The Oxera report on
the UK’s broadband state aid measure was published on May 19th. Due in January, it was hoped it would build upon the work of
the National Audit Office (NAO) and the Public Accounts Committee, where we learned in January 2015 BT had
inflated its 2012 cost models by 38%. We also learned the total average cost per
cabinet and fibre path was less than £25,000 before BT’s capital contribution
on the first 8,500 cabinets installed.
This contrasts with subsidies of more than £40,000 implicit within the
£1.2bn phase 1 BDUK contracts. This was
good progress and showed the importance of public scrutiny and how the NAO
could aid government departments in mitigating ex-privatised utilities abusing
their dominant position. The Oxera
report might shed light on the status of BT’s offer to match public subsidies with
£1bn of its own capital promised very publicly and repeated in Parliament.
The Oxera report is a
pre-cursor to EU re-issuing the state aid notification in July 2015 to approve
continuance of the state aid scheme so BT can be paid subsidies through to 2018.
Oxera does not mention
BT’s offer of £1bn matched funding. It does not mention any amounts paid by BT
so far but does make a reference to BT’s matched investment being less than
expected. But critically we get a hint in clause 3.83 on the bottom of page 44.
The Oxera report
provides one new crucial fact but fudges many others. The fudges include Oxera reporting on the
£780m central government funding, not the
total public subsidies of £1.7bn.
The impact of the latter is to understate what Oxera describe as BT bids
for cabinets. They reference a number of
circa £32k each. Using the £780m rather the £1.7bn implies some level of alignment within the
realms of a working contingency but this is misleading. This oversight should be corrected and the
report re-published with the full amounts,
but this rather large oversight can be forgiven if the new piece of
information they provide can be acted upon.
The key new learning, building upon the NAO work which focused on
the excess modelled costs, is found in
clause 3.83, page 44 of the report. It
informs us that BT makes no capital contribution to the project until three
years after each contract signature. The BT offer of £1bn matched funding to Parliament
has been reduced to no more than an
uncertain reconciliation process three years after the contract begins. BT is free to re-balance costs, self-certify
its own capital and add back any un-allowed costs presented and rejected for
payment previously. This new information
explains why so many rural communities are being excluded from the planning
This means the BT promised
capital contribution cannot be used in planning the fibre rollout. The whole Broadband project in its initial
stages is 100% state funded with assets gifted to BT. This is at the odds with the principal of gap
funding. How can this be compatible with
the principals of state aid where the subsidy should be the minimum needed and
does not substitute for private investment?
It is important to
note that the nature of the capital subsidy is controversial, as the upgrades are largely improvements in
BT’s passive infrastructure in the form of ducts, sub-ducts, fibre, the
cabinets, re-arranging copper. The
nature of the work is best considered as an allocation of BT’s existing
resources to work on one section of the network while delaying other work. It is important to note that BT’s commercial
investment in fibre access came from within existing capital envelopes, thus any state aid would act as a direct
substitute for BT’s own resources,
should BT’s contribution not be forthcoming when the work is occurring. In these circumstances it is quite easy to be
in breach of the principals of state aid.
The Oxera report makes
a virtue of the protections that the state cannot pay more than 100% of the
capital cost. This is certainly better
than paying standard unit costs unrelated which are unrelated to the underlying
costs as evidenced in some of the early contracts, but the switch to actual costs ignores the
gap funding principle and forgets BT assurance to Parliament of £1bn of matched
funding. The latter was never real but
the principle of BT paying a fair proportion of the capital costs is captured
in the intervention rates which suggests BT should be paying about 30% of the
capital costs. Can or should these
payments be brought forward?
Should the EU
Commission ignore or try a fix?
The focus for the last
2.5 years has been on putting the BT
abuses aside and getting on with the job of installing some 200 street cabinets offering improved connectivity to some 40,000 premises a week, should
customers want to order an upgrade. Just
as the NAO and BDUK have progressed the verification of actual costs, it seems appropriate that the EU Competition
Commission could amend the state aid measure to enforce the concept of gap
funding and thus extracting from BT the monies promised. But what form would this take? One example is outlined below.
The USA FCC rural
broadband scheme has this notion that the operator pays the first $70
cost per premise passed for rural broadband projects.
For the UK a
couple of things need to happen.
We need to stop
believing BT spent anything like the £2.5bn capital they claim to have spent on
the commercial rollout of circa 50,000 VDSL cabinets passing 19m premises. The NAO efforts to help BDUK to get to actual
costs suggest BT costs on the access network of cabinets, fibre, handover points is not more than £20k
a cabinet and fibre path. This would put
BT’s commercial investment in the access
network at a total of £1bn. BT may have
spent some more on Next Generation Access for content distribution platforms, wholesale
billing and provisioning systems but upgrading the exchange side of the network
with optical fibre paths has been quicker and cheaper than imagined. For simplicity this is about £50 per premise
passed and so BT should not be expected to invest more per premise to upgrade premises
in economically more challenging areas.
Thus, any state aid measure should translate into an
unambiguous BT contribution of c £50 pounds per premise passed. For the first BDUK funded 12,500 cabinets passing 2.5m premises to
April it would mean BT paying
£10,000 per cabinet, while the state
pays the remaining £14,000 each.
This allows a greater amount to be achieved with subsidies rising as the
geographies become more difficult to serve and cabinets serving fewer people. Currently, the state pays everything and
hopes to collect some money at the end.
This is likely to occur at the expense of those the programme was originally
intended to meet.
For the 5m premises in
the BDUK scheme this is about £250m contribution
from BT rather than the £1bn promised to Parliament, but it is a real contribution
as opposed to hot air. It is consistent with sums mentioned in the first
NAO report in September 2013. It is also
consistent with modelling work done to support similar subsidy schemes in the
The UK rural economy is now reliant on EU Competition Commission
spotting clause 3.88 of this report and insisting upon the necessary
changes. If it does, it finishes what will be three years of work. This was begun by the Public Accounts
Committee and the National Audit Office supporting BDUK’s efforts to force through actual costs rather
than tolerating standard unit charges unrelated to the underlying cost. The latter are evident in reports to North Yorkshire County
Council where a charge of £179 per premise passed was reported. The amendment to the state aid measure should
convert the £1bn promised to the UK Parliament and to the UK Government into
something directly tangible and usable.
£250m is much less than the £1bn promised, but it would be real and it
would get spent where it was intended to be spent.
Finally, if the cheapness is
understood, then UK policy makers would
not hesitate in calling for a more ambitious connectivity plan for all UK
businesses. It would also allow the
£1.7bn subsidies to be spent where intended with BT the asset owner making a
fair and just contribution. As we speak
such a contribution is not secure and thus many villages and hamlets are being
excluded from the opportunity of getting an upgrade.
To recap, the combined efforts of
the NAO and the BDUK value for money team was to secure actual cost data from
BT, so the state pays no more than 100%
of BT’s capital costs. This is a bigger
achievement than one might first consider,
as it reduced average per premise payments from a possible £200 per premise passed (£1.2bn contracted /6m premises ) to what should be a working total average cost of £125 for the first 2.5 million
homes. The next step is to extract from
BT a fair capital contribution which would be for indicative purposes would be
£50 a premise for the estimated 6m premises in the programme. No capital contribution has been paid yet and
it is unlikely anything will be paid unless the EU Competition Commission act or
the UK Parliament demands BT keeps its promises to Parliament by demanding
specific changes to its undertakings.
will know by July, but there is nothing
in BT’s annual accounts to suggest it will make any capital contribution to the
44 rural UK phase 1 projects.