Ofcom: Oi, BT! Don't be greedy – feed dark fibre to your rivals Britain's communications watchdog wants BT to open up its fibre broadband network to rival telcos. The one-time national telecoms giant responded negatively to Ofcom's proposals, which were floated on Friday. BT claimed that "mandating dark fibre" could "disadvantage" ISPs that do not have the "capability to deploy" the …
Hmm. There is an element of truth in that, but it's not the full story.
From my own experience I can agree with the bad customer service statement, although I have, on occasion, also had good experiences.
But I can also say that once set up their network provision via FTTC has been excellent, fast and stable.
So I'm actually cautiously optimistic about the prospect of TalkTalk running their own fibre network. At the very least it introduces some competition and price pressure into the market, which is something TalkTalk have historically been good at doing.
"TalkTalk do their own independent backhauls "
It's possible to connect with a range of telcos in BT's exchanges - the only limiting factor being if the other telco has actually got some network into the site. Verizon, KCom and Colt all offer services like that. I doubt the speed or reliability is higher, what might be true is that achieving a certain speed or uptime is cheaper than it would be with BT.
I don't know what to believe anymore.
Well it is a bit obvious, Talk Talk want infrastructure yesterday and if they can get someone else to do all the paperwork and legwork involved in physically laying cables so much the better for Talk Talk...
I will be interested to see what the other owners of major fibre infrastructures such as Vodafone (previously C&W), Interoute and Colt will say, as this proposal will impact their business'es.
This post has been deleted by its author
In addition, IIRC BT have favourable rules about business rates on fibre. I can't remember all the details, but I believe it was something along the lines of BT have preferential rates, along with only having to pay for lit fibre, where others have to pay for all fibre lit or not.
(Note as I say, I can't remember the details, but I do remember that there was an advantage along these lines)
"why aren't 'the rivals' investing in dark fibre networks of their own"
Because BT have some history of not fully utilising their assets, then when somebody proposes to offer a separate solution they'll move to undermine them using the assets previously left to languish.
You'd have to be mad to invest in your own new fibre if there's any dark fibre making the connections you propose to build, because you could sink millions into your project, the dark fibre suddenly gets put into service, and then your traffic modelling and whole business plan disappears up Brown Creek.
"Because BT have some history of not fully utilising their assets, then when somebody proposes to offer a separate solution they'll move to undermine them using the assets previously left to languish."
AKA "leveraging a market monopoly" and "anticompetitive behaviour"
It's totally illegal, but ofcom have been letting them get away with it for years.
Stupid question here but why aren't 'the rivals' investing in dark fibre networks of their own...
My guess is because BT is able to deploy dark fibre much cheaper thanks to all the ducting...
Dark fibre == fibre you have installed that is not lit with communications. If you install fibre and are using it, its just called fibre.
BT installed masses of fibre throughout their network that is not lit up, because installing a bundle of fibre cores is similar in cost to deploying a single fibre, and only lit up what they needed, eg 8 cores out of 48.
This is to mandate them to allow other communication providers to light up the cores that BT are not using at a rate that is considered "fair", rather than forcing those providers to buy IP services from BT at whatever rate BT think they can get away with.
Dark fibre means providing the customer with an end-to-end fibre strand which does not traverse any active equipment. Therefore it is lit by the customer's own equipment at both ends, and the customer can run it at whatever speed they like, only subject to the physical limitations of the fibre.
This is as opposed to a managed service where you are facing into a piece of telco equipment, and the telco charges more the higher the data rate you buy.
BT's rivals have been investing in their own dark fibre networks - the challenge is that they don't usually put significant capacity into BT exchanges as they are owned by BT. In general, the fibre goes in where the customers pay for it (speaking as a customer who regularly pays for new fibre installs almost every time one of our offices move).
BT on the other hand, developed the network of exchanges and ducting largely while it was a state owned business. While that might not have provided everything required, it was a very significant head start. If I want to use vendor X, they probably have their own fibre or will install their own fibre in most of the large metropolitan areas, but for the remainder of the country you are stuck with BT.
Personally, I think BT OpenReach should be sold off as a condition of BT buying EE - it would help address the issue of putting equipment and ducting into existing exchanges.
but why aren't 'the rivals' investing in dark fibre networks of their own
Answer: Time and money!
Building out a network requires capital upfront and then locks it up for the long-term - just ask: C&W, Interoute, Colt, Virgin Media, to name a few who have invested in infrastructure. However, being able to piggyback on someone else's network is significantly cheaper in CapEx terms, even if over the long-term it may have a higher OpEx.
Which bit of BT owns the dark fibre? OpenReach or someone else?
I wonder if they segregate their dark fibre between the stuff that's used for exchange connectivity versus more generic backbone type stuff.
Either way I suspect they are using it to throttle demand to keep prices high.
"Which bit of BT owns the dark fibre? OpenReach or someone else?"
I don't think any part of BT owns it today because it's not a product BT sells.
I think the way the network is divided up is that Openreach own the bit from local exchanges out to customers, and the bit from local exchanges into main exchanges. The bits linking the main exchanges belong to an internal only part of BT, but BT Wholesale can sell those bits to other telcos.
A recent conversation I had with a wide area network vendor, who is by the way installing their own dark fibre, lead me to believe that any spare dark fibre BT has, either does not exist, or they aren't even using it themselves, as I was informed, BT's fibre network is, in many places oversubscribed. (Pinch of Salt?)
But having worked in a large technology company where assets were not released, just in case someone important wanted one, I can well believe that one part of BT isn't releasing dark fibre, just in case someone important needs it. It will be very interesting to see the outcome of Ofcom's ruling, I can't think of any outcome that would surprise me.
If BT really do have a lot of spare Dark Fibre capacity, then it may very well have a catastrophic impact on smaller competitors, as the wholesale prices will drop considerably.
"Which bit of BT owns the dark fibre? OpenReach or someone else?"
Due to regulatory failure at OFCOM, Openreach isn't a separate legal entity from the rest of BT, so Openreach technically own nothing. A cynic would say that this is VERY convenient for bamboozling regulators, competitors, and politicians, and enables a multitude of sins to be hidden, such as the margins being made by Openreach, the cost of internally provided services and the like.
Consider BT are installing fibre in a duct and need X fibres to meet current and imminent demand. A beancounter looks at the job and decides that X+Y fibres should be installed, weighing the cost of installing that spare capacity and the cost of capital used against the potential profit raised by selling services using that fibre and the probability of that fibre actually being needed. If you are selling fully managed service the potential profit is nice and juicy, providing an incentive to make Y reasonably large in the first instance.
You then change the equation to make those potential future profits smaller - instead of fat profits on service you get much smaller amounts renting out bare fibre. The return on investment calculation shifts in favour of not installing so much spare capacity to begin with. In the long term that doesn't benefit anybody.
I'm not a massive fan of BT but equally I don't have a massive axe to grind about them - I'm simply looking at this from an economic standpoint. BT are a telecommunications company that makes money from selling service. They are not in the business of renting out bare fibre for chump change.
"The return on investment calculation shifts in favour of not installing so much spare capacity to begin with. In the long term that doesn't benefit anybody."
You'd be right if dark fibre cost serious money. It doesn't, because the cost of the actual fibre is a tiny part of the installation cost. Not only is the actual fibre optic cheap, the marginal costs of installing additional capacity when you're already on site to the job are very low indeed - another couple of hours on site when you've already got the people and equipment there.
"You'd be right if dark fibre cost serious money. It doesn't, because the cost of the actual fibre is a tiny part of the installation cost. Not only is the actual fibre optic cheap, the marginal costs of installing additional capacity when you're already on site to the job are very low indeed - another couple of hours on site when you've already got the people and equipment there."
I think you're missing the OP's point.
If any spare capacity that gets installed can be required to be sold at low cost to a competitor, it makes an awful lot of sense to no longer install spare capacity. The regulation acts as a disincentive.
"If any spare capacity that gets installed can be required to be sold at low cost to a competitor, it makes an awful lot of sense to no longer install spare capacity. The regulation acts as a disincentive."
Indeed. Even if the extra fibre only adds a few pounds to the install cost, why are any company going to spend that money just so the fibre can be essentially taken off them and sold for peanuts to a competitor?
I am not a massive fan of BT (quite the opposite, actually), but I don't really see how this is going to benefit anyone apart from other ISPs, and I think this action will actually negatively impact the consumer in the long run as it will dissuade ISPs from investing in their networks in the long term. After all, why bother to invest in brand spanking new hardware, and running new fibre everywhere when they can just bitch to Ofcom and get the fibre for next to nothing? Also, it will no longer be in BTs interest to invest heavily as they know they'll just get slapped for it by Ofcom before they can turn a profit on it.
"If any spare capacity that gets installed can be required to be sold at low cost to a competitor, it makes an awful lot of sense to no longer install spare capacity."
And this is why the de-facto monopoly must not be in a position to be selling at wholesale AND retail.
Openreach must be cleaved off. BT has been repeatedly noted to be abusing the market, but Ofcom keep pretending it doesn't happen.
"If any spare capacity that gets installed can be required to be sold at low cost to a competitor, it makes an awful lot of sense to no longer install spare capacity. The regulation acts as a disincentive."
It makes more sense to install the same capacity but just redefine it so it's no longer dark or 'spare'.
If any spare capacity that gets installed can be required to be sold at low cost to a competitor
I suspect you are jumping ahead of yourself. Firstly, Ofcom are talking specifically about "dark fibre" ie. fibre that is unlit and also unterminated(?). Spare capacity is something different, although dark fibre might be considered to be spare capacity. All BT have to do is to connect up their spare fibre to some live equipment and occasionally use it for something and it is no longer neither spare nor dark, it is just under utilised capacity.
The other big issue about this is duct capacity. I've had to deal with many issues in the past that are simply delayed due to certain parts of bt's network not having new capacity. Then theres the issue of old bundles that have had faults so they have respliced / rerouted around the problem.
Now if all of a sudden a lot of new players start buying up segments of fibre they are most likely all going to want it in the same general high population density area's.
Just how much spare dark fibre is there to support this demand? What will people say what faults or new provides take a lot longer when there is no new capacity and its going to require a totally new fibre pull/duct installation to simply provide their service or fix a fault?
The other big issue about this is redundant routing of trunks, or the lack of.
Currently, it is relatively 'easy' to confirm that a circuit from BT takes a different duct route to one from another provider. I can see with this new directive in place, life will get a lot more difficult. And many companies will only discover they didn't have alternatively routed trunks when another Holborn happens...
Another big issue is going to be end-to-end connectivity, as BT's dark fibre will be between BT premises and cabinets. Hence a third-party is probably going to have to use some creativity to adapt BT's network to their needs and may find that there is insufficient space capacity in parts of BT's network to enable them to build an end-to-end connection.
Openreach is deliberately restricted in what it can do by BT head office - artificially restricting supply means that prices can be held high.
Once it's a separate company the brakes can come off development.
This is what happened in New Zealand - and the separation there was based largely upon observations of BT's ongoing market abuse over the last decade since Openreach was created.
Predictions were that the newly separated lines company would be a dead duck(**) but it's now highly profitable and responsive whilst the remaining telco is in trouble.
BT is possibly concerned the same thing may happen. At the moment everyone deals with "BT" because they have to, but given a real choice of supplier there's no guarantee that things will stay that way.
(**) This was thrown about so much that many shareholders of the newly split off company dumped their shares and the price dropped 30% in the first week of operation. Those who held on (or hoovered up the sold ones) are making a nice income tyvm.
New Zealand's mean broadband speed is lower than the UK's;
http://www.netindex.com/download/allcountries/
The subsidy from the NZ government ($1.5Bn) looks to be a lot money in a country of only four million inhabitants - that's about £300 for every man, woman and child, give or take.
Just two ISPs have more than 75% of the market in New Zealand - one of them the former incumbent;
http://au.pcmag.com/networking/23456/feature/fastest-isps-2014-new-zealand
There might be a better model of last mile ownership somewhere in the world, but I don't think it's the New Zealand model - if those articles are right, there's less choice, lower speeds and a £300 tax bill to pay.
I think a public ownership model might work better - access to state funds to build and no profit margin to keep wholesale prices low. That would tend to damage the business model for Virgin and others though - building your own network is a bad call if you can lease space on someone else's for less money. Maybe the answer is to incentivise infrastructure investment amongst BT's rivals by having the regulator raise their prices instead of keeping them low?
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
