Wholesale failure

In arguably the most telling indictment of the DSL wholesaling business model, AT&T last week bought most assets of bankrupt NorthPoint Communications but decided the DSL provider's customer base was not worth the trouble.

For $135 million, AT&T purchased NorthPoint's DSL equipment and co-location space in central offices across the nation — assets the IXC will use as AT&T's consumer division rolls out its new DSL offering to residential customers. But it left NorthPoint's customers on the table.

“We basically got everything except the customers,” said an AT&T spokesman. “We did not choose to have the customers be part of this deal.”

NorthPoint's business-heavy customer base does not fit the residential market that AT&T Consumer will target in its DSL initiative, an AT&T spokesman said. When asked why AT&T did not buy NorthPoint's customer base and give the business customers to the AT&T Business group, the AT&T spokesman said, “We thought it was in [AT&T's] best interest not to.”

In an industry where customer acquisition is highly valued, declining customers at a fire-sale price is very surprising, according to Gary Kim, president of NexGen Data Research.

“Usually, in a deal like this, what you really want is the customers…. Typically, that is the only reason you make the deal,” Kim said. “This is the first time I've seen anything like this.”

“My guess is AT&T ran the numbers [on NorthPoint's customer base] and they came out negatively. The real implication there is AT&T's opinion that the wholesale DSL business model doesn't work.”

Such a conclusion speaks volumes about the financial problems inherent in the wholesale DSL business model, which relies on three entities — the DSL provider, the ISP reseller and the ILEC that controls the copper lines — cooperating and maintaining profitability. Noting that DSL does not generate enough revenue to satisfy all three, Kim applauded AT&T's strategy to remove the ISP reseller from the formula.

“The important thing is getting out of the wholesale model and selling retail direct, which takes an entire step out of it and makes it easier to be profitable,” Kim said. “Under the old [wholesale DSL] model, three people had to be fed; with retail direct, only two people have to be fed.”

This reality explains the recent strategy of Covad Communications, which is trying to get the customers of failed ISP resellers — and now, NorthPoint — to switch to Covad.net as part of its attempt to emerge from its financial struggles. About 80% of Rhythms NetConnection's revenue comes from retail-direct customers, which is why it is has been more fiscally stable than its DSL competitors, Kim said.

“And [Rhythms] still may run out of money,” he said.

That possibility is troubling to business customers, which value the data CLECs' ability to provide symmetrical DSL. Thus far, established carriers have offered only asymmetrical DSL service — a strategy critics attribute to these providers not wanting to cannibalize their high-margin T-1 businesses.

In the residential market, RBOCs have a better formula than retail direct — using its own lines, meaning they get to keep all revenues derived from DSL service. For this reason, AT&T Consumer's strategy of pursuing the residential DSL market is questionable, according to Drake Johnstone, first vice president for Davenport & Co.

“No one besides the RBOCs have had any sustained success selling DSL,” Johnstone said. “RBOCs have not proved easy to deal with for others trying to access their networks. I don't see any reason AT&T would be any different.”