Reaching out for fibre

Published on:

It's the ultimate outsourcing move in the industry: BT's local loop business is now run by a separate unit, Openreach, which offers equal access to all competing operators. Now it is exploring a strategy to build fibre into the local access network. CEO Steve Robertson reveals to Global Telecoms Business that the company will develop a business plan for fibre. Interview by Alan Burkitt-Gray

Steve Robertson: Openreach is a very healthy business. We generate cash. We've got very good cashflow

Steve Robertson is engaging and loquacious. It's still less than a year since BT outsourced its local network to Openreach, which he heads. The purpose is to operate them, in the old cliché, with a level playing field.

The official term is "equivalence of input". It means that no carrier, not even BT itself, has preferential access to Openreach's services or its 30,000 staff. It's a bold move, carried out by BT in the face of a threat by Ofcom, the UK regulator, to refer the incumbent to the competition authorities.

Almost all the 30,000 staff used to work for BT: just a few of the most senior have been hired from other companies in the industry. Now they've been trained to say "Openreach" when a customer answers the door — a customer, that is, of BT or any one of a number of competitors. Openreach's aim is to work for all telecoms operators equally.

It's a move that seems to have the support of Europe's über-regulator, Viviane Reding, the European Commissioner responsible for telecoms and the media. She and Robertson will share a platform at a Global Telecoms Business conference in April 2007.

They — and other industry leaders — will be talking about broadband in the access market. Throughout the world users are demanding more bandwidth. Fibre will be key. According to this interview, Robertson and his team at Openreach are now developing a business plan to build fibre into the UK access network.

There's been no formal announcement yet, and the business plan is still embryonic. But they're working on it. You can read it here first.

Openreach will be a year old in January 2007 and already questions are starting to form. How is it performing — and are all operators in the competitive market getting equal access?

Most operators say yes, though there are some exceptions — notably Cable & Wireless. Most appear to be pleased with the access they're getting, though it's been a difficult year. Openreach has been focussed on separating out its systems from BT's, so that no customer-facing BT executive can see what Openreach is installing for a rival telco, for example, and building an Openreach corporate culture.

Rebranding the vans

When engineers knock on doors, to install services for customers of any number of companies — Colt, Tiscali, Easynet, BT and the others — they announce themselves as Openreach. The vans are all being rebranded with the Openreach name.

And it's been a busy year for local loop unbundling and for the broadband market generally. Openreach's staff in the telephone exchanges have been kept busy disconnecting — with wire, not software — customers from the BT network and connecting them to the systems of other operators.

There's a shortage of people to do it and, say some, a lack of physical space in the exchanges at times to do the job.

Robertson has set up Openreach's headquarters in one of those telephone exchanges, at a respectable distance from the rest of BT's corporate headquarters: there can be no suspicion that staff can meet old BT mates around the watercooler or in the local pub or in Starbuck's and accidentally drop information about competitors. Openreach is being ultra-scrupulous and wants to be seen to be ultra-scupulous.

He was appointed to be chief executive of Openreach after a spell as managing director of operations at BT Wholesale, where he was responsible for the overall performance and operation of BT's network in the UK. At 48, he's spent most of his career in BT, but he had an extended spell with rival pan-European operator Colt Telecom.

We talk at first about the nuts and bolts — plugs and sockets might be a better term — of what Openreach has been doing since it was formally set up in January.

Local loop unbundling has boomed, though "personally I would like to have seen even faster take-up of LLU", says Robertson. "The take-up in wholesale line rental has been really brilliant — I'm delighted about that."

And the wholesale line rental figures imply there's a filling reservoir which indicates that LLU is going to speed up, he says. "I don't think we've seen the peak of this. The quicker the better, because the process of migrating customers into the LLU environment helps us to bulk up the infrastructure. I would like to look back in six months time and see it accelerate even more."

His customers have changed in the same period. Orange, owned by France Telecom, has launched a converged service of mobile and broadband. Carphone Warehouse, a mobile service reseller, launched spectacularly into broadband. O2, the mobile operator owned by Telefónica, has bought a broadband company, Be. Tiscali has bought IPTV operator Video Networks to create a converged offering. Vodafone is to offer home broadband services to customers. Satellite TV company Sky has bought Easynet and is now offering broadband services. BT Retail, for it is just one among many customers now, is launching IPTV later in 2006.

And Robertson is pleased that Openreach hit a significant target on June 30, when BT Wholesale also became an LLU customer. Equivalence of input means that the unit of BT, run by Paul Reynolds, that is running the 21st Century Network project must now use the same systems to get local loops as the rest of the industry.

"We can say we have an equivalence of input system. That was really hard work. It was really, really, really hard work. We consulted with industry. We did the development at breakneck speed. It was a massive effort. And we hit the date. It helps us build confidence in ourselves."

Executives in the market suggest the initial impact of Openreach's arrival has been a levelling down of the service. In other words BT customers got preferential access in the past and now all are waiting a similar period: another pressure on Robertson.

Twin market dynamics

He describes convergence and consolidation as "the twin market dynamics". There will be more mergers, and more companies will extend their range of services, he thinks.

"I haven't spoken to any directly, but it wouldn't surprise me because everyone's looking for a competitive edge. Infrastructure players can move aggressively into service provider spaces where they can start looking at content players. Content players can think of getting into this world too. You'll probably have to think of changing the name of your magazine, won't you?"

Interesting idea, but we leave it well alone. But what's all this consolidation doing to the market? "When you've got lots of businesses moving into new spaces some will be successful and some will not be successful, for all sorts of reasons, and that can be everything from operational issues to changes in company strategy. It's far too young to expect stability. It's not something that particularly bothers me."

Look at the big picture, he says, and some characteristics are beginning to emerge. "There are some service providers moving into the content world. There are other service providers which are expanding a converged connectivity portfolio — like Carphone Warehouse. There are other service providers with an existing media business looking in the opposite direction. You can see these emergent models starting to happen."

People are dipping their toe into the water, he says. "Some things are going to fly. Some things will be more challenging."

And all of them want to use Openreach — or most, because Virgin Media, the new name for the merged cable operations NTL and Telewest following its takeover of Virgin Mobile UK, owns its own local broadband infrastructure, built at great cost in the 1980s and 1990s.

The state of the copper

So what's the physical state of the local loop network that Openreach has, and on which so many businesses are depending? How much of it is up to the job of being able to deliver the real super-broadband services that people are talking about and how much will need renewing over the next few years?

Robertson looks a tad uncomfortable. This is where we're getting to the heart of what Openreach is all about. Officially the company is responsible for the copper — well, pedants say "metal", because a percentage of it is aluminium-cored cable laid in the 1970s as a cost-cutting move and now turning to aluminium oxide — between the main distribution frame in the exchange and the customers' junction boxes. Not optical fibre, just metal.

Yet just as Openreach has been created half the telecoms world is looking for ways to move fibre deeper into the local access infrastructure, and the other half is thinking that maybe wireless — either cellular, mobile technology or new fixed technologies such as WiMax — might replace copper in the "last mile".

So is Robertson merely the custodian of an obsolescent distribution system, the rutted cart tracks or overgrown canals of the information age, while all around its customers are planning new superhighways with broadband local networks based on fibre and wireless? These are thorny questions, and Robertson recognises that.

"It depends on the thresholds," he says in answer to the question about the current state of the network. "It's a multi-dimensional issue because it isn't just an issue of the copper, but of what technology do you put on it? It's not just an issue of what technology you put on it but of the tolerance of the applications that you put over that technology."

Deep breath. "And it's there's also a huge variation depending on the type of modems, the customer equipment, at the end of this. It's a very multi-faceted issue."

But ... ? "At the moment it's still best endeavours, so you get the copper that you get. As long as it can support a basic voice service then you add what you can add and you get what you can get from it."

That's the broadband environment that we've grown up with, he says. "I think you have to question whether that is always going to be the situation."

But presumably people are questioning it and are starting to look at the alternatives? "To be honest, in terms of my customers talking to me right now it's not a big issue. It's not a big issue." People are not worrying about the state of the copper, he says.

He is, though, looking over the horizon. "It's an inevitable consequence of the evolution of the market. If we are moving into a converged marketplace the characteristics of the bearer are very important."

Higher broadband speeds

Let's push a bit harder here. A decade ago everyone was thrilled to get speeds of 56k. Now many think one megabit is slow. "Yes, it is slow," he agrees. "So from 56k to a meg, to two megs." The dynamics are "largely going to be driven by the end user", he says. He accepts that users will want higher and higher broadband speeds from their providers.

"Right now I think the state of the basic infrastructure in general, in general, if you take a snapshot, is probably adequate for what's being used."

In many cases applications to consume all this bandwidth need to catch up, he suggests. "We know what human psychology is like. I drive a car that is capable of going much faster than I ever really want to go, but that's the car I want. I think there is a state of the market now where I would like to have a very fast broadband link but I might not be entirely sure what I want to do with it," says Robertson.

So the real pressure is not necessarily there at the moment, he suggests, "but that will come along". And that means "if we're not having the discussions just now, it's only because we've postponed them for a while", he says. "The discussions will happen, without any shadow of a doubt."

And then? "When we get to that discussion it is going to be about the combination of technologies that allow things to operate effectively," he says. "In the mean time, there are a lot of very basic things that we're doing."

In the first six months of Openreach, "we've been looking very heavily at basic network reliability", he says. "If we open up the network, let's make sure we always close it up again properly."

The development of LLU and the demand for broadband does mean that there is more work going on in the network than before. "If we look at some of this infrastructure that has been around for a long time, there are certain types of joints that are sealed with resin. Let's reseal them with a modern compound that doesn't deteriorate and doesn't leak. There has been a lot of stuff like that, which is basically groundwork preparation. There are some housekeeping things we've been able to address much more effectively as part of Openreach than we could do when we were all separate parts of BT."

Is he doing that progressively through the network? "We're doing it every time we touch the network," says Robertson. "We are looking for opportunities to improve the quality of the network."

But he then introduces the big question: "Do we run out of road in terms of the existing infrastructure? We have to think ahead."

There are questions Openreach needs to look at: "Things like: are we going to take fibre to the cabinet? As an example. That's potentially a huge investment. Are we going to have active MDFs? If we take fibre to the cabinet, do you still do active MDFs?"

An active MDF? What's that? "It turns the main distribution frame into an active network component. In this market, where we have lots of service providers that are LLU operators, and customers want to change services and service providers, the systems are there to do that on a zero-touch basis," he says.

Up to a point. "But at the heart of this, there is physical intervention, because you take a pair of wires from one place and put it somewhere else."

The main distribution frame inside a telephone exchange has historically been quite a static thing, he explains. "Because you were connected, and you were connected, and you didn't change service provider every five minutes."

Systems to handle churn

But now there's a market. "Consumers will have options and smart companies with lots of great ideas will be selling them things, and they will want to change from one company to another."

That means Openreach has to have a means of handling the churn, migrating customers from one service provider to another "as efficiently, effectively, seamlessly as possible". Maybe Openreach can continue to do this manually, but clearly Robertson is doubtful. "These are the sort of technology issues."

And that means Openreach has to create a business facilitating operators' churn. How much does it charge for the service? "Of course, of course," Robertson agrees.

So what are the conclusions so far? "It looks like, pretty much, if we invest money in Openreach we get a pretty decent rate of return."

Openreach's figures are identified clearly in BT's accounts. "It is a profitable business," says Robertson. "They're published as a separate line of business."

Figures published after our interview for the six months ending September 30 2006 show that Openreach generated total revenue of £2.53 billion and earned an operating profit of £577 million. Its capital expenditure was £550 million in the same period.

"It is a very, very healthy business," says Robertson. "We generate cash — we've got very good cashflow — and we're more or less in the right ballpark for a regulated rate of return, which is 10% return on capital employed. The regulatory accounts are based on current cost accounting. If you value our assets at the current value, we're making a healthy return."

So when he and his colleagues start to think about investment strategies, in active MDFs or anything else that they might want to do to the network, would Openreach be able to do that from the existing revenue? And would Openreach be able to go to different sources, outside BT?

"Our funding is always going to be part of the BT group funding," says Robertson. "I'm not going to set up a treasury department or anything like that. It will be part of the BT funding."

New capex strategy

That means the assessment of Openreach's capex strategy would be by the BT board? "By the BT board," agrees Robertson. "The process for this is quite transparent." He develops the business plan with his executive team. "It has to be agreed by the main BT board."

However it also is shown in its entirety to the equality of access board. This is a committee of the BT board but one that has responsibilities to the industry and to Ofcom to ensure that all parts of BT are abiding by the terms of the settlement of the strategic review which led to the creation of Openreach.

"Everything that the BT board sees, the equal access board sees too," says Robertson about his business plan.

The EAB doesn't have authority to change the plan, "but it is visible, the investment levels are visible, what we need to do to succeed as a business is visible".

The board is chaired by a non-executive director of the BT group — currently Carl Symon, formerly a senior IBM executive. There is one other BT person, Sally Davis, who has worked in the cable industry and in Cable & Wireless's former unit, Mercury Communications, in the UK as well as in Bell Atlantic — now Verizon — in the US, before joining BT in 1999.

And there are three independent members: Bryan Carsberg, the UK's first telecommunications regulator from 1984 and now a professor of accounting; Stephen Pettit, who has worked at Cable & Wireless; and Peter Radley, formerly a senior executive with Alcatel.

Robertson makes the point that "the things that we will need to do to support industry as a whole are clearly essential for the success of other bits of the BT group as well". He recognises that the BT group, like any other business, has to make decisions about priorities. "As long as we run Openreach successfully and sensibly, there is no reason for it to be disadvantaged compared to any other line of business in BT in terms." It's part of his job to ensure that continues to be so.

He didn't say so, but it must be in his mind that at some future date there might arise conflicts about BT's spending plans. How will the BT main board determine its responsibilities to shareholders when it has to allocate capex between customer-facing parts of BT, such as BT Global Services or BT Retail, versus Openreach, whose activities provide infrastructure for BT's competitors. It almost inevitably will become a difficult question. The implications must be that the current status of Openreach as an integral part of BT will be questioned.

But that's in the future. Let's get to specifics about the business plan that Openreach is developing. There is continued pressure, we have agreed, for higher and higher bandwidth. Does Openreach have a plan that by, say 2008, or 2009, or 2010 some homes in the UK would want a certain bandwidth, in terms of megabits, and how is Openreach going to provide that to its own customers, the service providers? And what are the technical opportunities?

Critical decisions on fibre

"I think that one of the critical decisions is, how deep do you take fibre?" says Robertson. "Do you take fibre to the kerb, do you take fibre to the home? Or do you not? Do you bond copper, and all that malarkey? That's one of the critical strategic issues."

At the same time there's a race between the amount of capacity "and the efficiency with which end user equipment can use that capacity", he notes. Developments in codec technology means that high quality video needs less capacity than it would have needed five or 10 years ago. "How much capacity do you need to achieve a given task? That's changing all the time. The tasks that have got to be achieved are changing all the time."

So on the one hand there's a move for more and more bandwidth, while on the other hand there is better and better compression, so it's possible to do more with the bandwidth that's available. "That dynamic has to be taken into consideration as well," he says.

Where does that leave us? "Having said all that, fibre has to come into this picture," says Robertson. "That's my opinion. Fibre has to come into this picture in some shape or form. That is one of our critical strategic issues."

The two big operators in the US are doing fibre in quite different ways. Verizon is building fibre into every customer's premises, at great expense. AT&T is planning fibre to the kerb, "to the nearest flexibility point in the network to customers", says Robertson. "Then they're reconditioning copper, or whatever they need, to offer 20 megabits, 24 megabits, something like that."

And, at last, he volunteers a number, a target for Openreach's bandwidth, though by indicating agreement with AT&T. "That's the ballpark that, you know, roughly speaking, I think you have to expect to be in that space. How we get there is something we're going to have to resolve fairly quickly."

That will lead to questions about the reliability of equipment on the kerb. "There are a number of technological options we'll need to look at," he says. "There's a whole bunch of stuff that needs to be put into this."

One thing is for sure, he says: "We are an infrastructure business." He hesitates. He's choosing his words very carefully at this point. "The investment we make, the investment decisions we make, it's very important we get right. It isn't a business that is short term and you make all your money back in six months, so it doesn't really matter. This matters a lot."

It's clearly important. Let's try to get an idea of the timescale of all this. When will he have a firmer idea of how Openreach will go ahead?

"I think it's the next couple of months," he says. Oh, that quickly? After all those comments earlier in the interview about postponing decisions on faster broadband "for a while" and his remark that this is "something that we have yet to explore properly", such a short timescale comes as a shock.

Deciding now

He fills in a bit of detail about this schedule. "By the time we get to the end of this calendar year we have to start firming up our view. We have to be in that position."

And, he points out, Openreach is already running trials of fibre to the cabinet. "There are various other initiatives we're looking at, commercial initiatives — fibre to the premises. We're not starting with a blank sheet of paper on this at all."

A lot of people have suggested that Openreach would be a model to export to other parts of the world. It's widely believed that Commissioner Reding in Brussels favours the idea. "We've heard various noises" from the Commission, he admits.

"Just as we are looking out, others are looking in, and we've had a lot of interest from other parts of the world. My reaction is the same as when I look at them: it's horses for courses." But there are some principles — higher bandwidth for one. "And if you truly have got an economic bottleneck and you want to encourage sustained infrastructure-based competition, then in terms of the Openreach model, so far so good." It's not that the job is done, he concedes.

It is a very different model from that which still applies in much of the world. In many countries the incumbent — whether it is privatised or still state-owned — tends to be regarded as a national treasure to be protected against competition. In Germany, Deutsche Telekom wants a regulatory holiday before it will invest in broadband, and in Australia there are similar requests from Telstra. Maybe Openreach is a way of keeping the local loop as a national resource while allowing competition at the services level?

"Across the world people still have their national airlines and there's a big emotional attachment. British Telecom is British Telecom. You can expect an element of that. But you have to ask if true open, competitive markets end up delivering something better for the health of economies. If they do, you know what? It happens. It does take a lot of time, and sometimes it's painful. Sometimes it needs legislation, but breakthroughs happen, and they happen in unusual ways."

It is within living memory that people got their black Bakelite telephone from the telephone company. "It's not that long ago. But markets will continue to open up. It's not a smooth process." The underlying economic imperatives will prevail, he believes.

There's something else that has changed in the past six months, says Robertson, speaking "from Openreach's point of view, the worm's eye view if you like."

Does he feel that Openreach has established its credibility in terms of equivalence and being separate from BT? Do they share things with the unit?

"I think it's a journey that we have now moved a few stops down," says Robertson. "We're not in any way near the end of that journey, but if I compare the quality of our relationship six months ago, it's a much higher quality now. I use the words carefully, because that doesn't mean it's a less tense relationship. As you get to know each other better, and you get more interdependent, then you become more intimate. The depth and quality of the relationships, I believe, in general, are much better. I'm probably the wrong one to judge."

Viable business

One of the things he's most pleased about, since Openreach was set up, is that it has identifiable figures in the BT results. "One of the ways of giving confidence that we will continue to invest is to be able to say we are a viable business. We are not a drain on the rest of the group or a poor relation."

It is, says Robertson, "a business in transition". When it was set up in January "it was only the start". It is "the vertically integrated telco" — BT, in other words — "taking itself into its constituent parts to ensure that there is a level playing field".

The next target date is at the end of December 2006, when BT Retail itself starts taking wholesale line rental directly from Openreach.

"They will be buying the same lines product as everybody else. There won't be PSTN and wholesale line rental, where your wholesale line rental is what you sell to other service providers, and PSTN is BT gets. There'll be wholesale line rental, and that's what everybody in the UK will be consuming. It's a huge thing for us. You just cannot overestimate what we're doing here. It's a huge thing."

What are the parameters for judging performance in three or four years' time? Is it number of broadband connections, the economy of the country, or the value of Openreach in BT?

"I would like to see a thriving service provider industry which is delivering a full set of converged services to consumers and to businesses. That's why we're here. If we don't do the right thing we are capable of screwing it up. We can't guarantee success but we can guarantee failure if we screw it up. So far, so good, but it is only so far."

And if he does screw it up he'll have Viviane Reding to contend with. GTB

Latest Openreach results

Second quarter ended September 30 2006

Half year ended September 30 2006



Better (worse)









External revenue







Revenue from other BT lines of business














Operating costs before leaver costs







EBITDA before leaver costs







Leaver costs









Depreciation and amortisation







Operating profit







Capital expenditure







*Restated to reflect changes in intra-group trading arrangements.

Source: BT Group results