Skip Navigation

Reasonable progress in Wales says Wales Audit Office
Thursday 28 May 2015 10:40:28 by Andrew Ferguson

While the Wales Audit Office has not given a simple A to F grade to the Superfast Cymru project and BT, but the headline of reasonable progress is hardly a resounding endorsement.

"The report also found that:

  • there were initial weaknesses in the programme and project management but the Welsh Government now has clear and appropriate arrangements in place;
  • some local authorities, businesses and residents have not been satisfied with the communication about the Superfast Cymru rollout;
  • take-up of next generation broadband is rising, but there are no take-up targets in place;
  • the Superfast Cymru contract has already achieved most of the expected direct benefits related to jobs, apprenticeships and work experience opportunities; and
  • although arrangements to support and measure the wider benefits from the public investment in digital infrastructure have been weak, the Welsh Government is now developing a national project for the exploitation of next generation broadband by businesses and a plan for public sector exploitation.

The report makes a number of recommendations, which include:

  • improving communication about the local rollout of Superfast Cymru;
  • ensuring that 40 per cent of premises in the intervention area are capable of receiving download speeds of 100 Mbps or more; and
  • monitoring and supporting the take-up of next generation broadband; and improving the delivery of the full benefits of the public investment."
Summary from WAO report

The reasonable progress is of course no comfort to those who are yet to see any progress towards better broadband being available, but there is still time for significant improvements, e.g. Conwy before Christmas was devoid of superfast broadband but now has 55% of premises with access to speeds of 30 Mbps and faster.

Overall Wales has 74.9% coverage at superfast speeds (30 Mbps or faster and based on thinkbroadband analysis as of 26th May 2015), and FTTP coverage has improved some way since the WAO report was written rising from 325 to what we believe is 1,300 premises now. Of course the changes of this increasing to 40% of the Welsh intervention area looks impossible, so the 100 Mbps available to 40% is either going to be down to vectoring or some early deployment of cabinet based in 2016 or a fudge due to poor wording in the contract allowing Fibre on Demand to qualify.

30 Mbps and faster broadband coverage in Wales
Click image for full size version.

The often talked about 96% target in Wales is the level of fibre based coverage without any speed qualifier, the contract calls for superfast to 90% of premises and the Wales Audit Office is using the lower 24 Mbps definition.

FTTP coverage in Wales May 2015
Click image for full size version.

The spread of native Openreach FTTP across Wales is limited so far, but much more is believed to be on the way, resources rather than will power are probably the limiting factor there. So while it is entirely possible Openreach may miss its end of 2016 deadline, with 18 months to go it does still look very possible. The problem is that even if Openreach/BT meet all the contract goals, there will still be the 4% with no access to anything fibre based and 10% getting speeds below 24 Mbps, so the complaints may be less in volume, but probably more vocal as those left out will feel even more abandoned.

When reviewing the problem of Wales only having one bidder, there was three other bidders at one time, but worries over the amount of funding they would have to put into the pot due to the £205 million cap on public money, concerns that Wales was looking for the lowest cost bid and too high risk levels due to size and complexity meant that BT remained the last human standing.


Posted by TheEulerID about 1 year ago
My bet is that the 100mbps target could be met by offering VDSL bonding as a GEA product. It would cost the consumer quite a bit (as there's an SLU wholesale cost for the second pair), but I would have thought it's technically viable.
Most houses are already served with a second "spare" pair; certainly way over 40%, and as not many would take it up it might well meet the letter of the contract.
Posted by ValueforMoney about 1 year ago
They have confirmed a premises passed payment £81,000,000/345,000 *.9 = £211. At 200 premises per cab, they are colecting £42k per cab, £10k above the Oxera bid prices and £20k above the costs identified by the NAO.
How can you hope to get service in the final 10% if your handing over double in subsidies?
Posted by ValueforMoney about 1 year ago
The report makes clear they have no way of enforcing the 100Mbps and BT have no resource to do the work, but the costs are bundled in the payments passed per premise.
Posted by TheEulerID about 1 year ago
If there's no way of enforcing it, then it's not a contractual requirement, merely an aspiration. Unfortunately, we don't have the detailed wording of the contract.
Of course there will still be a (political) price to pay if that 100mbps to 40% can't be met. Hence the fudge.
nb. a bonded VDSL GEA product is not a bad idea. It would suit many businesses
Posted by WWWombat about 1 year ago
Make naive calculations of a "per cabinet" cost, get naive answers. Are your simple per-cab costs calculated the same way theirs are?

For example, one way for SFNY figures comes up with £47k per cabinet. A different way comes up with £37k per cabinet. Because neither take account of the fact that 40% of the budget is not spent on deploying fibre or cabinets.

Elsewhere, I've seen costs of £18k per cabinet, £38k per fibre spine, and £300k per head-end.

Vastly different figures per cabinet. All valid, but they're apples and pears, and can't easily be compared.
Posted by TheEulerID about 1 year ago
One thing that @VFM always neglects is these are 7 year contracts. In other words the public money goes to subsidising both the capital and operational costs during that 7 year period after, of course, netting off the revenue. The "BT contribution" equally covers that combination. Per cab costs are even less meaningful given that is far from the only element.
Posted by ValueforMoney about 1 year ago
@Euler I think the reconciliation is three rather than 7 years for capital, but at least your begining to acknowledge the state pays everything during that time. This is step forward. It is also a mockery of the state aid condition on gap funding, if not a breach of the state aid condition.
Posted by ValueforMoney about 1 year ago
@WWWombat indeed but in most cases the components are the same and I hope your not questioning the NAO report in January.
Posted by TheEulerID about 1 year ago

The three year point is the (usual) point when the claw-back element will first be reviewed. And, no, the state does not "pay for everything". It pays whatever was in the framework agreement formula for that particular contract. That is one of those elements which is covered by commercial confidentiality clauses as it contains BT's internal costs.
Whilst claw-back is often quoted as being just on take-up rate, it's actually more complex as revenue from other uses of the BDUK infrastructure is taken into account (like use in leased circuits).
Posted by WWWombat about 1 year ago
I question everything that doesn't make sense.
Posted by ValueforMoney about 1 year ago
EulerID - bT's internal costs can be anything they wish.
So everything in this case is where subsidies of £40,000 can be gleaned for cabinets identified in other parts of rural Britain at c£24k.
It can be seen in total volume of state aid reported in BT's accounts.
I support fully BT doing the work but the current arrangement is in breach of the state aid concept of gap funding as BT inflated its costs models as per NAO report.
Posted by ValueforMoney about 1 year ago
Then ask how is there any room for BT's capital contribution given the scale of state aid reported in BT's accounts, relative to what's been delivered and relative to what BT did commercially?
How can this be state aid complaint when there is not even a reference to BT making a capital contribution in the notes of the accounts?
Posted by TheEulerID about 1 year ago

BT's internal costs can't be "anything they wish". All costs incurred, including allocated internal manpower are subject to audit and cross-comparison by the BDUK and the auditing bodies. Further, these figures are benchmarked by consultants employed by the NAO. It's not just an arbitrary figure plucked out of the air.
Posted by Llety about 1 year ago
As the talk about accounting and value for money continues, teenagers whose lives and friends are online find they can't exist and move out of rural communities, houses can't be sold and communities age and disintegrate. The Audit Office did a terrible job, they ignored the human cost and the impact on wider communities of continued delay and uncertainty. There is a huge social and financial value to the community cohesion that exists in rural areas (such as rural Wales), don't underestimate the value it has and the impact of poor infrastructure (not just broadband) has on destroying it.
Posted by ValueforMoney about 1 year ago
@The Euler The allowable costs are audited, BT's contribution is self certified.

@Llety I concur.
Posted by TheEulerID about 1 year ago

The degree of BT contribution can be assessed by the auditing authorities based on the difference between submitted costs and what the public authorities provide. Of course it's all subject to judgement, but not of the auditing reports have raised this as a concern, or have I missed something?
Posted by ValueforMoney about 1 year ago
@EulerID -I think you have. NAO confirm the £24k is all costs - even the 20p in diesel receipts and the chargies for PMO, core, provisioning, hop, spines. (based on the first 8,500)

BT's contribution might eventually be a proportion of this, or it could be as you other self certified costs that BT may wish to attribute.

The situation in Wales is worse....
Posted by ValueforMoney about 1 year ago
@EulerID - The standard per premise passed charge arises directly from the excess cost modelling of 38% identified by the NAO.

Hence the Welsh are paying this £40,000 + subsidy, as I think (but not certain) are the other 7 contracts not in BDUK milestone to cash process.

This is denying/delaying rural areas the services they should be getting. There is absolutely no need for this.
Posted by WWWombat about 1 year ago
That's precisely the question I keep asking you. I look, but reach a different conclusion.

Hard to do this in 600 chars, so I put a post in the forum:

Summary: I show 5 sniff tests to determine whether BT's capital expenditure makes sense in light of the state aid figures.

I think they do make sense, but please take a look.
Posted by WWWombat about 1 year ago
Where do you get the "NAO confirm the £24k is all costs - even the 20p in diesel receipts and the chargies for PMO, core, provisioning, hop, spines. (based on the first 8,500)"?

All I see the NAO report is an average of £21k, which didn't include PMO or core.
Posted by TheEulerID about 1 year ago

That looks like a very useful analysis. As you point out, the capex is not apportioned to what are or are not public supported projects, so you have to infer what you can from the accounts.

I think the point I'd make is that the NAO has looked at this, used consultants and, as far as I'm aware, have not raised any concerns over the level of BT's contributions. They have far more access to actual figures than we do.
Posted by ValueforMoney about 1 year ago
You should get to £24k further in the document, the latter comment was made to me when checking the detail so it's an anecdote. It is worth cross referencing with Oxera, although these exclude the first 8 contracts which if like Wales are stuck on the unit costs directly related to the 38% excess modelled costs.
Posted by ValueforMoney about 1 year ago
Euler I wrung the NAO and they do not wished to be quoted but the notion of BT contributions is odd as Oxera's primary claim was the process ensured BDUK paid no more than 100%. Unfortunately this cannot be the case in Wales because of the per premise cost.
Posted by WWWombat about 1 year ago
I know you think Oxera's report, section 3.83 means that BT don't pay capital for 3 years. I think it means something different, and solves the problem of your 2 posts...

What I think 3.83 means is that, for projects using estimated unit costs, it ensures excess costs (beyond the actual cost) are repaid by BT at the end of the deployment phase (Max 3 years; likely average 1.25 years). That's why Oxera claim no more than 100% ... but it is only correct after a true-up reconciliation.
Posted by ValueforMoney about 1 year ago
WWWwombat I wish you were correct. Project were priced using models with 38% excess costs meaning the allowable costs to be billed against were already inflated above any contribution BT would make, hence BT contributions would never be used to say plan a bigger rollout. It's ugly.
Posted by ValueforMoney about 1 year ago
@LLety What I do not understand about Audit Wales is why they could not simply say c1700 cabinets installed, passing 345k premises and the £81m paid means our total cost for these easy areas is twice being paid in other BDUK projects and then ask the question why?
Posted by AndyCZ about 1 year ago
VFM - Seriously, at least try to post correction information.

The models were not priced with 38% excess costs - the reported capital spend in phase 1 was 38% less than the 'estimated' cost and this does not include work in progress/not invoiced. There is a massive difference between what's reported and what you're saying (even mathematically, you're wrong!).

BT will be investing further in BDUK areas, including deploying vectoring in the areas that will benefit most. None of these costs have been published yet as the capital investments have not been made.
Posted by ValueforMoney about 1 year ago
@AndyCZ really serious clause 3.7 BDUK’s analysis of the cost information it has received suggests that so far actual phase 1 costsare significantly lower than BT’s financial model. As at September 2014, BT’s total reported capital spend on phase 1 of the Programme was £142 million (38%) under the estimated price, including work in progress not yet invoiced.
Posted by AndyCZ about 1 year ago
Come on, this is basic GCSE mathematics. 38% excess costs does not equal 38% under an estimated price.

Aside from the maths failure, you're using numbers incorrectly to suit your claims - it very clearly states that work is not yet invoiced, yet you claim "project were priced using models with 38% excess costs".
Posted by ValueforMoney about 1 year ago
AndyCZ - please that's clause 3.7 cut and pasted. Wales are paying based on a cost per premise passed. Divide £81m/ 345k premises and approx 1700 cabinets and tell me what you get. It is that bad. I wish it was not.
Posted by fox-uk about 1 year ago
It's dinosaur politicians trying to comprehend 21st century internet.

Most have no comprehension of what the internet is let alone how fast it should be or how much it should cost.

They have failed to achieve and have tried to cover that simple fact with the 'Reasonable Progress' twaddle.

Too many PPEs from Oxbridge and too few with the nouse to smell a rat (even if one does bite them in the bum).

I am ashamed to be Welsh born and bred when I see such idiocy.
You must be logged in to post comments. Click here to login.