreach drop their prices to be more competitive, with Openreach still profiting despite lowering prices, then the Altnets and investors should have forseen Openreach/BT fighting back.
The problem with that Ex telecom engineer is that BT have a built in cost base advantage and existing penetration so can destroy that competition, then raise prices later with no alternative to keep them honest. Competition benefits the consumer, Openreach fought against any upgrade of their network for decades saying ADSL was all anyone would ever need, then the same with FTTC. The early alt nets (Hyperoptic and the like) showed demand was there but infrastructure upgrades take a huge amount of money and time to get built so there needs to be some element of business environment that’s needed before any of these builds can happen. Openreach dropping their prices to cover their operational expenses only destroys competition and once that is done Openreach remain as the only company standing and then prices rise and product offering remains static
@Ben – how does that explain why 3 altnets are building along the same streets which are also due to see OR? Why do they need any support?
@Somerset – I agree that doesn’t make any sense at all, I expect in 5 years time there won’t be 3 alt nets in your street. The more general point to say Openreach shouldn’t be subject to price control. The Alt nets aren’t getting support as much as Openreach aren’t allowed to destroy the market case for having as much of the market not subject to a monopoly due to significant market power (ie Openreach already has the overwhelming majority of people on its network) which it is able to abuse to crush competition due to an inbuilt advantage from when it was state owned
City Fibre wholesale pricing is below current Equinox and offers better speed. I don’t see the entry point product on OR reducing much if at all. It is not just overbuilding against OR, CF are overbuilding against VM. I would like to read the report in full as the main competition in my view to CF may come from other Altnets using OR PIA (e.g Cityfibre, VM and FW in Crawley).
An additional factor is the overall ISP price offered and whether the ISPs using CF can gain market share which CF are not in control of. Our economy will be stalling for effectively 3 years and consumers are likely to keep to the entry level products unless the ISPs take the hit price wise.
With OR and VM targeting the majority of premises most of Cityfibre will be overbuild. OR will only have significant market power in non-competitive or BDUK areas.
These AltNets are shoving in their network whatever the cost, destroying existing OR network as they go, but that doesn’t matter as BT are evil right..
Literally none of it is to standard, cables are left flopping around, unlabelled and unprotected.
Their subcontractors are risking their own lives and of the public’s installing these cables without adequate guarding/TM as well. Numerous instances of this and no one cares.
Good luck if you’re investing in or using these networks.
@Bob
Then they should be kicked off of the PIA product. No excuse for not following OR guidelines which they should sign up to when they apply to use the product. OR should and are made to allow access but there has to be responsibility taken by the Alt nets when they use the network, otherwise they lose that access privilege
‘Literally none of it is to standard, cables are left flopping around, unlabelled and unprotected.’
Can’t say that’s been my experience with the builds in progress around here.
Name names given literally none of it is to standard?
If Cityfibre think Sky is going to save them from poor sales, think again. Sky have spent millions re-training engineers to install Openreach FTTP from the pole to end user.
Cityfibre offer no value to Sky in the long term. Especially with Equinox 2 coming.
Vodafone and TT better pull their socks up.
@John Karidis – Current customers 144k/2.1 is <7%. Your projecting a build of 1m a year and 15k orders a month. Customers numbers remains low. What is the ambition? What is needed to survive?
OR have just reported 26k FTTP orders a week, although less reported is that FTTC is reducing by 16k a week.
The reference to Gigabit! Do you think Gov will actually fund overbuilding recently subsidised FTTC (30-80Mbps) services? Consultations since 2019 and a few picture opportunities for Boris/Nadine does not suggest any sort of great public need is being met.
I am surprised they’re still at £500 per premise passed.
Seen reports, mainly from OR that CF are at £1400 per prem.
Not sure where the £1,400 figure for urban build comes from (seems unlikely as an average), although I do think CF have probably been spending a little bit more than they planned on civils. On the flip side, CF are trying to build as much of their own infrastructure as possible, so they don’t end up being too over-reliant on PIA. The latter is a regulated product today, but that may not always be the case.
It was a bit sticky however increased use of PIA alongside moving to one and done digging rather than digging the metro core is helping. They started off basically zero PIA then moved to using it for drops and are now using it, where economical, to get from their metro core to an Openreach chamber and from that chamber to homes, with the only deeper dig being a cabinet and duct to connect that cabinet to Openreach plant.
The city I live in they are doing pole lead-ins and drilling into Openreach plant as they dig the metro. The one I lived in previously the ‘main road’ outside where I used to live is receiving its fourth dig in the same place from CF.
£1,400 per premises seems way too high though. VM’s early hybrid fibre-coax Lightning build, full size ducts and involving zero PIA, came it around £660 per premises passed, CF’s early zero PIA build would’ve come in at less given it made use of microtrenching and shallower depth.
I suppose it goes without saying but Openreach are hardly likely to give generous numbers for CF’s build costs.
In the latest BT Group report they state that the OR FTTP per premises has reduced to between £200 and £350. Partly because they are using basic GPON and existing infrastructure
Cityfibre have chosen to only use only the OR pole for PIA and are installing their own distribution duct and patching cabinets. The additional civics costs will mean their costs are higher but the UG micro duct lead in to the customer premise boxes is barely 2inches below the surface so more liable to damage going forward.
Hi all, Mark’s piece quotes quite a few parts of my research. Therefore, if he wishes to do so, he is welcome to provide a link to my entire note. For one, this may answer the question posed by ‘NGA for all’…
Hi John. Is there a public link to it? The report was emailed to me, but as a rule we don’t host third-party reports on our own server.
Sadly not, Mark. The link is for Numis clients only. However, as you have it already, I don’t mind if you wish to make it available. I totally understand and respect what you say though. Never mind. Thank you Mark.
CityFibre enjoy saying how much the local economy will benefit from their rollout, ignoring all the other suppliers.
The study by the consultancy Hatch, (Economic Impact of Full Fibre Infrastructure from CityFibre’s Network), estimates that, over a fifteen-year period, the positive impacts of CityFibre’s £22m investment in Bath will include £122m in productivity and innovation gains, £27m from a widened workforce, £5m in Local Authority efficiency savings and £141m in increased housing value.
Yep all double counting and ignoring the actual customer demand for speed. Across the board from broadband to HS2.
I bet their business cases do not include the VM expansion and migration to XGS-PON nor that OR will switch to XGS-PON at some stage and already have the filters installed for an overlay on existing FTTP.