Cover Story: Will DSL Survive?
Back in the late 1990s, the phone companies were quick to oblige Internet surfers with a product they call DSL, short for its lackluster name: “digital subscriber line.”
It quickly became a hot seller, and was briefly considered a threat to cable modems. But in a few short years, as consumer demand was met by the superior speeds offered by cable operators, DSL's growth trajectory plunged.
Today, DSL claims 30 million U.S. subscribers and few doubt that its growth days are over. The number of new customers fell from a peak of 4.15 million additions in the second quarter of 2006 to 1.46 million additions in second-quarter 2008, according to a recent research report by investment firm Sanford Bernstein.
By comparison, cable-modem customers numbered 35.3 million in the second quarter, and cable outpaced DSL in terms of net subscriber additions by more than a 3-to-1 margin — 671,426 net additions for cable versus 215,514 for the telephone companies.
“So the elephant in the room is, what are the telephone companies going to do now that we're getting 75% share or whatever the share was last quarter? I think the answer is there's nothing they can do, because the problem isn't price,” Comcast chief operating officer Steve Burke said at the Merrill Lynch Fall Media Preview conference in September.
At the time, Burke said that Comcast felt at an advantage because it was competing against DSL in most of its footprint at a time when consumers are seeking ever-faster speeds to download video and other applications. Showing his competitive side again, Burke said on Comcast's second-quarter conference call that the MSO's strong broadband results “continue to suggest, as some have said, that DSL may be the new dial-up,” a dig referring to the super-slow telephone wire still used by 15 million subscribers.
DSL may be dying, but it certainly isn't dead. AT&T, which lost DSL customers in the second quarter, improved to 148,000 additions in the third quarter. And Verizon Communications also improved to a smaller loss of about 96,000 DSL customers in the third quarter in the same period. The three largest telcos — AT&T, Verizon and Qwest Communications International — had about 26 million DSL subscribers at the end of the second quarter, or about 88% of U.S. DSL subscribers, according to Leichtman Research Group,
Multichannel Newsletter
The smarter way to stay on top of the multichannel video marketplace. Sign up below.
DSL's decline stands as an unavoidable truth to the nation's telephone companies, which are racing to build fiber lines to compete with cable. In its second-quarter conference call with analysts, Verizon chief operating officer Denny Strigl admitted the company had focused more on its fiber-optic FiOS product than DSL, which he vowed to change. AT&T has also said it placed more emphasis on its fiber-to-the-curb network U-Verse, which has been a factor in the DSL decline.
The simple logic of DSL's downfall is speed. Compared to DSL's top speed of 7.1 Mbps and a low of 256 Kilobits per second, cable-modem service varies from a low of about 5 Mbps to 30 Mbps for “Power Boost” service. And several MSOs are in the middle of rolling out DOCSIS 3.0 high-speed networks that will offer speeds of about 50 Mbps — Comcast expects 20% of its footprint to be DOCSIS 3.0-capable by the end of this year.
Those speeds are matched by Verizon's FiOS product — between 30 Megabits per second and 50 Mbps — and AT&T's U-Verse, which comes in around the middle with a maximum speed of 18 Mbps. But for the vast majority of the telco footprint, DSL will be the only telco option for high-speed Internet service.
The belief that every Internet user will be hungry for more bandwidth speed is a misplaced one, said Leichtman Research principal Bruce Leichtman, who doesn't believe that an explosion in video downloads has created a huge demand for speed. He cited research which states that the average U.S. consumer spends less than eight minutes per day downloading video.
“People are getting way too carried away with that,” Leichtman said.
Indeed, telephone companies may find a niche in customers who have no need for speed. Verizon spokesman Bob Elek said that consistency of service is more important to his customers than speed.
“It's not the demand for speed as much as it is a demand for consistency,” Elek said. “With a DSL connection, you can sit down and you're going to get very consistent performance out of that connection up to or at the speed it's advertised at. The shared cable-modem architecture doesn't generally produce that. It can, but you certainly can't be getting online when a lot of other people are online as well.”
While that argument has been used by telcos for years, Leichtman said that as the industry has matured, new customers have different priorities.
“When you're the 66 millionth customer, there's a reason why you've held out so long. Fifteen Mbps, 50 Mbps, you don't care about that,” Leichtman said. “What you care about is does it work, that it has a good value.”
The telcos are counting on the continued relevance of DSL for many reasons, but perhaps the most important one is that for most of its footprint, it will be the only broadband product it can offer. According to Sanford Bernstein's Craig Moffett, Verizon plans to build its FiOS fiber network to reach 18 million homes by 2010, leaving 15 million homes to DSL. AT&T expects its U-Verse network will touch 30 million homes by 2012, leaving 22 million households fiber-free.
After accounting for all the homes expected to be reached by fiber, about 71.5 million households — or 60% of the country — will not be directly connected to fiber, according to Sanford Bernstein. “In the majority of the U.S. — and by a large margin — the telcos are betting on legacy DSL,” said Moffett.
Moffett points out that cable's broadband advantage didn't just happen in the first half of this year. “The shift towards cable — and away from DSL, in particular — is no longer a question, and indeed is now more than two and a half years old,” Moffett wrote.
Perhaps the telephone industry's greatest concern is access-line losses, which totaled 2.3 million in the last quarter. Although telcos argue that those losses are former wireline customers who've cut the cord and switched exclusively to cellphones, Moffett argues that about two-thirds of access-line disconnects represent defections to cable.
“As more consumers disconnect their telco phone lines, there are fewer and fewer left to subscribe to DSL,” he wrote.
Access-line losses started to become a major issue for the telcos just as cable began offering its own voice service. Residential access lines at the major telcos have dwindled from 101.8 million in 2003 — when Cablevision Systems began offering the first $90 triple-play package — to 72.4 million in the second quarter, according to Sanford Bernstein.
“Cable bundling their service with phone has absolutely helped them,” Leichtman said. “Cable has phone available to over 100 million households. They can tie the broadband into those 100 million households. Phone has fiber available to 20 to 25 million [households]. That's the big difference.”
But Leichtman said the decline in DSL subscriber additions in the past was mainly due to a conscious decision by the telcos to not pursue less-profitable customers. “What happened in the last two quarters wasn't that cable suddenly got faster,” Leichtman said. “It was that the telcos stopped going after the low, low end.”
Leichtman added that the telcos realized that the game wasn't about gaining share at any cost, it is about gaining high-quality customers.
“If DSL is killed, it is a self-inflicted wound,” Leichtman said.
To offset the losses, telcos are trying to boost DSL speeds and package the service differently. Just last week, AT&T kicked up the speed of its highest-tier DSL service by 80%, introducing a broadband service with 18 Mbps downloads available only to customers who also subscribe to its U-verse TV. Previously, its fastest retail offering has provided up to 10 Mbps downstream.
The new 18 Mbps service is an attempt to squeeze the last ounces of juice from DSL and should be some help in stopping customers from fleeing to cable-modem services, according to analysts. While AT&T expects to see continued gains in DSL, the telco said it is also providing customers with several options, including wireless.
AT&T spokeswoman Jenny Bridges noted that AT&T added 2.8 million wireless customers with broadband-capable devices in the third quarter and that the telco bundles free Wi-Fi connectivity with its broadband and Laptop Connect plans, among others.
Telcos are also starting to borrow a page from the cable playbook with DSL, bundling it in different tiers with satellite video and voice service. AT&T began offering a so-called “naked DSL” product in October — one that does not require the subscriber to buy a separate phone line — that it said is proving to be popular with customers.
In early November, Verizon cut prices on DSL, phone and DirecTV bundles and expanded the availability of its fastest DSL tier. The telco introduced an entry-level triple play — 1 Mbps DSL, phone and 150 channels of video for $79.99 per month. It also dropped the price of some of its other DSL bundles by as much as $18 per month.
In addition, Verizon nearly doubled the availability of its highest speed DSL service — with download speeds up to 7.1 Mbps — to 6.6 million homes, versus 3.4 million homes in June.
The telco boosted the speeds of its entry-level DSL Internet service to 1 Mbps downstream and 384 Kbps upstream, up from 768 Kbps/128 Kbps, and is offering the service for $9.99 per month for the first six months with a one-year contract. After the six-month promotional period, monthly pricing for the remainder of the annual plan will range from $19.99 to $25.99, depending on the market.
Verizon just started offering the lower-priced bundles, said Elek, so it's too early to gauge their success. But so far, “it looks pretty good,” he said.
But Elek stopped short of predicting that the tide has turned completely for DSL.
“I don't know for sure,” Elek said. “The market changes from quarter to quarter. We still think it's an extremely viable product for us and our consumers. But cable is going to have its successes too. It's a very competitive industry.”