Openreach FTTP – Final 10% of UK Likely to Cost £4000 Per Premises
The CEO of BT Group, Philip Jansen, has revealed how Openreach anticipates that reaching the final 10% of UK premises with 1Gbps capable Fibre-to-the-Premises (FTTP) broadband ISP technology is likely to require an “outlay of around £4,000 each to pass” (vs build costs of £300-£400 for the first 50% of premises).
At present Openreach aims to reach 4 million premises by March 2021 with FTTP and there’s an ambition for 15 million by around 2025 (out of c.30 million total). After that they also have an aspiration to reach “the majority of the UK, if the right conditions to invest are in place” (e.g. 20 year business rates holiday on new fibre, more skilled labour, easy migration from copper to fibre, easier wayleaves, mandated full fibre for new builds and access to large buildings with absentee landlords etc.).
The latest cost estimates for this are close to Ofcom’s arguably more optimistic model of FTTP deployment (here), which indicated that the capital expenditure needed to do FTTP for the first c.20 million UK premises is under £500 per premises but above that point it quickly rises to nearly £2,500.
We’ve always felt that £2,500 was a bit too conservative but accurately estimating such things is rarely easy and will vary between operators, depending upon the approach to deployment and business model of the operator. Clearly £4,000 is reflective of Openreach’s modelling but other operators may be able to do it for less (e.g. B4RN’s volunteer built network).
Jansen, who was speaking to shareholders after Friday’s results announcement, also confirmed that Openreach’s latest full fibre build update (here) would boost their rollout pace from 20,000 premises passed per week today to 30,000 per week by the end of this financial year (March 2020). So far 1.5 million premises have already been completed and Openreach’s Clive Selley said they added c.350,000 in the last quarter alone.
Philip Jansen, BT Group CEO, said:
“Alongside these enablers across the country, public funding will probably be needed to bring FTTP to the hardest to reach 10% of the UK population [mostly rural areas but a few urban ones too] and to address enduring mobile not spots. It is crucial at this that this public funding that is used as efficiently and effectively as possible without undue bureaucracy and duplication to make sure that the UK gets the maximum digital bang for its buck.
We agree that full fibre can be the platform for the UK’s future prosperity and we’re determined to lead the way. No company is doing more.”
Jansen also added that the operator was “ready” to train, recruit and re-prioritize their skilled labour into places that that would “help us to build at a pace that would meet Boris Johnson’s ambition for a huge step change by 2025.” Meanwhile the wait continues for the Government to reveal details of how it will achieve the new 2025 target for universal full fibre coverage, which is something that we examined in a bit more detail on Saturday (here).
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 and Linkedin.
« The Challenges of Boris Johnson’s 2025 Full Fibre for All UK Pledge
Latest UK ISP News
- FTTP (5538)
- BT (3518)
- Politics (2542)
- Openreach (2300)
- Business (2267)
- Building Digital UK (2247)
- FTTC (2045)
- Mobile Broadband (1978)
- Statistics (1790)
- 4G (1669)
- Virgin Media (1625)
- Ofcom Regulation (1467)
- Fibre Optic (1396)
- Wireless Internet (1392)
- FTTH (1382)
Looks like a lot of people will have to pay to get a FTTP line as according to the Ofcom website:-
“A consumer or business will only have to contribute to the cost of a connection if it exceeds a threshold of £3,400. Above that level, properties can still be connected if they pay the excess costs or do some of the work themselves to help bring costs down.
There is also the option of using alternative technology, such as commercially available satellite outside the USO scheme, or keeping their existing service that delivers a lower speed.
To help as many USO connections as possible fall below the £3,400 threshold, the USO providers are required to assume and aggregate the level of local demand. This potentially brings more connections under the reasonable cost threshold by sharing the costs.”
At this stage the USO doesn’t really come into the Government’s 2025 target in the sense that the former is a legally binding minimum and will mostly be delivered via 4G, with only a little bit coming from FTTC or FTTP.
It may also be possible to stack the USO £3,400 test limit with Gigabit vouchers, however Ofcom have not been able to confirm, or deny if they are stackable at this time as it is a decision for government.
It would be interesting to know what the average cost per home would be for the entire 100%. Of course there are the “cheaper” ones at £300-£400 but if they were budgeted at for example £600 each then BTOR would surely be money up by a fairly large amount and are effectively asking for money they have already have and are just making noise over having to spend it?
@Steve: Good question! What would be the average deployment cost per premise?
Make me wonder how they ever managed to build a telecom network for the final 10% in the first place? Something isn’t right here.
Well, some of them aren’t? There’s the same service obligation with a phone line, but again they’ll charge for the cost over £3400.
Most people have chosen to live in places with phone service. For new builds, it’s part of what you pay when buying the house, and it usually costs less because you’re building more than one house at once, so the cost is distributed.
Some individual homeowners may have paid a lot in the past to get it. I don’t know what the policy was when it was owned by the GPO but I doubt even then you could get a free line to nowhere.
@Steve
No commercial organisation would budget properties at an average of £600 if they believed the average was £300-£400.
If they budgeted the higher figure, it would mean they had less to invest in other parts of their business and it would make the business case for commercial FTTP deployment worse.
Steve, let’s break this down.
‘It would be interesting to know what the average cost per home would be for the entire 100%. Of course there are the “cheaper” ones at £300-£400 but if they were budgeted at for example £600 each then BTOR would surely be money up by a fairly large amount and are effectively asking for money they have already have and are just making noise over having to spend it?’
So you’re saying Openreach will somehow be in the money by artificially budgeting more for the Fibre First urban properties and using the amount they’ve ‘underspent’ to deliver FTTP in harder to reach areas? They’ve have more money by budgeting higher pricing?
BT Group would be sued to hell and back for attempting to do this, it would be lying to investors, and there’s not some magic money tree where Openreach budget something and the money appears in their hands. Much of it’s being funded by debt at the moment.
The budgeting isn’t done across the entire country, it’s done per project. Leeds, etc, were selected as their costs could come in under a certain price per premises, and the exchanges within there for specific reasons.
By your logic Openreach could budget 20k per premises and would be seriously in the money. They would also build nothing as there is absolutely no prospect of them making enough back on the asset to fund the debt the group would have taken on. Openreach are only able to sell to other providers within a regulated framework and their spend has to reflect this. Virgin Media can spend more as they sell the services over the network themselves. CityFibre can spend more as their model is also different.
Not possible to simply throw them all in a pot and average them out – while it’s equitable, socialist, etc, you would be looking to renationalise Openreach and others and follow the Australian / Singaporean NBN model. This would be outside the scope of commercial operators.
No @CarlT you’ve clearly mistaken what I’ve said. I am asking if BT have worked out an average cost per property?
Then part 2 is how many properties have come under that valuation in comparison to those that come over? We can clearly see the range is some £300-£4000 already but it would be interesting to know the figures used in the notebooks in BT planning.
This is the second comment of mine you have taken the wrong way and then tried to speak down to me. If you want to be superior then go for it, just make sure you understand what the adults are saying first.
‘Of course there are the “cheaper” ones at £300-£400 but if they were budgeted at for example £600 each then BTOR would surely be money up by a fairly large amount and are effectively asking for money they have already have and are just making noise over having to spend it?’
Was hard to understand and was misread by others much as I did.
BT will have estimates but, of course, not exact figures as they haven’t surveyed everything.
The average itself is quite irrelevant. What matters is how much Openreach are prepared to invest per premises. Everything above that is into the territory where Openreach would want subsidies. Whether that’s the 3-400GBP they’re spending to build in Salisbury or more is something none of us here will know.
BT underspending in some areas where they’ve come in under budget is also irrelevant. They may choose to broaden their coverage with that money or they may choose to keep it. They aren’t obligated to spend a certain sum of money on the project and have only mentioned coverage, not the amount of money allocated.
The commercial rollout isn’t BDUK. There are no contracted coverage or spending commitments. There’s apparently no pot of cash dedicated to FTTP – note Openreach gave exact sums they were spending on the FTTC project but have not mentioned anything regarding FTTP for good reasons.
I would recommend a read of the research at https://www.nic.org.uk/wp-content/uploads/Cost-analysis.pdf – it’s eye-opening.
Lastly I apologise for being patronising but could only work with what I had. If I’m misunderstanding you please consider writing more clearly so that you get your message across better.
£12bn for 3m lines does not look credible. If you were to plan each line individually and assuming you had no existing network then this might be credible. But the existing BDUK 350k + Cornwall 90K FTTP cost a fraction of this and there is plenty of this in very awkward areas.
We are back the to £100k a cabinet myth (2013) or the equivalent £500 per premise passed for FTTC used by BT as late 2016 in CMS SC inquiry into Broadband.
Quoting £12bn at this stage is not helpful.
They were selected based on value for money for the taxpayer. The whole point was it was more viable to deliver FTTP to them than FTTC.
They are not a valid baseline. The potential costs of the last few percent are well documented. If you were to capitalise all the free labour B4RN are using or require them to either pursue charged wayleaves, hard dig or use poles their numbers would be right up there.
EDIT:
‘Quoting £12bn at this stage is not helpful.’
Assuming it’s accurate the truth’s job isn’t to be helpful, it’s to be the truth. You may not like it, it may disagree with the message and agenda but if it’s accurate there’s no point in trying to pretend otherwise.
for example CDS Gainshare planning is £6M+ for 2000 prem – or £3k each average. Similar costs are coming through in all the contract extensions reinvestment.
If 50% of the country costs £300-400 average and 50% costs £4000 average yet the final 5% is over £3k you can easily work out the final tail is horrendously expensive to distort the average that much (for openreach, others may be able to do it cheaper by other methods)
EDIT (typo)
If 50% of the country costs £300-400 average and 10% costs £4000 average yet the final 5% is over £3k you can easily work out the final tail is horrendously expensive to distort the average that much (for openreach, others may be able to do it cheaper by other methods)
That 3k doesn’t include Openreach’s 3-400GBP per premises budgeted to of course.
These are the real numbers on the ground. If they are incorrect I’m sure a judicial review will be incoming, however these are comparable to similar nations.
Yes I suspect it’s been vastly over estimated, so the share holders can pocket the substantial difference!
Some countries to build an entire fibre network from scratch for that much..
Typical Openreach though. Over priced and with a never ending timescale..
‘Some countries to build an entire fibre network from scratch for that much..’
Countries where wages are lower, safety and employment standards are far lower, and more of the population live in apartments?
Singapore managed it with a taxpayer subsidy of about 450GBP per premises. Singapore is, of course, a city-state.
Go outside the hotels and tourists areas in many countries and you can find things incredibly cheaply relatively to what we’re used to paying for things. There’s a reason for that.
I thought 97% of the UK now had access to fibre broadband according to Openreach? How has this number jumped back upto 10%?
As someone in the 3% (or 10%!) living in rural Aberdeenshire, I find it slightly puzzling why Openreah cannot use a lot of the existing network. Given that we pretty much all have copper lines running to our properties, why can they not use the existing ducting / telegraph poles?
Hi
They will of course use existing ducting if available, or use the existing poles. The issue is a lot of these hard to reach places may have directly buried telephone cable so no ducts, and the information of where that is exactly may not be recorded that well. Overhead poles often will need replacing, especially in our health and safety conscious times before someone can climb up it. This all adds time and cost.
I quite agree though they do have an existing network that helps them, and Openreach may spend more for these harder to reach areas for FTTP, but they don’t mention that of course they will also save more on future maintenance costs.
If they want to retire copper, then I think the legislation should be where there is copper already it has to be replaced by fibre before they can decommission the copper part of the network.
‘If they want to retire copper, then I think the legislation should be where there is copper already it has to be replaced by fibre before they can decommission the copper part of the network.’
It is. They can’t switch off copper full stop right now and where there are plans to, Salisbury, they stop selling new copper products and, eventually, people are compelled to move to fibre.
There are no plans, it’s not on the agenda, to allow Openreach to simply stop providing copper – they must replace with fibre before any decommissioning in all scenarios. At the extreme the same kind of court orders allowing utility providers access to meters would be used to fit ONTs.
@Stephen: “I thought 97% of the UK now had access to fibre broadband according to Openreach?” Openreach is nowhere near 97% fibre coverage, more like just a few percent. I think you are wrongly equating “fibre” with VDSL. You wouldn’t be the first one, thanks to ASA not doing its job in preventing misleading adverts and claims.
“It is. They can’t switch off copper full stop right now and where there are plans to, Salisbury, they stop selling new copper products and, eventually, people are compelled to move to fibre.”
I’m confused, so why are talking about USO over 4G/5G rather than FTTP being deployed for those people? Openreach long term aim is to retire copper as it’s becoming expensive to maintain, is only degrading more and more as each year passes, and isn’t fit for the purposes of reliable and fast data connections. If every copper telephone line has to be replaced by fibre before they can retire copper, then all these people on very long telephone lines in remote areas are going to get FTTP, the question then is ‘when’ rather than ‘if’. Openreach are going to have to pay for it either way.
@Phil
“I’m confused, so why are talking about USO over 4G/5G rather than FTTP being deployed for those people?”
The USO provider is BT Group not Openreach. It will use Openreach for any fibre provision and EE for 4G, depending which is the best value for money to a given location.
Tying together copper retirement with FTTP deployment could yeld cost savings. For the most part I have a very good idea where the copper line is located for the last 3km, at least within 1 meter, due the the number of faults over the last 15 years, as it snakes under the verge at the road, officially it will be considered direct buried, though I suspect it was direct flung in the ditch and as it silted up over the years it got buried. The many joints boxes and where its flung over a dry stone wall confirm its location. If the new ducting was mole ploughed in the only service present is the existing phone cable, which would more than likely destroy it, so its retirement would be very convenient.
Doesn’t this mean anyone who is willing pay a small fee of £600 in the USO area could be possible to get FTTP?
I think the £4000 is an average while the amount charged under USO will be specific to each property – so I think some properties will require a top-up of much more than £600 to get FTTP.
No, definitely not.
It’s £500-£600 per premises IF they roll out to an entire area. That brings the cost down considerably.
Economies of scale.
FTTPoD (On demand) cost from £1,000’s to £10,000’s.
Even in urban areas with FTTP nearby these bespoke FTTP builds tend to cost over £5k.
There will be nothing in the US0 where you can insist on the method of deployment, or make the operator do your whole area so it is cheap.
That last bit is somewhat wrong John, the USO very much states that operators are to assume and aggregate potential local demand to bring the most properties under the cost threshold,
Remains to be seen how that works in practice.
I do love a good Average, If that’s what it actually cost I’d write a cheque today.
Yes £4,000 is of course an average.
My view is that it is good to remind the politicians and others of the likely real cost of what they are saying. particularly if their intention is that future broadband costs for the 10% are to be kept in line with urban broadband or where the Government will continue to pursue a competitive model. They need to be clear regarding any direct funding model or a cross subsidy model as well as realistic timescales.
The model used in rural Wales (Ceredigion) actually works quite well, where overhead reinforced fibre is brought to the existing telegraph pole terminated to a Green Fibre Splitter. Because the cable is reinforced, it does need to avoid overhead electricity cables, but large sections get installed fairly quickly otherwise outside these hazards.
This allows BT to bring fibre close to the premises without having the full cost of installation/connection to every home, where a customer hasn’t yet taken the service.
BT could offer homes a discount, if a community of homes are finished off as a group, with a higher connection charge where the owner decides to wait.
The cost can’t average £4000 a home, based on this model of rollout, BT are pushing for handouts as per usual. BT have already stated they can’t operate both Fibre and Copper Networks simultaneously, so why should pay BT to remove their legacy kit (they have to anyway).
We’ve already paid BT vast amounts to supposedly install Superfast “Legacy Copper based’ Fibre-FTTC, instead in Wales at least, where FTTP was to be deployed, many areas have been left with unfinished fibre coils left to decay, with no legal enforcement on BT to tidy up their siteworks (No sure why this hasn’t happened).
Too many clueless MPs/Politicians believe that “BT will do the right thing”, this is not the case – “BT will do the right thing for BT, end of”. We need to stop giving them handouts to get rid of their legacy kit, for basically sitting on their hands.
That’s making the (very incorrect) assumption that all properties have telegraph poles in the first place.
Many rural lines have copper cables buried directly in the ground. No ducting and no poles.
New telegraph poles would be needed or installing new ducting would be required.
At around £500 a telegraph pole or over £110 per meter to dig on a carriageway it’s very easy for the average cost to be £4000 per premises.
Some will indeed cost less than this, some will cost CONSIDERABLY more though.
That’s what an average is.
I don’t think OpenReach are claiming every property would cost £4000.
As you get down to the last couple % of rural homes the distance needed to cover increases while the number of homes covered decreases.
https://www.nic.org.uk/wp-content/uploads/Cost-analysis.pdf
Last 10% = ~3 million premises.
Page 57 gives a cost to serve those premises of 8-9 billion assuming extensive infrastructure re-use is possible, page 54 gives 12-13 billion assuming no infrastructure re-use.
The 4k per premises for the last 10% is not outside of the range of that independent research. Obviously the figure is at the higher end of the estimate, as you’d expect, but is neither unreasonable or unrealistic.
Sorry – didn’t mention. The Openreach estimate will also include the operational expenditure as they did for FTTC. That’s another billion on top of the build costs give-or-take.
With that in mind the 4k per premises, adding together OpEx and CapEx, seems a fair, if somewhat high, guesstimate.
Reality doesn’t care how unhelpful it is unfortunately, NGA4All, it is what it is. The costs could be substantially reduced however that would require extensive reform to wayleaves and streetworks methods. Give BT B4RN-style access and they’ll be able to deliver much more cheaply.
That typically 11 years of operational costs, and so we are back to some contribution to the incremental average costs £2,000- £2,500.
There are an outstanding 400,000 premises under existing BDUK contracts, taking us beyond 97% for superfast but using FTTP to do so.
There is another c£670m of Capital Deferral + Stratum (£165m) and R100 in Scotland to come. Suffolk, Cumbria and Essex all doing more FTTP at the fringe.
If BDUK were to keep going and conclude at 99% ‘superfast’ by doing 2% (600k) FTTP at the fringe, the government would not need to get involved in the fibre transition.
In English, please, alongside the relevance of BDUK clawback when not yet committed.
Also, evidence that the costs are as you describe. All the research I can find points to 3-4k per premises for the last 10% with much of it the last couple of percent.
Just remember the 100 in R100 refers to coverage, at a minimum of 30Mbps which leaves open FTTC and FWA as solution technologies not just FTTP, so as far as I can see removing those numbers from your list cannot be guaranteed.
There is no point in some speculation about operational costs to bring down the notional capital costs as no network provider is going to put forward a case based on cost of kit alone, it has to have customers provided, kept running and regularly maintained as well as some power to light the fibre and handle the data and the usual depreciation etc as well.
There is also no reason not to believe that Pareto would not apply, in which case 80% of the final 10% would come in around £1,000/premises and the final 20% of the last 10% (which spookily is 2% of the total) would be around £16,000/premises. Just to say that this game, especially at the tail, is not linear and applicable to simple pro-rata.
Gadget – R100 has some £600m available and I think <30Mbps is est 170k premises for Scotland, so the problem or most of it is addressable. The problem is resource and perhaps opportunity cost than actual cash. We can actually take 60k off this number as they SCot/BT announced another 60K mostly fttp from existing funds.
Carl T ..I will try the English. It may take another go. The Capital Deferral is relevant as it is monies intended for Broadband. The CD is already capitalised and included in the cost recovery for broadband – so why would you not find a way to actually spend it on Broadband – election agreements like Stratum is one way, you can trace to the EFRA SC in 2017.
If BDUK can contract another 500k premises where needed in England which will be mostly FTTP.. then just about anybody can order the connectivity they need.
If this occurs, the Gov do not really need to be involved in what will be a BT/Openreach/Industry debate on transition and product definitions. Government only needs to be involved where the supply of fibre is artificially constrained. Completing in-fill where needed would make a big contribution. If Suffolk have a plan, then why not the rest?
There could have been more of a cross over in rural as you should doing things once every 25 years. Hence upgrades should always have been a good mix of cabs and lots of in-fill and planned that way. We only got some of the latter (so far) as BT bid strategy for BDUK was a gambit, based on the original EU Cornwall contracts where subsidies and confidentiality was maximised at the expense of a more complete engineering outcome. The coverage templates were product of the gaming of costs. Cornwall which we must salute has still a long way to go, largely because of the lack of scrutiny on the funding.
The Gov now does not need a transition plan to fibre, only act selectively where services cannot be ordered. FTTP is not an end in itself. Its original role in a 'mixed economy solution' (remember that) was fine. Those interested in rural are still waiting for the mixed economy solution to be delivered something lost in time about 2013.
If the Government needs to do nothing on transitioning to full fibre then why do we have things like LFFN and RGC and a Prime Minister pushing for 100% coverage?
On your accusation that “Suffolk have a plan, then why not the rest?” can I suggest you take a closer look at what is going on across the English and Scottish local authority areas.
I provide a number with a reference behind and could point to others and get this back.
Unless you’re in charge of the programme this level of fixation with BDUK can’t be either healthy or profitable.
Unless it’s in the contracts BT don’t have to do a thing with the gainshare money beyond not bill it as they have been or return it to the public sector. It should not and cannot be taken for granted that it will go into broadband.
Andrew .. Suffolk is an example. (NI , Scotland already referenced) .. but UK rural is 500k ish short in England in terms of rural premises not under contract. If PM understood he might re-direct his attempts. Vouchers look like a poor substitute for completing works at the edge. LFFN looks like encouragement for mostly Cityfibre to offer alternatives to private circuits in public sector buildings.
Carl T .. no.. but the money is in the accounts and is being charged as a cost to Broadband users. A job, given the money is in the system should be finished.
Of course always more can be done, but to the lay person reading what you are saying it comes across as if nothing is happening apart from in Suffolk. You did say the rest had no plan.
In my e-travels I don’t think there is any county/nation where I’ve not seen FTTP infill and can keep finding little bits of it – apart from Wales and probably Devon and Somerset but the later is waiting on Gigaclear build which is starting to appear live in areas.
using Wiki https://en.wikipedia.org/wiki/List_of_islands_of_the_British_Isles England has 19 inhabited islands, Scotland has 97 – I think that will add quite a bit to R100 which is quite a problem to be dismissed as “addressable”, of course it can be addressed but at a more expensive cost.
We have been quoted £80k by Bt (yes 80,000!!) to upgrade our tattered copper line to fibre. Despite the fact it’s already been partially installed to the road side near our building….. BT’s response was “well, somebody has to pay for it !”