Blog Post

How Realistic Is BT's Fiber Broadband Plan?

Stay on Top of Enterprise Technology Trends

Get updates impacting your industry from our GigaOm Research Community
Join the Community!

Updated at the bottom: Unless you’re using Enron math, BT’s new plan to connect 10 million homes — roughly 40 percent of the United Kingdom — with fiber networks at a cost of £1.5 billion doesn’t quite add up. At today’s conversion rate, that’s about $3 billion — or $300 to wire up each of these proposed 10 million homes.

BT hopes this will help it stave off competition from rivals who have started to use their new backbones and the latest technology to eat into its broadband business. Cable operator Virgin, for example, plans to use DOCSIS 3.0 to compete with BT. The incumbent has been reticent about upscaling its infrastructure over concerns that it would spend billions and then be forced to share with upstarts, the way it does now. By comparison, the new plan is closely tied to regulatory concessions and includes some sort of investment protection from Ofcom, the British regulator.

The Guardian writes:

Under the current regulatory regime, BT must allow rival service providers to use its network on the same terms as its own retail arm. There would be a huge outcry if that “equivalence” was lost, following the battles between BT, its rivals and the regulators at the start of this decade when Broadband Britain was just an ambition.

Nevertheless, BT’s announcement is full of more holes than a wheel of Swiss cheese. Lets look at the deal from a distance: 10 million homes for $3 billion. In comparison, Verizon is spending about $22 billion to fiber up some 18 million homes. That’s a cost improvement of 9x, which means BT’s plan just doesn’t make sense, even if you take into account that somehow it will get massive sops from Chinese equipment maker Huawei.

BT plans to sell 100-meg connections to homes it will connect with fiber (FTTP) using mostly G-PON technologies. Other homes, which will be connected to special cabinets on the curb (which are, in turn, connected to the Internet using fiber), will get a top speed of 40 Mbps. So in a sense, the plan is a blend of broadband strategies being used by Verizon (all fiber) and AT&T (combination of fiber and copper.)

Having followed this business for some time, I know that neither of their strategies are cheap. Verizon spends close to $1,400 per connected home (assuming that everyone is going to sign up for the service). AT&T’s numbers are also higher than $300 per home.

According to my sources, it costs just north of $500 to get the network ready to offer households super broadband, or what is generically known in the industry as homes passed. This doesn’t include laying fiber to the home, its associated labor costs and the on-the-premise gear. All that costs between $750 and $1,000. The on-the-premise ONTs cost between $150 and $200 alone.

Given that the network is scheduled to be rolled out in 2012, let’s assume that by then, prices decline by half — but the numbers still don’t add up. It could be that this $300-per-home-for-fiber is on top of the previously announced spending on BT’s broadband buildout as part of the 21CN. But even taking that into account, I’m not ready to buy BT’s splashy announcement. I would like to know from BT the exact breakdown of the cost structure of their network.

BT’s new CEO, Ian Livingston, whom I had a chance to meet back in 2006, is a sales maven, given his background with a high-street retailer and an upstart ISP. Some say he’s so good he could sell ice to Eskimos. Of course.

Update: My good pal, Dave Burstein, who writes the influential newsletter DSL Prime, wrote in to point out why the news is spin. “There is nothing in the announcement that wasn’t discussed by Christopher Bland with Andrew Parker a year ago,” he wrote. Dave tracks the industry closely, so I’m not surprised he found the “spin” in the news. He also pointed out that by 2012, less than 1 million will be on fiber, and mostly new fiber.

And Andrew Odlyzko, the authority on broadband and networks, in an email to me noted that the incremental 100 million pounds in capital expenditure increase for this promised network upgrade is a mere 3 percent, and even that is contingent on regulatory relief from Ofcom.

Q: Is this investment dependent on Ofcom creating a new regulatory framework?

A: Yes. The right regulatory environment is vital for anyone seeking to invest. The funds required are extremely large and companies need confidence that risk-taking can be appropriately rewarded.

Image courtesy of BT plc.

12 Responses to “How Realistic Is BT's Fiber Broadband Plan?”

  1. As you’d expect there’s plenty of puffing up in the BT press release, but the announcement could be a belated but genuine first step towards fibre.

    However, of the £1.5bn promised spend, only two-thirds of it is new money, and to do everything BT promotes in its story headlines would also require an easy ride from the regulator and — perhaps more importantly — lots of public money…

    Similarly, it’s going to be principally an investment in fibre to the cabinet rather than the home. FTTH might happen at brand new sites, like Ebbsfleet in Kent or the Olympic Village, but at this stage it doesn’t look like BT’s planning to replace copper in the local loop unless local authorities partner up and pay for it.

    There’s quite a detailed look at the proposals here, if anyone’s that interested…

  2. Corrupted Mind

    OM I think we can both recognise a bit of user bloat when we see it (wink). BT’s endgame is to sew up the high-end homes in the south-east and city, the “can pay, will pay crowd” and use them plus the government to fund fibre across the rest of London. The pop math is straight forward, the SE of England excluding London is 8M. London is a v.conservative 8M (more like 10+) – if BT used its Kent bridgehead and manages to join it to the city of London (cutting through Greenwich, Lewisham and Tower hamlets as well as the new Olympic homes) they open up a market of highspeed internet of about, say close to 15/16M people – more importantly lets not forget that to the people of the southeast, in many cases BT (with either its retail or wholesale hat on) would be their sole provider therefore in a position of strength to focus on its competitors in the other london boroughs.

  3. I’m left with more questions:

    1. What is a “home”? A city or urban setting (especially in Europe) can be very dense. Hence is a “home” 1 apartment building? Or 1 apartment (unit)? If it’s per unit, one building can be 50+ homes right there. In a city environment that’s pretty easy.

    It’s getting the rural areas where you have to deploy fiber several miles to reach 1 home that it’s gets costly.

    2. Are they leasing capacity? They could make considerable income by leasing some unused capacity on their network to others and subsidize the cost. They will only need a small percentage of the fiber in 2012. They could arrange for a 5 year agreement on transit from metro A –> metro B. To help cover the costs. This brings you much closer to having high utilization at day 0, and would drastically cut the time to recover the investment cost.

    3. Is Verizon a good comparison? Verizon’s got a whole bunch of different issues than BT’s going to face. First of all, I believe most of VZ’s labor is unionized, and has all the complexities and costs associated with that. Secondly, the US is much more spread out. Verizon is gunning for Comcast’s hot markets. Often those are rich suburbs (spread out), and can be spread apart quite a bit themselves. Britain on the other hand is known like most of Europe for it’s density (UK ranks 51 compared to 180 for us for density). Not to mention Verizon started a few years ago, BT is starting later. Surely VZ’s total cost is higher due to earlier adoption.

  4. @Corrupted Mind… good points you make, and thank you for that. Just to clarify, are you saying that those areas are 40% of the UK households. If what you say is true, I wondering if the guys from BT are kicking sand in Ofcom’s eyes and “giving concessions”. Advise.

    @ Andrew,

    They are pretty explicit in saying they are going to be building FTTP connections. I think Ofcom relief is contingent on that. And If the whole thing is entirely VDSL, I am not so sure you can eliminate the costs – labors, modem and what not entirely. I think even 10% of those homes being fiber makes the cost estimates questionable.

  5. I’d guess that the 10M are city dwellers. Modern British cities still have large areas where offices and apartments are adjacent. (the office next to ours changes from apartments to offices every few years).
    BT (and the late NTL) have been providing ISDN services to businesses by running fiber to the cabinet
    for years. (We have both sorts of fiber into our building).
    BT is probably just going to start offering services over this pre-existing fiber – about time too,

  6. Corrupted Mind

    Methinks you have grabbed the very end of a wonky stick. Basically, this blog post illustrates a lack of local knowledge, BT’s plans focuses mainly on the legacy olympic homes -that will be pre fitted with fibre – the densely, pre fitted (and densely populated) area from canary wharf to the city, and eastwards to Kent and its highspeed pilot. This project has the look of a “join-up and upgrade” of possibly the most densely pre-fibred parts of London and the southeast. Admittedly, upgrading the exchanges will be pricey but its hardly starting from scratch, like Verison or AT&T. I would be surprised if they overran budget.