TAP INTO THE UK’S BROADBAND MARKET
TRANSCRIPT OF SEMINAR HELD ON 18 SEPTEMBER 2001
INTRODUCTION BY CHAIRMAN - DAVE TOMAN, RA
Right everyone, can we make a reasonably prompt start. We have got a pretty
packed programme and it is going to be an effort to get everything done and
dusted by lunchtime. Good morning and a warm welcome on a chilly day to the
Westbury Hotel. A particularly warm welcome to our American colleagues who
have obviously braved much more to be here this morning. This is the Agency's,
with the support of KPMG Consulting, BFWA seminar, and what we intend to do
this morning is to try and give you as much information as we can about the
forthcoming award of 28GHz licenselicences. To give you an opportunity as well
to raise any queries that you might have about the processes and the procedures
and anything else that you might want to raise.
My name is Dave Toman. I head the Unit in the Agency responsible for
Broadband, Broadcasting and Programme Making, and it is my privilege to chair
the event this morning. Hopefully, I will try and keep us to time in what is a very
full and congested programme. First of all, can we run through some house
keeping procedures which are particularly important at this time - a continuous
ring, if you hear a continuous ring, then can you all make your way to the exit
through one of three doors, there, that corner, and this corner. Please don't use
the exit in that corner as it restricts movement for others. When you leave the
building can you assemble outside Russell and Bromley’s on the corner of Bond
Street and if it is an alarm of a serious nature can you report to me so I can tick
you off the list of attendees. Mobile phones, can you please switch them off.
More important than anything, coffee and lunch will be served in the same room
as you registered this morning. We are also going to record this event and put a
transcript of the proceedings on the website, You may want to bear that in mind
when you ask questions. Obviously we will want to keep the proceedings or at
least your participation as confidential as you would like. Question and answers
will be governed by the usual rules. We will have a radio mic: please don't talk till
you receive the radio mic, and when you do receive the mic can you give an
indication of who you are. You should all have a package that comprises the
Iinformation Mmemorandum for the auction, - a very important document that you
should read avidly, - publicity packs, and also copies of the presentations that are
being made this morning. If there are any problems please contact one of the
agency staff, there are quite a few of us buzzing around.
Right, on to the main programme. Can I point you to the agenda: we have made
a couple of minor changes to the agenda. Hazel Canter, the Agency’s Director of
Spectrum Services, will introduce the proceedings and she will be followed up by
Martin Heath, a partner in KPMG Consulting and Martin will give us a flavour of
the market opportunity at 28GHz. Then we will break forAfter coffee at about ten-
fifty and I would be very grateful if we could reassemble at five past eleven on the
dot when Steve Talbot, who is head of Broadband Engineering at the Agency,
will give us a flavour of some Broadband technology and spectrum management
issues. Finally, Joe Sonke, head of the Broadband team, will give us the latest
information on development in the 28GHz auction and attendant licensing,
processes and procedures - a very important presentation for you all. We have
extended the timetable for questions and answers and lunch is now at 12.45.
Now, these occasions are also opportunities for you to talk to like-minded people
and also to ask questions of as an Agency and wider questions that you might
have. We have Steve Jones in the audience who is leading the licence award
procedures for public fixed wireless access at 3.4 and 10 GHz, and he will be
avaliableavailable to talk to anyone about that., and I understand that we have
one or two colleagues from mainstream Department of Trade and Industry who
will be able to answer any queries that you may have on general communications
policy. Without further ado can I welcome Hazel to kick off the proceedings.
OPENING ADDRESS - HAZEL CANTER, RA
Morning everybody. Some familiar faces and quite a few not familiar, and it is
very good that there is such a good turn out, braving the threat of terrorist
activity, weather, central London travel, all sorts of things. Anyway, I am really
glad that you are here and I hope that you have a useful and productive day. and
Pplease make sure that you collar all the staff and ask any questions that you
want to know about the process that we are going through. It is a really posh
area around here isn't it - one of the reasons I am afraid that I can't stay for the
rest of the morning is that if I stay for lunch I might feel the need to go shopping
afterwards. The only consolation in that is I found my downstairs neighbour
standing outside Tiffany & Co - he has got a part-time job there and I walked
away thinking discount, discount… but never mind.
This seminar is a key part of our preparations for the award of the 26 licences
that remained unsold at the end of last November's auction. At the end of the
auction and after the process had been gone through a lot of people phoned us
up and said “'if only we had known”' and you think “N, no”. However, we had
decided that as a result of that, there probably was some latent demand for the
unsold licencses and certainly it is the case that there is quite a lot of interest in
bBroadband rollout. I don't know about any of you but you have to be a more
patient person than me to cope with using the Internet at 56 kilobytes a second
or so. I just get so frustrated, I go away and do something else. I do think that
there is a need out there, and it is a matter of making the means available to
achieve that. Today is really the culmination of our attempt to try harder to make
sure that as many interested companies as possible are aware of the opportunity
that the licenselicences present and I am really glad to see that there are many
of you here. It is also the starting point of the award process - the hefty
Information mMemorandum has all the information potential bidders might need
and we will soon be making regulations that will allow us to begin the award
process in mid- October. It has effectively become the second stage of the
process we began last November. As I was trying to explain to someone
yesterday, in a way what we are trying to do is move from making the spectrum
available on a sort of saleroom fail room basis to running a shop really. You can
go in and get some when you want it, should stocks last and all of those normal
things that you see in shops. But it is a matter of trying to create a situation
where people can pick up spectrum when they want it rather than at the time that
we decide it is appropriate to sell it or auction it. We believe that the
licenselicences provide a good opportunity for companies to enter the Broadband
access market, and that there is a high level of demand, so we want to give you
a chance to develop the market and provide services. -- I can hear someone
opening, open it, do come on! It is very noisy! -- So we are hopeful that the award
process will enable you to respond to both changes in market conditions and
your own company’s interest over the next year. All the presentations will go in
to more depth on these points and give you an opportunity to raise any questions
that you might have. Anyway, welcome and lets get going shall we?
THE MARKET OPPORTUNITY FOR BFWA - MARTIN HEATH, KPMG
Alan Horn from Interconnect Communications based in Wales, sflaw BTony Ballard of Field Fisher
Waterhouse. If I may, James Bird from the consultancy ReguTel. You gave us this promise that
technology presentation, a tall view of the market. But where Rob House, freelance
I am Martin Heath and a partner with KPMG. I run our communications and
content practice and we have been working with the RA over a considerable
period advising on some of the issues surrounding not just Broadband but
around the Broadband market place. What I want to do over the next thirty
minutes is give you my view of what I think is one of the biggest opportunities
available in the communications market at the moment. That opportunity is
about transforming UK business and I make no apologies but this morning I will
not be talking about technology - that comes later. I am very much going to be
talking about the opportunity around Broadband, about creating Broadband
Britain and the size of that opportunity, and the size of that market - and in
marketing size is always important.
SLIDE 3: The presentation covers….
Before I start there are three things I want to cover. Firstly a quick update of
what’s happened since the last auction, this won't take long. The middle point is
about the emerging Broadband market and around Broadband Britain, the way it
is going and the impact it might have. We need to remember that broadband
Britain is not a policy, it is a fundamental change in micro and macro economic
drivers, driving the need for a economic transformation in Britain. Finally, I intend
to talk about broadband fixed wireless access’s role within the creation of the Formatted
new market place, the new market space.
SLIDE 4: The broadband market is set to be the fastest growing segment…
What are the major points? Well for those of you who had an early start you can
listen to the next three or four minutes and then go to sleep and wake up again in
half an hour, these are the things that I want to cover. Despite economic
uncertainty over the last eight months, if not a year, in the telecom field this is still
a growing market, a rapidly growing market. We are still potentially looking at 11-
12% compound growth rates in this market place, but I think we are going to see
some major structural changes in the market, the days of narrowband, the days
of analogue are over. The voice market if anything has contracted in revenue
terms, the future is going to be about digital, it is going to be about data, about
Broadband. This is the fastest growing sector in the telecommunications industry.
SLIDE 5 & 6: Summary of key points
It is also about some of the macro/micro economic drivers, and the drivers for
enabling business. What we have seen over the last five or six years, is non-
inflationary or low inflation growth, across the global economy. The belief is that
much of that is being driven by the investment in technology. Technology, not
just in PC's, not just in micro-processors but technology such as ERP systems,
CRM systems - these are the ones that are really driving the technology growth
within the US economy and now the UK economy. These technologies actually
strip out huge amounts of cost from business, making them more efficient as they
e-enable their internal processes, remove transaction cost and outsource non-
core activities - that process is now going to continue. However much of that
change has been driven by large corporates installing high bandwidth, private
and virtual private networks (VPN) linking themselves internally as well as with
suppliers, customers and partners. This is the networked economy.
However I think we are going to see a shift down the size chain. We have seen
much of that change within large corporates, we are now going to see it in small
and medium size enterprises, we are going to see it in the small home office
market place. Those micro and macro economic changes are also being driven
by global competition and the need for all companies to become far more
efficient, to take out transaction costs.
So those are some of the macro/micro economic impacts that are happening.
Putting that in context some of the big barriers in actually getting this economic
miracle pushed down into smaller companies are probably around two things,
software and bandwidth. Software prices are coming down, coming down rapidly.
Large scale CRM and ERP software is now becoming affordable to smaller and
smaller companies, particularly as the application hosting model kicks in. But the
final barrier remains and that is the cost of bandwidth. The ability for small
companies to communicate internally and even more importantly for the whole
economy to communicate across itself - companies talking to suppliers, to its
customers, and to its employees is being prevented. The big barrier here is still
the cost of bandwidth - it costs £5000 for a 2MB link, that is a major barrier for an
SME. But we are now beginning to see the technology that can drive down that
price from £5000 a year to £25 a month, and there is a lot of evidence to
suggest that Broadband a highly elastic good, and that at practically a tenth of
the price or where the price has been cut by 90%, demand will explode. We
would see a huge explosion in demand for that Broadband, if that’s what
happens to prices.
What else has happened? The closure of financial markets - and I don't mean the
unfortunate closure over the last few days, I mean the closure of financial
markets to the Telecommunications industry in general -this has shut off their
major source of growth funding, particularly in the cable TV industry. Certainly
the UK's cable companies will no longer be able to drive out the networks, their
focus over the next three to five years is about sweating the assets they already
have. They are not going to roll out their networks any further than they have
already got to date. If they are, very marginally. That leaves perhaps fifty-sixty
percent of UK business without Broadband access, via cable.
There is huge untapped demand for Broadband, in urban, semi-urban and rural
environments - and there are a surprising number of UK business in those
geographies and - again something else I will not apologise for - I will speak
almost exclusively about the opportunity in the business market, for Broadband
access. Our belief is that this is where we will see the first real demand for true
Broadband. It does mean that the residential market does not offer a huge
We will also see that large tracks of the UK do not have any competitive
provision of any form of Broadband service and won't for a significant amount of
time. There is a huge opportunity in terms of time to even address the markets
that are incapable of coverage. Cable modem growth is now becoming quite
significant, both in residential and business market but it is confined to those
geographies that have cable. So, that is what I wanted to cover and, as I say, if
you have had an early morning then you can sleep and I will give some
background and depth to those opening comments.
SLIDE 8: Round-up since the last auction
Since the last auction, quite a lot has happened in terms of the telecom market.
The critical ones are quite clearly the failure of local loop unbundling But also the Formatted
slower roll out of ADSL - it is almost a race to be the last to market from that
particular segment at the moment. So that is the downside, these are the issues
to be addressed. But on the positive side, we still have seen the huge growth of
internet and intranet, not just in the large corporates but now increasing in the
small business place, and even the self-employed. So a huge potential untapped
So that is what has happened, what is happening in the Broadband market, as I
said in my opening comments the telecom market continues to grow despite its
difficulties over the last eighteen months. Perhaps a consolidation view of
broker's reports and various other industry views of how the market is going to
grow. Clearly, there is still huge growth, people even suggest 11% compound
These markets are also clearly undergoing very large structural change, and the
biggest two areas of structural change are the growth of mobile and the huge
growth in fixed data -around £13 billion to £22 billion. And what is driving that
growth? Massive micro and macro impacts are driving that need for fixed data
into UK corporates and eventually into UK households.
Slide 10: There are a number of key drivers underpinning broadband
These are some of those factors.
Globalisation and global competition means that almost everyone has to
become more competitive and that means driving down costs.
We have seen the impact the technology investment has had on the
activity growth in the US in particular, but also in the UK.
Those factors are now moving down into smaller companies, they are going to
have the latest techniques in communicating within their business, e-enabling
their supply chain, their demand chain, their communication with customers and
suppliers. So there are some of the big macro effects and indeed Mr Greenspan Formatted
(Federal Reserve Bank) has made it clear time and again that he will continue to
reduce US interest rates to stimulate non-inflationary technology investment.
There are some big micro impacts, that need to make business more efficient, to
reduce some of the drag and friction that is in the economy, not having to
transact via non-electronic means. There are some very powerful drivers, driving
the need for fixed data.
SLIDE 11: Broadband access will become an essential commodity for all
I think we can go down several layers lower and begin to understand some of the
detailed applications within the corporate market which is now very well
understood, most UK corporates have e-enabled themselves internally. With
intranets, with ERP and CRM we are now seeing that spread across the rest of
the UK economy. The introduction of VPN's to actually allow home working for
instance is another major impact.
So we are beginning to see that spread from the corporate sector into the rest of
the UK economy. But we are seeing the spread of the need but not the spread of
the technology to meet that need, and that is a point I will come back to.
There are other slides in you pack, which give more background which I won't go
into today. (SLIDE 12: Strong growth in the IP-VPN market)
SLIDE 13: Availability and the level of competition are the principal barriers to
broadband uptake in the UK
We can also look at what is happening elsewhere, examples across the world of
Broadband penetration. Clearly some of our competitors are well ahead of the
UK - the UK is poorly performing in terms of Broadband access. The US,
Canada, The Netherlands, are far in advance, in terms of how much cable
modem access they have, how much ADSL access they have got.
And we can do a lot of thinking about why that is happening but fundamentally
there seem to be two reasons:
they offer cheaper prices than we do and
it is more available
It is pretty simple.
And price seems to be the major factor.
SLIDE 14: The UK scores poorly in almost all measures of local access
It is also about competition and availability and we need to look at how the UK
stacks up compared to some of our European competitors. The biggest barrier to
e-enabling the UK is in the final mile, and that final mile does seem to be the
problem of a monopoly power in the UK. To give some depth to that statement,
see how the UK ranks against some of the other slowly liberalising neighbours,
clearly there are some big issues.
Slide 15: Lack of competition means that high broadband pricing in the U.K is
prohibiting growth in demand
The other issue with the final mile is that prices are still very high, too much for
most small sized companies. Compared to US our prices are remarkably high,
even though the UK has the advantage of a geographic concentration which
should actually make our local loop much cheaper than in a country as diverse
and as sparsely populated as US.
Slide 16: Nevertheless, internet penetration levels shows there is latent
broadband demand in the UK
Just a one last look at some more statistics on internet penetration. The
hypothesis here being that internet usage is a driver of broadband penetration.
You can plot many countries on this graph although there is a pretty good
correlation between household internet and broadband penetration, and the
interesting point here is just how for the UK is off the curve with penetration of
internet in homes of about 33% which would suggest that we should have very
nearly twice the Broadband penetration in this country that we currently have.
Two points to take from this: firstly that the UK seems to be lagging and that
latent demand is not being met: and, secondly we are close to an inflection point
which means that we may begin to see Broadband demand grow quite
explosively up to US-like levels. And broadly that would suggest that over the
next three years something like a tripling of Broadband demand in the UK.
Again the next couple of slides are in the pack and I don't intend to talk about -
although they are there for your reference.
Slide 18: Data access price sensitivity suggests broadband is an elastic good
I made a statement that Broadband prices are high – well, here is some more
evidence, based on BT's latest leased circuit prices. Around about £4,500 just to
install the circuit, somewhere between £250 and £600 per month in rental, which
is pretty high. Again, a lot of the research that we have been seeing suggests
that it is a highly elastic good. If we can get to a tenth of the price, it would mean
a huge explosion in Broadband demand. Price sensitivity is the issue, and that
can be taken two ways. Does this mean that prices will collapse and then there is
no market for broadband –this is one way of looking at it. But a more positive
way is to suggest that there is a huge price umbrella available in the market
place and all new entrants can take advantage of that, whether it be from cable
modems, ADSL operators or Broadband fixed operators. Again that demand is
there in the business sector.
So that was a very quick look through what is happening to Broadband in Britain
A strong latent demand for e-enabling business driven by micro and macro
economic events and very high current prices and very poor availability which
suggests to me there should be quite an opportunity for someone coming along
with a new technology, and with a new entrant strategy. What might that look
like? Well, perhaps I will spend the next two minutes going through that.
Slide 20: BFWA could capture a significant share of the market ahead of ADSL,
cable, and satellite which are slow to roll-out
This is perhaps the simplest but most graphic way of explaining the positioning of
fixed wireless access, two axes, very simple. The vertical axis shows the size of
a particular company and the horizontal axis the geographic region of the
country. The numbers represent business sites that exist in the twenty-six
licensed areas. Broadly, we can see satellite has a franchise to cover 100% of
those regions, about 1.2 million addresses exist in these 26 franchise areas.
Current penetration of Broadband in the grey area is broadly non-existent. Fibre
has been rolled out now to most major UK conglomerations even down to quite
small sizes and that is 120 thousand business addresses in that 10% area, pretty
high current penetration. Perhaps an area that wireless access may find hard to
penetrate. In other areas, cable modems are predominantly available. If we are
to believe that the financial markets are to be closed for the immediate future, for
the cable operators, and that they will not be rolling out the networks any further
than they are now, and they are sweating that asset it suggests that about 50%
of UK businesses will at some stage be covered by cable modem. This
potentially represents about six hundred thousand businesses. Currently less
than a thousand households have Broadband access and that is not because
they don't want it, because the demand drivers are there - it is because they can't
get it or they cannot afford it, it is too expensive. That figure will grow rapidly.
NTL/Telewest are about to launch or relaunch their Broadband business
services. Clearly ADSL, assumed to cover 80% of the geography of the UK at
some stage, if we ever get to there, represents close to one-million customers in
those areas, with current penetration of about less than 5000. BFWA can cover Formatted
about 90% of the business in those 26 areas and served by a reasonable fixed
wireless service. It is an addressable market of something like 1.1million
organisations or business that will require broadband - they just might not be able
to get it.
Slide 21: The market opportunity for BFWA is potentially lucrative
Perhaps a slight health warning on what the indicative market might be. The
number of business that exist within the 26 franchise areas, is about 1.2 million. It
gives us a view of the potential market might be. In total £4 billion is spent by
small business on communications in those 26 areas. If you go back to the first
chart I showed you, you can see how the structure of the £4 billion is going to
change, there will be massive growth in that £4 billion anyway, but much of the
growth, not all of it, will be in fixed data. We can divide those franchises into two
groups. Hot spot groups - these are areas with a high density of customers - we
would expect cable modems and ADSL to get through quite quickly. And then
the rest of the franchise area which is low density, but even here an area
probably not addressed by cable modems and unlikely to be addressed by ADSL
in the immediate future, there is still a £2.7 billion market now, potential growing
at 11 % compound. Of course we have not factored in the residential. So,
perhaps a substantial market opportunity may well exist.
Slide 22: BFWA presents a number of key advantages
I said I wouldn't talk much about technology and Broadband wireless, but
perhaps just some pointers to remind ourselves of some of the advantages that
fixed wireless access will give us: how rapidly we can speed up roll out, the
scalability, even more important, the way we can build up incrementally over a
period of time, picking off area after area, is perhaps something that is important
in managing risk and investment costs in this particular opportunity.
Slide 23: With the opportunity to address high business density “hot spot”
clusters in broadband poor areas
For example, it begins to look at the density of businesses by postcode in this
particular franchise area. Clearly there are those hot spots, but interestingly even
within those hot spots there is still very low cable modem penetration at the
moment and very low ADSL penetration. My belief is that this is not because of
lack of demand for these services - there is - it just has not been penetrated with
those technologies yet. Nevertheless, over time the hot spot areas are likely to
become more competitive, although BFWA will be able to capture a significant
share of this market, particularly if it gains an early foothold in the market. The
question is also how far cable and to a lesser extent ADSL can roll out to the less
urban areas, which represents a greater opportunity for BFWA.
Slide 24 & 25: Summary of key points
Wrapping things up, I hope I have demonstrated the fixed data market is the
fastest growing market by some considerable margin, which is the major growth
area. We are seeing demand for that fixed data, being driven by some clear
micro and macro economic factors. We need to invest in technology, we need to
be able to compete globally and we need to try and drive cost out of business -
we need to become more efficient. We need to network more, clearly, with
customers, suppliers with employees and with partners, as it is becoming a real
need. We are becoming a networked economy. We have seen in the large
corporate sector that e-enablement has already happened, it continues to
happen, it is happening internally within organisations and they are building
networks around themselves. Those same drivers are now being seen again
with the smaller sized business market, so the macro and micro factors are, I
think, very key.
The major barrier to the introduction of network economies within that sector is of
course the cost, but we are beginning to see that prices are coming down. But
that latent demand and that large price umbrella does suggest that market
opportunity is there. The closure of financial markets has prevented, and will
prevent cable TV organisations, from getting further in their network build.
Talking about the large price umbrella, we are talking about something like a
tenth of the price of these circuits. In terms of the competitive environment, most
UK businesses outside the metropolitan areas will have very limited competitive
access to Broadband technologies. Large areas will have very little or no choice
How elastic is Broadband? What will be the effect if prices come down by a factor
of ten? If it does what will happen to demand? Clearly we will be seeing what
happens in regard to the competitive environment, perhaps it is not such a
competitive environment as we suspected 12 months ago. Of course I have not
talked about the residential market which represents further opportunities for
broadband. Perhaps if I had to sum it up into my final point, in three key points,
technology driving down the cost of entry, we see is currently very high prices for Formatted
Broadband access in the UK, giving that price umbrella, and we see low
competition. Any high priced, limited competition market with falling barriers to
entry is one worth exploring. In my mind there is enough of a case to suggest
that this is something worth looking at. And I will leave my presentation at that
point. Many thanks for listening.
Thank you very much Martin for that, we have ten minutes for questions.
I am very concerned about the financial models for telcos and the flat rate
charges that you have mentioned via the cable operators that you have already
mentioned - I question whether they are financially long term solid models. In
other words, the telcos are not making money, can we sustain these low sums?
If we can't, then is the wireless access going to get offered at a substantially
lower cost base in order to make it viable, or indeed as an industry have we got
to increase expectations of what people have to pay for access not to rely upon
advertising or content revenues?
I can't comment on the individual business plans of the cable modem operators,
for instance, but there are several things we would need to consider. Let's
assume that they had set their prices at a viable rate - and there is no reason to
suggest that they haven't - which suggests that that is a real, a viable floor for
Broadband prices. I think the point I would convey is that we have this huge
price umbrella - it is not so much competing with cable modem operators, it is
being an operator, a service at all in a relatively low competitive market. If we
can come in at lower than £5000 a year, we would see a pick up in demand. The
other point is why are telcos beginning to move to a flat rate price model, in turn
in an IP environment. The fixed costs are now becoming quite substantial and it
is more appropriate in some cases to charge on a fixed basis and it can reduce
costs quite considerably. An example is in billings systems - it may not sound
significant but a billings system can cost £30 or 40 million: flat rate billings
systems are a very simple way of reducing costs. But I think there is
apprehension whether they go for flat rate or volume base, but I think flat rate is
where we are at the moment If it proves to be over time that flat rate is not alright
then the market will change. It needs to be considered but I think it is only
advantageous that we have a flat rate, as it makes life simpler, for consumers as
well for that matter.
I would like to follow up the question about whether the current prices are
sustainable. They will not be sustainable if the capital costs and the operating
costs exceed the revenue that can be generated. You have told us in a
fascinating way what the commercial opportunity is, such that one would like to
throw away my legal papers and go and invest in a local access network if I had
the skill or the money. How much money would I need? I imagine that you
would have done some financial modelling of the various technologies of cable,
ADSL and fixed wireless access - are you able to tell us anything about the broad
models, without commenting on particular business plans, the kind of thinking
that you have done in relation to the relative costs of the various technologies
that are available for running a local access network?
Yes - you are absolutely right, I can’t comment on the details, but we have to
certainly look at what we believe to be the cost of processes which gives us an
understanding of how much it will cost to roll out Broadband fixed wireless. We
have certainly looked at the capital cost, we have looked at the operator costs,
internet/intranet and given ourselves a comfortable view that these are the sort of
costs that will be required to drive out a Broadband network. The return that is
available is sufficient to reward the risk that is being taken. We haven't looked in
great detail at cable modem operators - we don't need to, we know what their
prices are - and we need to convince ourselves that there is a reasonable
temptation that the prices at Broadband will be under £5000 a year which is the
ceiling and can come in at £25 a month which is the floor because that is what
cable modem operators charge. For a significant amount of the geography of the
UK, there is very little competition anyway.
What I see in your presentation is a lot of overlap with conventionally what was
deemed to be the fixed wireless access market. I see the competitor cable
modems as being the fixed wireless 3 5 GHz this I don' think will ever be priced
at £25 a month - I don't think it needs to be. There is much more margin in there.
If someone could just give us five minutes as to where we are with the 3.5 GHz
because there is, from a technology point of view very little overlap, but from a
market quite a big overlap.
Can I stop you there. We will start the question and answer session later this
morning with a quick 5 minutes on the 3.5GHz. Because, as I say, the overall
market view is true, but I think certainly some of the operators here need to see
where we are with that new technology, put it in context.
What is your view that OFTEL or Brussels will force the unbundling of the local
loop, which will enable ADSL to massively accelerate its roll out?
Well, that sounds like a question for the regulators. They have introduced local
loop unbundling as a concept, and what else can they do in terms of actually
enforcing it - I am not sure if there is anyone that has an understanding of the
finer points in the EC competition law that might be able to answer that. But it
seems to be stalled, in this country anyway. I won't comment, because I don't
have a strong understanding.
Thank you for that, lets break for lunch. Thanks to Martin for his clear
presentation, I think the message for you is he who dares wins.
Can we make a start? Just before we go to Steve Talbot’s speech, can I just
raise one issue - quite a few people have asked for a list of attendees. Now, we
are more than happy to provide a list and we will do so, but if any of you don't
want to let anyone know that you are here, then please let one of us know and
we will take you of the list. Can we move on to Steve Talbot, who- he is head of
bBroadband engineering. Steve is going to talk about bBroadband technology
issues and some related spectrum packaging matters.
BROADBAND TECHNOLOGY - STEVE TALBOT, RA
Good morning. I should just say that the unit that I am in, is concerned with the
allocation at 28 and 40 GHz, so any of the engineering issues that come up here,
they really are concerned with 28GHz Broadband, and therefore try to give
people, who are not primarily concerned with the engineering aspects, what
SLIDE 3 – ‘Broadband’ – the term
Basically, in the old days when we used to listen to the radio or watch TV, it was
an analogue signal, we now of course digitise things. [TALBOT SLIDE 3]. We
used to have a piece of spectrum where we could put a radio signal and that was
pre-ordained but now we can encode these signals and we can put them across
a delivery method, be it hard-wired, fibre, radio or wireless, and as technology
increases, we are learning how to increase those data rates and one of the
examples is digital TV. Where we once had a piece of spectrum that could
deliver one channel, we now have an equivalent piece of spectrum where we can
deliver three/four channels. Arguments about quality - we will leave that up to
[TALBOT SLIDE 4– ‘Broadband’ – the term]
A little chart to show how Broadband sits in the tier of things. As we move up the
scale we have the data rates for ISDN, then to higher bandwidth these are the
data rates of FWA, and then we move up to broadband where we are at 28GHz,
then we move up to optical and fibre optic.
[TALBOT SLIDE SLIDES 5 AND 6 – Current Broadband delivery methods, and
Comparison of technologies
Looking at the different types of different delivery methods, the two wireless ones
there that stand out are satellite two way interactive and Broadband fixed
wireless. Very similar in their design except the base station for the satellite is
rather further away than that for Broadband. An emerging technology is optical
links, which are also slowly being delivered. [TALBOT SLIDE 6] Again, some
more data rates, we have sat ourselves pretty much in the middle there.
[TALBOT SLIDE SLIDES 7 AND 8 – BFWA general system architecture]
When we use a wireless delivery method, what do we actually have? The more
common analogy is that where we have an access point transceiver, which then
squirts info to the customer equipment. This is the last kilometre of the exchange
to the customer and we highlight the different types of services that could be
offered. [TALBOT SLIDE 8] Mainly, in the past, where we would have a data link
between two premises it would be delivered via point to point links - BFWA more
lends itself to having a point to multipoint system, and in the future multipoint to
multipoint. Basically, it allows connection between a customer’s premises and
the network without the need for a hardwire connection.
[TALBOT SLIDE 9 – Advantages of wireless delivery methods]
These are some of the possible advantages of wireless delivered services: some
of the hurdles of hardwire are the logistics of installing the network, digging up
the ground, annoying the locals, wireless tends to make itself more useful where
logistics don't allow and also you don’t have to lay a complete network.
[TALBOT SLIDE 10 – Topologies can be designed with the end-user in mind
] I spoke earlier about multipoint to multipoint, this diagram shows a typical
system. You would have a central station and its own services delivered, this
could be fibre optic, and then this relays information to the residential subscribers
or the commercial ones. And that gives the operator the possibility of delivering
quite high bandwidths to each individual user. You could also argue that a
disadvantage is that remove the base station and you remove the entire network.
If you move to the mesh network - this doesn't have any central base station, it
allows for growth. Where there is no central station, it can re-root.
[TALBOT SLIDE 11 – Operator issues]
Wireless isn't a utopia and there are issues that a wireless operator has to bear
in mind. Certainly at the higher frequency in the 28GHz band , you will have your
radio frequencies affected by certain conditions and there is an effect known as
rain fade where, in heavy rainfall, the signal can be weakened. Operators would
build in a margin and use automatic transmitter power control (ATPC) which
would retain the link, and increase the power in times of outages. Planning
permission is in the media at the moment - it is another thing for the operator to
bear in mind. In the Information Memorandum, we explain that we look to
operators to share masts, to relieve planning issues. Of course, it will still require
a backbone infrastructure depending on whether you installed a central station,
and a point to point link, that would be the operator's decision.
[TALBOT SLIDE 12 – Key licence conditions
]The Agency has, after consultation, issued an information sheet referred to as
RA3909 – this can be found on the website and it aims to give operators some
guidelines as to how they go about co-ordinating their system. We will get onto
that later. Also to bear in mind, the equipment has to meet the interface
requirements, which are again our on the website. These give information on the
standards that the equipment should conform to, if they are going to be deployed
within the UK.
[TALBOT SLIDES 13 AND 14 – Radio use needs co-ordination and Same region
] Where we have the spectrum packages available these are allocated, three
packages per region, and you will run into two scenarios. One is where you will
have two operators that share the same spectrum and are in adjacent regions or
where two operators in the same region have adjacent spectrum.. Where two
operators share the same guard band we have said that if they can discuss
amongst themselves, there is scope for both to use the guard band. Boundaries,
this shows what I was explaining previously. There are three spectrum blocks
and shows the guard band between the two services. This channel plan is a
published ERC recommendation. This plan has been worked out on the basis of
FDD deployed systems. It is anticipated that the majority of services will
utilseutilise, FDD (frequency division duplex) equipment, but if the systems are
not going to be using FDD (there is an emerging technology known as TDD -
time division duplex), then there are certain other issues to bear in mind and this
may require adequate separation between transmittiiers, as referred to here.
These are only guidelines and these issues would have to be discussed with
your neighbouring operator, should you wish to deploy such technology.
[TALBOT SLIDE 15 - Same region adjacent spectrum
] This gives a quick overview of what we decided after studies and findings - this
shows the PFD (power flux density) level that is used as a trigger level . This is
what we would hope the operators would use when planning their system. The
co-ordination guidelines state that if you exceed the trigger level, at the border
then you should liase with you neighbouring operator.
[TALBOT SLIDE 16 - Radio use needs co-ordination
] We do, within the guidelines, state that we wish operators to liase, one could
argue that the RA should get involved initially but that would introduce an
unnecessary delay. As the information is avaliableavailable to know who your
neighbouring operator is we encourage you to liase directly. As I have stated,
although we expect operators to resolve problems themselves, the Agency can
arbitrate if necessary. We would look to see if operators had followed these
guidelines, in those situations.
[TALBOT SLIDE 17 – UK radio interference requirements
] The interface requirements for the use of the equipment in the UK. There are
two particular, again on the website. They make reference to various ETSI
technical specifications. These have copyrights attached to them so we cannot
put them on the Agency web site . These are the two for the equipment, for
multipoint, point to point and 28GHz radio systems.
[TALBOT SLIDE 18 – Key points
] That was a very brief presentation: here are the key points, particularly of
interest to operators. The fact that it can be argued that wireless applications
have various advantages over hardwired have been presented. We are in a
position to inform you that one of the licence award winners is currently rolling
out services at 28GHz. And again to make a level playing field, we are looking
for operators to resolve frequency co-ordination problems that they might have.
That wraps it up for me.
Thank you Steve. We have ten minutes for questions.
Joe Hall. What sort of radius could a successful licenselicensee hold, and expect
We are not looking for major networks and I wouldn't anticipate them being more
than 5 kilometres. You have to bear in mind that at 28GHz free space path loss
is greatly increased, due to the frequency band used and it is very much line of
sight, so if you can see the building you can normally work to it.
Does that mean if you don't have a line of sight that it won't work?
Well, it depends how far away you are. If it is line of sight and far away then you
have other things to bear in mind, but if it’s close and you can't see it, it depends
on material it might work through, but we are talking very short distances now.
Steven Lowe, Broadband Wireless Association and Telewest. II don't want to get
into the details, but your slide on comparison of technologies worried me slightly
because I may have misunderstood it. It has ADSL as being 2 upstream and 256
downstream and I had a feeling that most of us would reverse that.
I am going to reserve judgement on that and refer it to the editor. The slide was
provided for me so I cannot verify its accuracy.
Have you looked at the advantages, when you said there was rapid and flexible
deployment have you talked at all to people who have control over planning
permissions? You raised it as an issue, getting a site is anything from 6 months
to 18 months, and with the health issues that are arising in wireless, is there
cross conversation going on?
If I was speaking from an operator’s point of view, one of the issues is that of
planning regulations, which are changing all the time, and we are aware of the
press interest in the health issues, I don't want to get drawn into that, I am not an
expert, in that field, although the RA has been involved in the consultation
process, regarding any changes to planning regulations.
Shall we move on to Joe, who is head of the BFWA team.
AWARD OF 28 GHz SPECTRUM LICENCES FOR BFWA - JOE SONKE, RA
Good morning. I hope to give you an overview of the award process, how we go
about it and why we are now awarding licences.
SLIDE 4 – Government Broadband Strategy
I would like to put the process in the context of Government policy for broadband.
We work within the overall government objectives and priorities. We are not
simply out to award licences for its own sake but we want to help further the
government strategy. You may be aware of a White Paper that was published on
13 February 2001 – UK online: the broadband future - which set out the
Government strategy over the next few years. The overall aim is for the UK to
become one of the best places to conduct e-commerce in the world - quite a
daunting target. And to be more specific, it is to become the most extensive and
competitive broadband market in the G7 by 2005. When we think back to Martin
Heath’s presentation, you can see the size of the task. Nevertheless, it’s an
aspiration that we are working to and broadband fixed wireless access has its
place in achieving that target. That is one of the reasons that we are trying to sell
I've just got a few bullet points that show elements of the strategy. The
Government has established a £30 million fund to be used by the regional
development agencies and the devolved administrations to run pilot schemes to
find ways of getting broadband rolling out to the less commercial areas, the rural
areas. I think the fund is opening in October for bids. Another aspect, with a
similar aim, is to try to band together public broadband services to make them a
more viable and attractive business proposition to potential suppliers, and again
targeting the less viable areas.
SLIDES 5 & 6 – Objectives in licensing BFWA spectrum
The objective is to offer an award of licences to encourage the provision of
BFWA services throughout the UK. We are not simply in the business of selling
licences, and not in the business of trying to generate money for the Treasury.
The key driver is the achievement of this objective. We have other subsidiary
objectives that fit in with this and also address other points. One objective is to
make the best use possible of the available spectrum, and that was a factor in
the spectrum packaging. We also want to promote competition between the
various access mechanisms: we want to get BFWA established, to develop a
competitive, low-priced broadband market. The UK market appears to be over-
priced compared to the US market, we would like to see competition to drive
down those prices, and the award of BFWA licences I hope will be a part of that.
But we have valuable spectrum and we need to make sure that we get the true
value out of the spectrum. We believe the best way to do that is for people to
pay the market price for the spectrum they are purchasing and we think the best
way to set the market price is via an auction process. It's one of the reasons that
we ran an auction last November. These objectives have driven us throughout
the last two years.
SLIDE 7 – Auction 2000 – outcome
Just to remind you of the outcome of last years auction – we had 42 licences
available throughout 14 regions (three in each), and 16 were sold. Bids actually
were fairly close to the reserve prices, and it was an eye-opener for some
commentators - there was a fear that billions might have to be spent on obtaining
a licence. The auction that we ran demonstrated that auctions don't necessarily
lead to very high prices. Although we sold less than half the licences, in fact they
were in the most densely populated areas in terms of business in the UK and so
we had just over 55% of UK business located in those regions where licences
were sold. So it wasn't too bad a coverage in those terms. But we would like to
see coverage of 100%. We have 26 licences left for award.
SLIDE 8 – Auction 2000 – licences awarded
This is a summary chart, it shows details of the licences sold in last November’s
auction. You can see that the three high value regions went - all three licences
were sold in London, Manchester and the West Midlands. But the prices even in
those high value areas were at or a little above the reserve prices. The other
region to sell all the licences was Northern Ireland, which was the lowest valued
region, but it was a competitive one.
SLIDE 9 – Award process 2001-2002
If I can move on to the award process that we are about to embark on. Some
have asked why are we bothering to do this, and this should be answered over
the next bullet points. As soon as the November auction was over we had a
number of people ringing us and asking whether they could pick up the remaining
licences. I think some of them were hoping to pick them up below the reserve
price. But there is, I think, real interest in purchasing the licences which remain.
Martin explained how competitive the access market is - that was the case 12
months ago and there is still competitive opportunity.
The next bullet point is that I think there is now a realisation that if you enter into
an auction or award process you may not have to pay very high prices, people
see that and will not be intimidated and frightened off. Also, there is an
awareness of the market and the opportunity, we did little in regard to promoting
and advertising before the last auction. There appeared to be a buoyant
telecoms market, but that changed just before the auction. This time we want to
raise awareness to the fact that we are awarding licences. In July we issued the
information packs which you all have. And we have been meeting people that
may want licences and explaining the process, as we are doing today.
We also have changed the process. The auction we held in November lasted for
just over a week or so and people had a very narrow opportunity to build a
business plan and to decide if they wanted a licence. This time it is rather
different. Generally companies pick up resources when they think the market is
right, and we want to address that and reflect it in the award process by actually
having a 12-month window of opportunity. People can come forward within that
time frame and put in a licence application. So we hope that people will have the
opportunity over the next 12 months to develop their business case and to see
how the market is going and look at the activity in bids and this will help them to
develop a better based decision.
These are some of the reasons that we feel it right to embark on this award
process - we think there is a market opportunity and we want to give people the
chance to develop that opportunity. If we kept the licences on the shelf that
opportunity would be lost. It is up to the operators themselves to decide if they
actually want a licence - all we can do is put the resource in the window and
leave it there for people to pick up as and when they want.
SLIDE 10 – Regions with licences remaining
This is just a pictorial representation of what we have available and will be
available for bids. We will be open for bids from around the middle of October.
SLIDE 11 – Remaining licences Formatted
Here is a tabular representation of what is available. Maybe the key thing is that
the remaining licences have reserve prices of either one or two million pounds.
The more expensive ones went last year and so did the cheapest ones.
SLIDE 12 – Licensing basics
We have maintained virtually all the conditions that applied in the first award
process. Certainly companies picking up licences then may well have
challenged us if we made changes to the licence conditions or what might have
to be paid for the licence, because that might have put them in a
disadvantageous situation. There are also some companies that did not want to
take part in last year’s auction because of the licensing conditions - to change
them would have changed retrospectively the whole aspect of the auction.
So, just to run through some of the key features. We are offering 15 year
licences, which is important to remember. Bids are open from any corporate
body whether they are based in the UK or overseas. Again, only one licence per
region will be available to a bidder and if someone holds a licence which they
picked up in the last auction, they are not allowed to bid in the relevant region
this time. But at the same time there is no limit to the number of regional licences
held - national coverage is open to operators. The ‘use it or lose it’ clause is an
obligation placed in the licence so that within 18 months of issue the operator
must have established sufficient network to be capable of serving 10% of the
businesses in a licence region, though these may not all be customers. This is
meant to be a sign of commitment that licensees want to actually use the
spectrum. It’s not all about selling licences, it’s about selling them as a means to
developing services. The ‘use it or lose it’ clause ensures that. The licence can
be revoked if the obligation is not met.
Another key point is that licences are to be used for access delivery only. A
major issue in our initial consultations with industry was whether the spectrum
could be used for something other than broadband fixed wireless access.
Again, the objective is the provision of broadband fixed wireless access services
throughout the UK, so we took the decisions to say ‘No, we do not want the
spectrum to be used for 3G backhaul’, which I think was in some operators’
minds. We want it to be used for broadband access and this remains the case. It
clearly makes a big difference to the business case whether you have freedom to
use the spectrum for 3G backhaul, and no doubt we could have sold more,
maybe all the licences if the conditions were liberalised. But restricting use to
access delivery remains a key licence condition.
SLIDE 13 – Award of remaining licences
I'll move on to the award process - how we will actually award licences. As I
said, we are offering all the remaining 26 licences and there will be a 12 month
window in which people can apply for them. Within the next week we should be
tabling the Regulations before Parliament which will enable us to embark on this
award process. And 21 days after that we can start the process. We reckon we
will be ready to receive applications around the middle of October.
The awards will be on a region by region basis. If someone wants a licence in a
specific region they can apply for that region. The auction last year was an
auction of all regions in one go, and people could switch bids between regions.
Given the auction we were running, it made sense and worked. But for the shop
window approach, it makes sense to do it region by region.
We are inviting bids at the reserve price. The prices are fixed - we are not
intending to offer licences below the reserve prices. If someone puts in an
application then that will be taken as a commitment to take up the licence at the
reserve price if there are no other bids. An application will specify the licence and
the region. Once we get the application, we will put details onto our website and
publicise it in that way and invite further applications, either for that particular
licence or another licence in the same region. There will be 20 business days for
others to put in bids. At the end of that period we will consider the applications
and decide whether we need to go for an auction or not.
Basically, if there are competing bids in a region we will need to sort out who is
going to take the licences via an auction process. If there are no competing bids,
we will then - after having made checks on the bidders - award licences without
an auction. If there are two bids for the same licence in a region, then we will
need to go into an award process and auction. So really the number of bids will
decide how we actually award the licences.
SLIDE 14 – The licensing process is designed to be as simple as possible…
This slide goes into more detail. We check the applicants for their bona fides to
make sure they do not have, for example, directors that have been disqualified.
There are one or two other checks to make sure we believe it would not be
against the national interest to award a licence to the applicant. We also have to
check whether there are any associations between bidders in the same region.
We need to eliminate any bidders who are associated through common
shareholdings. This is quite a difficult question and the rules are explained in
detail in the Information Memorandum.
Assuming there are no associations between bidders, then we will see how many
applications we have and whether there are competing bids for a particular
licence. Then we can either award licences to all applicants or go to auction.
If we don't go to an auction, the award process is simple. Within five days, we
will notify the winners and provide each with a provisional award notice. So it is a
short process, during which the applicants have to do nothing really but wait for
the good news, and then within a 40 day period we will make sure there are no
pre-conditions preventing the award of a licence. Once this is completed we will
seek payment and, once we have payment of the licence fee, we will issue the
Payment arrangements - as with last year there are two options, either pay the
whole fee upfront, or an instalmentinstallment option of paying 50% upfront and
the remainder in years six to ten of the licence period. Again we charge interest
on instalmentsinstallments and also seek a bank guarantee. I think this is a
reflection of the lessons from other auctions, for example in the USA where less
stringent payment terms have on occasion been applied – our counterparts over
there warned us against this because there is always the danger that people win
a licence and don't lose a lot if they decide not to develop services. We need to
make sure we have committed companies picking up the licences.
SLIDE 15 – There is a Multiple-round auction…
Once we get to the auction, the design is similar to last November’s. People put
in bids in a series of rounds and the company that has put in the highest bid in
the round is the current price bidder and may be over-bid in the next round. If the
current price bidder isn't over-bid, he picks up the licence. One more thing - we
decided that the current price bidder at the beginning of the auction will be the
company that put in the first application for a licence.
Because we have a simplified auction process we are not having recess days or
waivers. If we do have to go to auction we would expect that an auction would
not go on for very many days. Again I think this simplifies the process.
Once we have decided via the auction who is to be awarded the licence we go
through a similar process to the one I have just described: checking for any pre-
conditions, getting in the money and then awarding the licence. There is more
detail if people want it in the Information Memorandum.
Perhaps I should mention one more thing. There is obviously some constraint on
when you can go for a licence. If you are looking in Wales, for example, and
somebody puts in a bid for Wales before you, you will have to decide whether or
not you are to go into the bidding process at that time. At the end of an award
process or auction, if for example one licence is awarded where there are three
in a particular region, then the two remaining licences remain on the table for the
rest of the 12 months, or until they are picked up.
SLIDE 16 – Summing up
If I can just sum up what I have been trying to get over. TWe think the remaining
licences offer the opportunity to develop a good broadband business over the
next fifteen years. Operators developing such businesses is what we would like
to see. TWe think the award process is a simple one which won't discourage
companies from entering and will give them the opportunity over the next 12
months to develop a business case and decide when they would like to go for a
That’s it, if anyone has any questions.
Could you just confirm the 16 licences let are still active licences, there are no
defaults on the payments, are there any of those going back in the pool for this
No we have had no licences revoked. If they are returned, and it is a big if, we
will consider and I think they would go back in to the pot. We would have to go
through some procedures in terms of modifying the regulations because the
regulations are framed in terms of the 26 remaining licences. We would have to
through some procedural things but in principle they would go back on offer.
The ‘use it or lose it’ clause as is it set out on page 22 of the memorandum, 18
months from the award of licence will be up half-way through this one year
period, so it is quite possible some licences may be released back in to the pool
during this one year cycle. Have you considered how you might put them back
The timing might be difficult, in terms of the end of next June, only three or so
months left in our 12 month window. We would have to through some
Parliamentary procedures to modify the Regulations, we would have to decide at
the time whether to extend the 12 month period, to allow bids in for licences, if
they are handed back. We are talking theoretically to some extent. In making
decisions potential bidders have all sorts of things to consider. We have got no
indication that any licences will be handed back or taken back.
It seems to me that I understand the complications if you decided to lower the
reserve prices, since there was already demand for these licences at the reserve
prices. But I also see that in 7 regions there was no demand whatsoever so
maybe you can explain why for these 7 regions you decided to stay with the
reserve price. Also I would like to add that the FCC, when there is a re-auction, it
usually has reduced or different reserve prices.
Certainly at the time that we were thinking through what to do with the remaining
licences, our view was and is - and this was in light of advice from our auction
designers - that it would be very dangerous to reduce the reserve prices.
Because of the credibility of future auctions, the danger is that if people saw that
we lowered the reserved price because we couldn't sell the licences they might
enter future auctions with that in mind. There may be some discussions between
bidders that if they held back, then they could get a lower reserve price. It is to
do with the credibility and integrity of the process and we decided not to reduce
reserve prices. I think that there were numerous factors as to why people did not
pick up those licences and the reserve prices weren't actually a fundamental
barrier to people. We set the prices at lower than KPMG's valuation of each of
the business opportunities in each of the regions and reserve prices in
comparison with other costs, like establishing the network, were relatively low
and not a deciding factor.
You touched on the restriction on use for access delivery. I think that that
decision is seriously flawed and that if you had allowed use of cellular backhaul
as well then there would be much greater interest . I think one needs to look at
what we are trying to do and that is get some spectrum out and used for
broadband access and you could have 100% of nothing under the current
arrangements, which is clearly unsatisfactory, rather than 50% of something
significant. I think you ought to go back to your auction designers and legal
people and say we want to remove this condition, tell us how we do it and that
then you will get something moving. And also on reserve prices, what are we
trying to do? Protect integrity of future auctions or are we trying to get
Broadband Britain moving. Lets focus on the second objective.
I don't think there is much more that I can add to me previous comments, but I
must say that it was a difficult decision originally, whether we should open the
way to this to 3G backhaul in particular. We looked at various options - whether
people might be able to use the spectrum for backhaul for the first few years or in
areas that might not be commercially viable for the FWA. But we had to take into
account broader spectrum management policy and issues within the Agency.
We have got bands designed specifically for fixed links and I know my colleagues
are looking at the requirements for 3G backhaul links and we believe there is a
place for them but it is not in 28 GHz. But I take your point, and we have to live
with our decision over the next 12 months and see in fact how wrong we might
be - or how right.
Thinking about this condition, if any new bidder wanted to consider a national
licence and you referred to national coverage. Perhaps we are being
discriminated against as the major licences have gone and why should we pay
the same as the previous people who had the opportunity to do a national
It was going through my mind that the door has shut on a national operator,
except of course anybody who holds a licence in the regions where we sold all
three. Early on in the award process we were asking people what sort of
licensing regime they wanted: did they want local, regional or national licences.
There were some indications that people wanted national but because of the mix
of requirements we designed a regional award process. Nobody bid nationally
though some seemed to indicate they wanted national licences when the auction
actually started. Theoretically you are right but you’re the first one to raise the
problem that any potential national operator is discriminated against. It’s a nice
point but in practice I don't know whether it actually arises.
Would you consider holding back some of the licences on the basis that
someone might start rolling out at 3.5GHz with a view for using 28GHz once the
density or the penetration in a region went up. Rather than get rid of them all, link
them to some of the other bands that might be coming up.
let theyextent. In making decisions potential bidders have all sorts of things to
consider. We have got no indication that any licences will be handed back or
Philip from Euro. It seems to me that I understand the complications if you
decided to lower the reserve prices, since there was already demand for these
licences at the reserve prices. But I also see that in 7 regions there was no
demand whatsoever so maybe you can explain why for these 7 regions you
decided to stay with the reserve price. Also I would like to add that the FCC,
when there is a re-auction, it usually has reduced or different reserve prices.
It is a very valid observation. Certainly at the time that we were thinking through
what to do with the remaining licences, our view was and is - and this was in light
of advice from our auction designers - that it would be very dangerous to reduce
the reserve prices. Because of the credibility of future auctions, the danger is
that if people saw that we lowered the reserved price because we couldn't sell
the s they might enter future auctions with that in mind. There maybe some
discussions between bidders that if they held back, then they could get a lower
reserve price. It is to do with the credibility and integrity of the process and we
decided not to reduce reserve prices. I think that there were numerous factors as
to why people did not pick up those s and the reserve prices weren't actually a
fundamental barrier to people. We set the prices at lower than KPMG's valuation
of each of the business opportunities in each of the regions and prices in
comparison with other costs, like establishing the network, were relatively low
and not a deciding factor.
Martin Burrage from Ericsson. You touched on the restriction on use for access
delivery. I think that that decision is seriously flawed and that if you had allowed
use of cellular backhaul as well then there would be much greater interest . I
think one needs to look at what we are trying to do and that is get some spectrum
out and used for Broadband access and you could have 100% of nothing under
the current arrangements, which is clearly unsatisfactory, rather than 50% of
something significant. I think you ought to go back to your auction designers and
legal people and say we want to remove this condition, tell us how we do it and
that then you will get something moving. And also on reserve prices, what are
we trying to do? Protect integrity of future auctions or are we trying to get
Broadband Britain moving. Lets focus on the second objective.
It is a fair point. I don't think there is much more that I can say, but I must say that
it was a difficult decision originally, whether we should open the way to this to 3G
backhaul in particular. We looked at various ways whether people might be able
to use it for the first few years or in areas that might not be commercially viable
for the FWA. But we had to take into account broader spectrum management,
policy and issues within the Agency. We have got bands designed specifically
for fixed links and I know my colleagues are looking at the requirements for
backhaul links and we believe there is a place for them but it is not here. But I
take your point, and we have to live with our decision over the next 12 months
and see in fact how wrong we might be or how right.
Steven Lowe, Broadband Wireless Association. Thinking about this condition, if
any new bidder wanted to consider a national licence and you referred to national
coverage. Perhaps we are being discriminated against as the major s have gone
and why should we pay the same as the previous people who had the
opportunity to do a national licence.
It was going through my mind that the door has shut on a national operator
except of course anybody who holds a licence in the regions where we sold all
three. Early on in the award process we were asking people what sort of
licensing regime they wanted, did they want local, regional or national licences.
There were some indications that people wanted national but because of that mix
of requirements we designed a regional award process. Nobody though seemed
to want national s when the auction actually started. Theoretically you are right
but you’re the first one to raise the problem that any potential national operator is
discriminated against. It’s a nice point but in practice I don't know whether it
Would you consider holding back some of the s on the basis that someone might
start rolling out at 3.5GHz with a view for using 28GHz once the density or the
penetration in a region went up. Rather than get rid of them all, link them to some
of the other bands that might be coming up.
It’s a good point - we think that there is still an opportunity for people to mix and
match, but we certainly aren't planning to hold back anything from 28GHz
everything is on the table, or will be as of October.
Can we swap over, and then questions can be asked to the whole panel
You stated licences are for use for access delivery and then you added the
words only, can you just confirm that is the transmission to be one direction?
No, sorry. It is both directions, I meant just that licences would not authorise use
not for other services.
Steve Jones who is head of the unit responsible for public fixed access, just to
say a few words about where we are with 3.4GHz.
LICENSING PUBLIC FIXED WIRELESS ACCESS AT 3.4 AND 10 GHz -
STEVE JONES, RA
I have no slides or presentation. Many will have been watching the website to
see where we have got to with 3.4 and 10. The last thing that you will have seen
on the webpage is a document dealing with a complex technical issue called the
duplexing arrangements at 3.4. We have been sitting on a consultation
document which would have given the full details of what we propose for the
packaging of the award and the award mechanism, since before the general
There were four principal issues that we had to sort. First of all is that the
spectrum at 3.4 and 10 is managed not by us but the Ministry of Defence as
military spectrum, and the arrangements of fixed wireless in those two bands are
by agreement with our colleagues in MoD and the other uses that the ministry
have agreed to share that spectrum. Clearly if we are going to offer that as a
commercial proposition that makes sense, we would want you to have the
conditions that are published openly, and available for you so that you
understand the arrangements for the management of that band. In military
spectrum prior to this we have not been able to do that in a fully transparent way.
An award of this kind means we would have to be open. However, the 10G band
has presented a number of new difficulties, principally because the MoD has
been reconsidering their options for 10G. That is the reason that we have been
saying that the 3.4 and 10 are now decoupled. This means that we are going to
get on with 3.4 first and sort out 10 as soon as we can get to it. So we anticipate
that 3.4 will be delivered first and then hopefully not long after 10GHz.
We have also been wrestling with competition issues when we should deal with
these issues, in advance of an award or after. That I am pleased to say has
largely been resolved. The so-called technical this so called 100 meg duplex
arrangement is perhaps the biggest barrier and perhaps the biggest opportunity.
The UK other than Mexico is the only one that has adopted a 51 meg duplex
arrangement. Since we don’t match the rest of Europe it impacts directly on the
business case, what we want to do is arrange it so that it is looking at 100 megs
duplex between the go and return parts. It would then be consistent with the rest
of Europe. The main difficulty is that to do that we have to come to an
arrangement with the broadcasters to move one of their channels elsewhere in
that spectrum so that we can slot the fixed wireless access spectrum in with the
100 meg duplex arrangement. Providing we can do it efficiently, and it would
probably mean a switch to digital technology. What we are trying to do is to
arrange for 2 times 20 MHz with a 100 MHz duplex spacing. It will be technology
The last issue and perhaps the most important, is addressing the market for fixed
wireless in a way that we haven't done it in the past. Since Joe and I started
these processes, a number of things have changed and one of the principal
things that has changed is the requirement for us to deal in a very joined up way
with the regions of the UK. We have been talking to the RDAs looking closely at
how their policy is developing and then trying to work with them to help you
publicise the opportunity that fixed wireless has to offer. Looking at the RDAs
and the devolved governments and their markets the interest they have from
their users. That clearly takes co-ordination.
These are the principal regions for any delay we have in the issuing of the
consultation document, now to the timing. We are still hopeful that most of these
issues will be resolved during the October timeframe, and should be put to our
minister during the month of October. We would have to address any issues
consequent to that. That means that we might reasonably anticipate that
applications for an award process might be towards the end of this year. And to
answer the Steven question directly, does that gives us scope to address 28 and
3.4 in a coherent way, yes I think it does, the very fact that the licences are left
on the table for 28 means that there is an opportunity for you to build a business
case at 3.4 if you wish to do so, and the 28 licenseLicences are still available.
One other matter - many would be aware that in 1997 we ran a competition for
spectrum at 2GHz. Two licences were offered and one was picked up and one
was not and that was the BT licence. In the end they declined to take it up. That
means we have spectrum available at 2GHz. It is subject to a number of
technical constraints. We therefore have in mind a consultation process for that
2GHz spectrum and reviewing the constraints that currently exist. Thank you.
QUESTION AND ANSWER SESSION
Tony Ballard. I am worried about what I have heard about the way in which the
shape of this auction is so constrained by the previous auction. I understand the
point about auction integrity, I am wondering to what extent we will be setting
ourselves a regulatory bound by following the former auction pattern this time so
that we find ourselves in a position were we can't freely take sensible spectrum
management decisions in in future allocation of spectrum in this or other bands.
As the market moves on, as conditions change, there ought surely to be some
adjustments to them and very simple example is the way in which the licence fee
is presented in an unattractive way, that you have to pay upfront, as the deferred
payment terms don't offer a serious alternative because of the cost of money. All
licences up to the beginning of the auction period following the 1998 Act were
charged on the bases of periodic fees and you paid annually and when the
licenseLicence ended the liability ended. Here we have the payment upfront and
that is on the basis of US experience, and indicates commitment. But market
conditions have moved on, as was said earlier, and the financial markets are
closed to telecom companies now. Surely this an example which cries out for
some regulatory adjustments, for the auctions process to recognise that
conditions have changed.
I agree with you in some respects. The only thing I would say is that firstly the
Government is not in the business of taking commercial risks so it is difficult to
get involved in the dirty business of commerce. We don’t understand markets, all
we can do is what we do do and make the spectrum available. The other issue,
to repeat what Joe said, if we find people worrying about how they are going to
finance the price of licences then, to be honest, given that the price is a small
part of the costs of developing networks, are they the right people that we ought
to be encouraging into the business? A controversial point, I don't know if Joe
I think you have made the main points and I think what the questionerTony has
said is a very valid case. Generally I agree that we mustn't put ourselves in a
straightjacket forever - there has to be some stage when we have to see whether
modifications are required to the terms of the offer if you like.
My concern is that the information memorandum that was out at the beginning of
this auction process has become a burden around your neck justinstead a way
of facilitating one particular auction.
Given the length of these licences, another point I would make is work that the
Agency is doing to develop spectrum trading. Again that is a loosening up of
what might be deemed straitjackets and hopefully we will make some movement
in spectrum trading in the next few months. That may be an issue for licensees
If someone hands the licence back then what happens to the money they have Formatted
paid? The licence says that on the surrender of a licence, no refund is available
except in accordance with Regulation 5 of the Regulations. Regulation 5 says
that the Minister has a discretion to give refunds. Can you give any indication as
to what criteria the Minister might apply when considering whether or not to give
a refund, bearing in mind that the exercise of that discretion could be constrained
by all the principles of administrative law and also the Licensing Directive?
I would suspect that we would since licence fees go straight down the Treasury’s
throat, Gordon will be very reluctant to look at whatever discretion is available in
terms of handing money back. I wonder if Joe has given any thought to that
I haven’t, to be frank. Our presumption is that circumstances will be such that
people either voluntarily hand their licences back in the knowledge that they
won’t get a refund or will have breached the licence conditions. Certainly, I am
working on the understanding that it would be wholly exceptional to make a
refund. I can’t think of what the circumstances might be where we would make a
refund , and maybe even if I could I wouldn’t really want to divulge them.
It’s there in the Regulations. It’s part of the framework and is a discretion that
must presumably be excercisedexercised - notwithstanding the reluctance of the
Chancellor - in accordance with normal principles of administrative law and
where the licences are handed back and you – the RA – get 14 years of fee.
There are some interesting precedents that might be raised.
I would imagine that certainly if somebody handed the licence back, we wouldn’t
offer the Secretary of State’s discretion. But if somebody handed the licence
back and asked for money, then we’d have to look at that particular Regulation.
And what Ministers might decide within their powers of discretion, we don’t quite
With respect to 50% of the llicenseicence fee under the deferred payment
scheme, can we assume that if licenselicences are returned before year 6 the
obligation for that 50% will no longer be in place.
There would be, that is the reason for the banker guarantee to ensure that there
is certainty of the remaining licence fee being paid.
On that note, can I wind up the formal proceedings. Please don't hesitate to ask
more questions. I hope you have enjoyed this morning's session and I hope you
have got something out of it in terms of how you are going to position yourself.
Can I thank you for you attendance, for coming into central London in difficult
conditions and can I wish you a good lunch and a safe journey home. Thank you