DSL Prime: British Telecom Considering 100/100 Mbps

Telcos that face real competition will deploy, and BT has to worry about new startups fuelled by true line sharing as well as a potential wireless broadband competitor called James Murdoch.

by Dave Burstein of DSL Prime and Future of TV [August 3, 2006]

British Telecom considering 100/100 rather than 15/1
Is the cost too high?
British Telecom at the highest level is reconsidering going to the curb/basement with high speed VDSL, but still are afraid to commit the capex. They haven't decided to change plans, but are "looking at the problem nearly every day." My take is the risks are bigger not going to the fast network, but many executives and investor blanch at the billions they would have to spend for a full buildout. Most UK field units have neither fiber nor power, so going to the remotes might be more costly than at AT&T.

Except for the last mile, BT is going at breakneck speed to build the best telco network in the world. Starting this fall in Cardiff, South Wales, they are moving everyone to VOIP delivered over DSL, and literally "switching off the PSTN." The timetable is to do the entire network (30 million lines), in five years. The projected savings are enormous and Reynolds tells investors the return is so fast they are ahead within a year or two. The new equipment is so much cheaper to buy, they are barely raising capex.

Reynolds claims savings so high I was thoroughly skeptical until last fall at USTA. Bill Smith of BellSouth quietly outlined a network transformation nearly as dramatic and as rapid as BT. Mark Wegleitner at Verizon projected network savings even higher, and not just from his fiber last mile. Paul Lacouture of Verizon earlier last year had confirmed "essentially, soft switches are now so cheap we can pay for them just from the maintenance savings from dumping the old Class 5's." Chris Rice of SBC and Balan Nair of Qwest are less ambitious in their plans, but agreed with their peers that savings of 30-60 percent were attainable.

Verwaayen's plan has been to use those savings to finish the "21st century network" without raising capital spending. They haven't been willing to jump to the 50/10 Mbps of Germany or the 100/100 Mbps of Paris, Amsterdam, Korea or Japan. Reynolds and others are been fascinated for years by the possibilities of much higher speeds. CTO Matt Bross was a fiber guy at Williams before he joined the Brits.

BT's tech people have played a leading role in the standards for FSAN networks, and I'm sure have drafted a detailed plan to run fiber throughout the country. So far, they've decided the cost of digging up the streets for fiber is just too high, would take longer, and has uncertain public sector support.

Ben Verwaayen's first point in the quarterly call was "I would like to draw your attention to the last one, the free cash flow. It is £100 million better than a year ago" (transcript at seekingalpha.com, a very helpful site). It will be very hard to tell The City that cash flow needs to be sacrificed for several years if they make that decision.

Carphone, BSkyB, O2, and others will use TV and mobile revenues to cross-subsidize the new networks. Over a billion dollars is already committed, and possibly over a billion pounds. They need to grow very fast, and will make offers that will win customers.

Carphone Warehouse was the first to go after BT aggressively, and signed up 350,000 DSL customers in two months. The deal's not bad, bundled with a reasonable mobile phone package. The service to date has been so bad even CEO Charles Dunstone calls it a "nightmare," but 10,000 to 15,000 more join every week. Ovum's Michael Philpott believes CW is making some headway, but one of the best informed in the UK tells me CW's actual service may be even worse than reported.

Now, James Murdoch is putting nearly a billion dollars on the table to win 3 million lines in four years, bundled with satellite TV. Both will reach 50 percent of the country very quickly, and probably continue unbundling to 70 percent or higher. Bulldog and Tiscali also have facilities, FT/Orange and O2/Telfonica want into the game, and cable is now nationally unified and testing 100 Mbps service. BT needs to find a competitive advantage rather than count on customers being too lazy to switch. By this time next year, it will seem like every other advertisement on British TV will be aimed right at BT.

The other networks are building for 15/1 Mbps, similar to BT. BT's response will be burdened by heavy legacy costs and debt. Telecom is a business of scale, which gives BT one great advantage. But if they don't hold most of their customers, the base erodes. My guess is that BT will stick with current plans. But I understand some insiders are active proponents of a much more aggressive strategy, and an honest presentation to investors that if they don't invest they will fall behind.

BT's Joe Kelly writes "Our 21CN will deliver UK customers with a radically enhanced customer experience. They will have single billing and a range of new converged services. We believe 80 percent to 90 percent of customers will choose the improved self service option. It will also deliver new products and services to customers much faster than we can today—from an average of 18 months to months or even weeks or days. By opening up the platform to third party application developers, it will also enable the introduction of new services from non-traditional sources, no longer limiting the innovation to BT's payroll. There's much life left in the old copper dog yet. BT already provides the highest stable speeds across the widest national footprint in the world (up to 8 Mbps). BT's policy is to provide the highest speeds in every territory, not just those that choose to live in densely populated metropolitan areas. BT will be state of the art and then some."

"Our 21CN access technology has been specified to work both in the exchange, and to be relocated to operate in the street. In addition, PON technology will work from the MSANs we have selected. We have trialed fiber, and will deploy it in some green field developments."

AT&T is similarly exposed, sticking with DSL speeds from 3,000 to 5,000 feet (20 to 25/1 Mbps), mostly from COs and remotes that already have fiber. (The "fiber to the node" as a fiber build is great pr but essentially false.) AT&T's culture makes it very hard to question the decisions at the top, however. It's a "very military-like organization," very top-down, say yes as soon as the boss finishes talking. Randall Stevenson has not yet officially taken over from Ed Whitacre, and is scared that if Lightspeed is declared a mistake Ed might choose someone else for the CEO spot. It's extremely scary for anyone at AT&T to speak up—two CTOs have been pushed aside already.

Copyright 2006 Dave Burstein.
The DSL Prime Newsletter is reprinted with permission.

"The power of the printing press belongs solely to those who own the presses"
—A.J. Leibling

The Internet is the cheapest printing press ever invented.