DSL Will Win Where It Matters

Digital doesn't mean anything without profits. Cable appears to be off to an early lead in the United States as of autumn 2000, claiming 2.2 subscribers for every person who receives broadband through Digital Subscriber Line technology. But the absolute subscriber base for broadband remains small, and DSL has plenty of time to catch up. At present, about 40 percent of all phone lines are ready for DSL, while 58 percent of all households can order broadband over cable. But the real action will come in the next four years, as broadband subscribership moves from fewer than 5 million homes and small businesses into the range of 40 million to 65 million. During that time, DSL-enabled lines will be installed at a faster rate than will cable lines. Indeed, market research from Morgan Stanley Dean Witter predicts that DSL subscribers will outnumber broadband cable subscribers this year. By the end of 2004, both technologies should be available to about 75 percent of all households. More to the point, the for subscribers may be a red herring. If we look to the future and focus on the race for profits rather than warm bodies, DSL providers appear to occupy the stronger position in the highly profitable midsize and small-business segment than does cable, because of the latter's origins as a conduit for pay television. Investment in installation To bring DSL service to a neighborhood, a provider must install a device called a Digital Subscriber Line access multiplexer (DSLAM) at the local telephone-switching office. Every DSLAM installed by the broadband provider can deliver service to about 250 homes, and providers can install additional DSLAMs in the switching office as subscribership grows. Each home that is to receive the service must be less than 15,000 to 18,000 feet from the local office (because DSL signals degrade over distance) and requires a special DSL modem, which until recently had to be installed in the home by a technician from the provider. DSL also used to need its own separate telephone line into the home, though new line-sharing regulations are eliminating that requirement. The first enterprises to pursue the DSL market aggressively weren't telephone companies but DSL-focused start-ups, such as Covad, NorthPoint, and Rhythms. Since the implementation of the US 1996 Telecommunications Act, the telephone companies have been required to let the start-ups install DSLAMs in local telephone offices on more or less unprejudiced terms. Companies such as NorthPoint have invested heavily in automating the cumbersome installation process, with impressive results. Although the regional Bell telephone operating companies (RBOCs) certainly got off to a slower start, they are now making genuinely serious efforts to compete by installing DSLAMs and modems, either on their own or with partners recruited for the purpose. For example, Verizon, formerly Bell Atlantic, is investing aggressively in new infrastructure to bring DSL to a wider range of its residential and small-business customers. Of course, the telephone companies, influenced by their huge revenues generated by traditional voice service, have a perspective different from that of a start-up. In their home states, they see DSL as a way of ensuring that customers don't develop relationships with other companies that (under the 1996 Telecommunications Act) also have the right to offer voice communications services. Outside their home states, they are getting into the DSLAM-and-modem business in hopes of winning voice and data revenue from other telephone companies. Thus, a telephone company could be an attacker outside its home states and a defender within them. Cable's lead is less solid than might be imagined. Although cable entered the residential market earlier than DSL did and had a more standardized--and thus easier-to-install--product, huge investments are being made in DSL's network infrastructure. …