Broadband News

Wales in negotiation for who pays to activate stranded broadband assets

Knowing what has and has been delivered via the BDUK projects is not always straightforward and it is taking the Welsh Assembly a number of months to determine with a high degree certaintity what has been delivered and which premises Openreach has invoiced for at a £300 each.

Julie James AM who looks after the digital infrastructure portfolio in Wales addressed the Welsh Plenary on 15th May and we recommend a read of the transcript where Julia James outlines three outstanding issues (jump to 17:45:38 in the transcript).

  1. Consideration for how Welsh Government can work with BT complete areas that have part build infrastructure e.g. fibre tubing coiled up on poles awaiting connecting together and fibre blowing and installation of splitters and manifolds. A decision on how to proceed is expected by end of May 2018.
  2. Work to go through all the invoices presented by BT in the last five years to clense and confirm the final premises count for those premises that gained access to 30 Mbps services via the projects intervention.
  3. Closure of the delivery phase means that the books can be balanced and there are several months work underway to ensure BT is not benefiting from any oversubsidy.

An very interesting point is the statement BT has delivered a significantly larger volume of premises than either party anticipated at the outset of the project, though no specific figures are given but this does have implications for what happens with the next phase which nominally has £80 million of funding available via three contracts. The nominal is because apparently around £62.5 million will go towards delivery of infrastructure via intervention, the other £17.5 million is being held back for use on bespoke community arrangements.

When the tenders for the next phase of work in Wales was announced there was in theory 88,000 premises split across three contracts covering North West Wales, East Wales and South West Wales. Our latest figures for Wales suggests 82,000 premises still only have access to a sub 30 Mbps option and this includes cabinets that only went live in April and scattered bits of FTTP that we are still spotting.

To outsiders it may seem simple to resolve the invoices against what has been delivered, but the complexities of some VDSL2 cabinets overlapping with Virgin Media cable access and what date the cabinet went live versus when cable broadband became available will make getting a 100% accurate figure difficult, add to this the fact that the number of premises is not static and over a five year period there will have been new premises added to cabinets originally delivered via the Superfast Cymru project. A major concern should also be what is the cost in terms of labour to re-check five years of invoicing and if as suggested some anamolies are only being made aware to the Welsh Government when they visit different areas paper based exercises may always have errors.

We're having a complicated conversation with BT around the connection of the stranded assets. There is a complex commercial—'negotiation' is the only word I can think of—going on about who should pay for them. So, BT have sunk an enormous amount of capital into the ground. They haven't got a penny from us for that, because they've over-bulked the programme. The conversation is: who should pay for the last bit of the connection? That's a complex commercial conversation that is ongoing, and as soon as we've reached the end of it, I'm more than happy to report it here in the Chamber. But I make no apology for the fact that, obviously, what I want to get out of it is the maximum number of premises for the least cost. So, quite clearly, where we're coming from is we want as many of those assets connected as possible for as little part of the gain share as is humanly possible.

Section 503: Julie James AM addressing Welsh Plenary

The situation particularly around FTTP (full fibre) which saw a number of areas drop back from in construction to no plans is the most difficult for residents and businesses affected, hence why we have included the word for word quote. Clearly Openreach and BT have expended labour and resources if there are part built networks, so the question around who pays to finish it, or whether (if BT win the next contract for an area) these are given priority in the next contracts are open for debate.

On the size of what has been delivered we believe there are 3,238 VDSL2 cabinets (note: additional cabinets for capacity upgrades are not in this figure) that have covered some 685,000 premises attributed to the Superfast Cymru project and superfast (30 Mbps and faster) broadband is available to 93.2% (638,000) of these. There is the additional FTTP footprint to consider where across Wales as a whole we have 62,200 premises on our books but some of that will be commercial and for the overall figure there is the question of the 6.8% overlap with cable broadband to be taken into account.

If one works to the assumption of £300 per premise and we ignore the cable overlap and that some of the FTTP is commercial we get a figure of £210 million, which is very close to the gap funding total of £205 million talked about back in 2012, there was some additional funding added part way through the contract if we recall correctly to add to this. We suspect the public funding though for FTTP areas is going to be higher than £300 but this may be balanced out by the removal of cable premises.

thinkbroadband analysis of Superfast, USC, USO and Full Fibre Broadband Coverage in Wales
with national rural/urban split and two selected 'county' areas
figures 16th May 2018
Area% fibre based
VDSL2 or
FTTP or
Cable
% superfast
30 Mbps or faster
% Ultrafast
100 Mbps or faster
%
Full Fibre
 FTTP
% Under 2 Mbps download% Below USO
10 Mbps download
1 Mbps upload
Under 15 Mbps download
Wales 97.4% 94.1% 33.2%

4.7% (all ISP)

4.5% (Openreach)

0.8% 4.0% 4.0%
Urban Wales (66% of premises) 98.7% 98.1% 37.2%

1.6%

1.2%

0%

1.3% 0.6%
Rural Wales (34% of premises) 94.1% 85.7% 13.0%

11.4%

2.3%

9.9% 10.7%
Urban Ceredigion (21% of premises) 96.9% 96.5% 1.9%

1.9%

0%

3.1% 3.5%
Rural Ceredigion (79% of premises) 86.5% 73.1% 21.6%

21.6%

4%

21% 20.6%

Urban Powys (13% of premises)

99.8% 98.3% 4.6% 4.6%

0.2%

0.4% 0.6%
Rural Powys (87% of premises) 88.9% 76.4% 19.6% 19.6%

4.6%

19.1% 18.9%

The above coverage figures are unusual in that we have split out the rural and urban areas for a couple of areas of Wales, this is to help illustrate that coverage is far from uniform and for those in rural areas missed out by the initial contract the rural area figures probably give a better idea of the situation. In terms of full fibre KCom with its City of Hull Lightstream converage is leading the local authority race at 76.9%, but in terms of purely Openreach FTTP Cornwall is the leader at 32.7%, followed by Anglesey 17.8%, Powys 17.6% and Ceredigion 17.4% but Powys is almost double the premise count of the other two areas. As someone is bound to raise it there are also the London Borough of Waltham Forest is in there at 20.4% and Spelthorne District Council at 27.4% which is part of Surrey, but Spelthorne is not a county like the others and Waltham Forest is purely commercial footprint, so as always when doing 'best of' lists how you define the list is important.

Back in 2012 Powys and Ceredigion both had no superfast coverage and are now in the top 10 council areas in the UK for full fibre coverage.

Comments

The way BDUK is supposed to work seems over complex to me. Surely it would be better simly to pay by results? To make sure installations are up to standard, payment would be be made only after the customer has enjoyed a trouble-free service for a year or so.

  • Pessimist
  • 3 months ago

Your suggestion would add an awful lot more complication, since you would have to survey or monitor every connection/customer.

BT is meant to only invoice for premises delivered and it does that on a quarterly basis, along with reports on what was delivered in a quarter. Seems either Welsh Assembly was just paying them and is only now making major checks OR given the complexities around infill cabinets it is wanting to recheck previously checked invoices.

While the BT interface was not great, the withdrawal from social media in the face of 'farce' posts did not help at all.

  • andrew
  • thinkbroadband staff
  • 3 months ago

Charging a unit cost £300 for passing 200 premises is £60,000 a cab. BT confirmed to the CMS select Committee Phase total of £26k.
The fact the actuals have not been reconciled and BT's capital contribution needs to be reported and the clawback suggests the level of FTTP can grow at a rate consistent with whatever resource BT can find.
It is not surprising the BT Capital Deferral is likely to continue to grow a bit more - £536m and counting. The increase in the volume of premise is not just overbuilding VM but a good deal of substitution of commercial investment.

  • ValueforMoney
  • 3 months ago

Re Andrew's comment: I don't think it would be a problem. All that is needed to get the £300 is for the ISP to show evidence that the customer has signed up and paid for the minimum qualifying period.

  • Pessimist
  • 3 months ago

£60,000 for a cabinet where 200 premises can get superfast speeds.
Just £7,500 for a cabinet where 25 premises can get superfast speeds.

Averages do not mean every cabinet is of the average size.

If BT start delivering FTTP off their back now in Wales they CANNOT invoice for it, since there is no agreement in place for this sort of - automatically keep on spending until you have spent every single penny.

  • andrew
  • thinkbroadband staff
  • 3 months ago

But that is a very different payment system to now, so with takeup running at 42.5% a lot less money would be forthcoming and if this was on the table back in 2012 I would bet either (a) BT would not bid or (b) the price they would negotiate per premise would be much higher.

The gainshare mechanism takes into account take-up and is reported to projects and we have our own estimates to ensure that this is not massively under reported.

  • andrew
  • thinkbroadband staff
  • 3 months ago

Not to forget that a significant number of premises have been enabled through FTTP, so simply dividing the total amount by the number of cabinets to get and average is simply wrong. FTTP is disproportionately expensive.

Also, the cost to enable a cabinet is going to depend a lot on its location, remoteness from fibre links, power delivery costs and so on, all of which vary a lot.

Also, is that £300 per premises before or after any savings or is it just based on the public funds in the original contract, not all of which have been spent (not counting gainshare - a different issue).

  • TheEulerID
  • 3 months ago

Not to forget that a significant number of premises have been enabled through FTTP, so simply dividing the total amount by the number of cabinets to get and average is simply wrong. FTTP is disproportionately expensive.

Also, the cost to enable a cabinet is going to depend a lot on its location, remoteness from fibre links, power delivery costs and so on, all of which vary a lot.

Also, is that £300 per premises before or after any savings or is it just based on the public funds in the original contract, not all of which have been spent (not counting gainshare - a different issue).

  • TheEulerID
  • 3 months ago

The £300 comes from the Plenary session i.e. "To date, we have paid BT £300 for every premises, however, the closure of the delivery phase means that we must now work together to balance the books and ensure that all expenditure is eligible and fully evidenced. This process will take several months to conclude, but it is essential to ensure that BT does not benefit from any oversubsidy."

  • andrew
  • thinkbroadband staff
  • 3 months ago

TheEulerID - its 700,000 premises passed with c3,000 cabs. Andrew's last response shows the £300 payment per premise, this was also referenced by Audit Wales. The reconciliation is against actuals invoices not unit costs, thankfully and the BT capital contribution (c700k x x£70 needs to be evidenced) A least the clawback £80m referenced is a function of the full capital deferral not the £130m places likes Sommerset and Devon are still quoting.

  • ValueforMoney
  • 3 months ago

TheEulerID - its 700,000 premises passed with c3,000 cabs. Andrew's last response shows the £300 payment per premise, this was also referenced by Audit Wales. The reconciliation is against actuals invoices not unit costs, thankfully and the BT capital contribution (c700k x x£70 needs to be evidenced) A least the clawback £80m referenced is a function of the full capital deferral not the £130m places likes Sommerset and Devon are still quoting.

  • ValueforMoney
  • 3 months ago

I'm not a big BT lover but let's face it that a large area of rural wales lags far behind the rest of wales and the UK thanks to Julie James and her labour cronies. Whenever you write to her, she's like a broken record highlighting how many homes have benefited from super fast cymru, do I care no as I'm not one of them. This exercise is going to slow down the next phase of installs, so Welsh Government get a grip and award the contract to BT as no-one else wants the work!

  • withoutfibre
  • 3 months ago

@VFM

3,238 cabs was what was listed, but you seem to be deliberately ignoring my point that a significant number have been enabled through FTTP which means that your simplistic calculation on the cost per cab doesn't work if the number of those is material. Andrew mentions 62,200 FTTP premises in Wales in total, but what we don't know is how many are BDUK. What we do know is that the cost per premises on FTTP is a lot higher (on some bids in remote areas gap funding in the region of £1.5k on some contracts). 50k FTTP premises at that rate would be £75m.

  • TheEulerID
  • 3 months ago

At the end of the day the 'rough' figures based on the £300 mentioned and numbers we believe delivered show that things appear in line with expectations.

What is the cost of several months work for a few accountants?

  • andrew
  • thinkbroadband staff
  • 3 months ago

The EulerID I am not ignoring it, it remains too small. Ignoring it causes more problems for BT. It is allowable costs. 3,238 cabs/paths cost £100m actual being generous, then take off BT's contribution. ON FTTP, unless the planning is outragous they will be struggling to find £1k a premise before BT's contribution. There is no final drop, just the connectorised joint on the pole.
All other projects had 'efficiencies' and this one of the few where inflated unit costs were applied rather than actuals - hiding the badly named 'efficiencies'.

  • ValueforMoney
  • 3 months ago

TheEulerID 'unit costs' disappeared for a reason from the BDUK Framework. They bore no relationship to the component costs. The BDUK Framework was sufficiently inflated that not only were all projects report 'savings' but BT is beginning to write cheques to LA'a as reported in the Q4 statements. The 'unit costs' observed will need a large reconciliation payment. If dealt with earlier more work could have been planned.

  • ValueforMoney
  • 3 months ago

FTTP Planning Outrageous? And there was me and others thinking that the costs were down to the distance the fibre had to run and time it takes to do it. So how much to get FTTP to LL28 5RD?

Some cabs no doubt cost what you suggest, did they all? Unlikely, since for some those very expensive planners would have more to do Shirley.

At the end of the day for all your wasted key presses, the savings are appearing and therefore why current DCMS hopes are 97 to 98% superfast once everything has run its course.

  • andrew
  • thinkbroadband staff
  • 3 months ago

Andrew; You just need to read BT's three submissions to CMS Select Committee (2006) to confirm average cab and fibre path costs.
97-98% came after the consultations of the B-USO talking of 95% coverage. DCMS and Ofcom needed to be challenged.
The savings in Wales have yet to acknowledged.
If fully explored and this applies to all the unreported 'true ups', then the funds would exist to complete the in-fill work.
The benefit to Openreach and UK Enginerring is to be prepare for full transition to fibre.

  • ValueforMoney
  • 3 months ago

Andrew - sorry 2016 submissions. You could also read the WPQ (Hancock to Timms) correcting Ed Vazieys evidence to the same inquiry where he thought BT had made a capital contribution at the time of the inquiry. It is on the public record.

  • ValueforMoney
  • 3 months ago

Well done for not answering the question and trotting out your usual references.

So to repeat...
FTTP Planning Outrageous? And there was me and others thinking that the costs were down to the distance the fibre had to run and time it takes to do it. So how much to get FTTP to LL28 5RD?

  • andrew
  • thinkbroadband staff
  • 3 months ago

Andrew - references are a matter of public record. They should be used to inform your articles on matters of coverage and potential coverage. FTTP planning reference is from CDS public comments.

I will look at LL28 5RD but start with the £2-£3 a metre where overhead infrastructure is re-usable plus a distance based connection charge. How many live in that community?

  • ValueforMoney
  • 3 months ago

Are so are you saying I am not informed on matters of coverage and potential coverage.

As for how many live there, was not aware of anyone doing costs per premise, surely the question is how many premises and as you proclaim expertise since you dispute everything it should be easy enough to determine this without my input.

  • andrew
  • thinkbroadband staff
  • 3 months ago

Hi Broadband Watchers.
To cost the fibre route you have to know the direction it is coming from a good example is Cab 7 at Wormley all the customers have acesss to FTTP this is not feed back to the Exchange and covers a large area.

  • Blackmamba
  • 3 months ago

Well the four premises in GU8 5LD are 1km from nearest other postcode, so even then its £2000 to £3000 split 4 ways as a minimum, i.e. FTTP costs soon add up

  • andrew
  • thinkbroadband staff
  • 3 months ago

Hi Broadband Watchers.
I think you will find that the Post Code Gu85LD is on a private land and the distance for fibre is .5 of a mile to the road which is park lane where the fibre runs. I am surprised that the estate has not provided ducting to this area to cover the four units there even may be a grant for this work. I think this covers my other remarks.

  • Blackmamba
  • 3 months ago

First thing to remember is a straight arithmetic average can be rendered relatively meaningless if there is a skewed or irregular distribution.

And secondly it has been reported here (https://www.theregister.co.uk/2018/05/15/nbn_q3_2018/) that the NBN average cost of FTTP brownfield is currently running at 4,396A$/prem which at £1=A$1.8 is still a very big number.

  • Gadget
  • 3 months ago

actually wormley 7 was deployed as FTTC under BDUK phase 1 think these are additional premises off that cab (that have been covered under a small pon that did not get benefit under the initial BDUK enabling of PCP 7 back in 2014 or so so more misinformation / disimformation as ever

  • fastman
  • 3 months ago

Hi Andrew Staff.
Please can you check the data on Gu85LD Post Code one tile shows fibre in Black.

  • Blackmamba
  • 3 months ago

@Blackmamba and the problem is what for GU8 5LD data says FTTP available, checkers say it is too

  • andrew
  • thinkbroadband staff
  • 3 months ago

@fastman sample postcode GU8 5UU pretty close to PCP cabinet and no VDSL2 options show up just the FTTP

Usually if FTTP comes to a subset of a VDSL2 cabinet the VDSL2 service remains as an option

The Surrey project did deploy FTTP under its phase 1 roll-out it was not all VDSL2 cabinets.

  • andrew
  • thinkbroadband staff
  • 3 months ago

think that cab was as FTTP but done in stages over a period on years I think

  • fastman
  • 3 months ago

wormley 7 massive area would have been horribly expensive (lucky to have been built as probably well outside the BDUk (maybe why done in stages - will have been many km of fibre to be run as well

  • fastman
  • 3 months ago

Hi Fastman. This Cab 7 was done in one hit with fibre and the customers on long lines payed if the fibre was down long drives by contractors. I will check in the next few and see if they have provided a FTTC at the moment I think there is not one. Most of the fibre is overhead so much cheaper than providing a Cab. They only need to cross the A3 with fibre this would open up more customers that have slow speeds.

  • Blackmamba
  • 3 months ago

blackmamba it has been done in stages over 2 years and big hit at beginning and a mumber of smaller ones later - there is no FTTC in this cab area -- re you comment on A3 are you offering to fund that them personally ? sure openreach would love to talk to you about that (because some one will need to fund it)

  • fastman
  • 3 months ago

Hi Fast. Thanks for that information save me a trip to Wormley CAB 7 as for the customers on the other side of the A3 they only will require a bespoke CAB off Cab 4 in Elstead area to cover 80-100 Addresses.

  • Blackmamba
  • 3 months ago

Hi Broadband Watchers and Fast.
After checking Post Code areas GU84,85, the results are not up to date ever results close to the new Cabs showing incorrect data.

  • Blackmamba
  • 3 months ago

@Blackmamba There is not any results for GU84 and GU85 postcodes or speed availability data because those postcode areas do not exist in the postcode dataset.

If they are totally new postcodes they will be added once the ONS dataset is released and subsequently processed.

  • andrew
  • thinkbroadband staff
  • 3 months ago

Hi Andrews Staff.
I think when the new ONS datasets are released you may find that the ((genarial area coverage )will improve as it will cross borders as you push down on the 10 meg target and this will remove many orange post codes.

  • Blackmamba
  • 3 months ago

Post a comment

Login Register