En route to real competition

The latest round of squabbles between incumbent and competitive local exchange carriers has centered on data unbundling.

CLECs have grown accustomed to buying network elements from the incumbent and reselling the incumbent's services. These CLECs see no reason why data should be treated any differently. Incumbents have argued that they would have little or no incentive to invest in broadband technologies such as digital subscriber line if they have to resell those elements to their competitors at or near cost.

Late last week, the FCC formalized its response to these squabbles:

Incumbent carriers should be exempted from sharing their DSL access multiplexers (DSLAMs) with competitors if they deploy them through a separate data subsidiary that is subject to the same terms and conditions as CLECs. The subsidiary would have to buy unbundled loops and collocation facilities from its own telco operations, just like its competitors.

To date, most RHCs have deployed DSL through their telco operations, but if the proposed guidelines stick, incumbents most likely will choose to use a separate subsidiary. Ameritech, in fact, already has done just that-and has done so as a CLEC using unbundled loops and collocation space from its sister unit.

The guidelines have drawn the customary "their grass is greener" kind of commentary. Incumbents say they're disadvantaged because only they will bear the extra costs of a separate subsidiary. Competitors say the incumbent subsidiary's reliance on loops and collocation from its sister organization does not prevent the sister organization from charging exorbitant prices for those offerings. The money will simply flow from one part of the parent company to another, competitors argue.

One class of carrier is not complaining, however. Several CLECs-including Covad, Northpoint and Rhythms NetConnections-have focused on deploying DSL and already have made a substantial investment in collocation facilities, unbundled loops and DSLAMs. These DSL-centric CLECs would like to recover their investment by wholesaling DSL lines to other carriers. It's good news for them that the incumbents will not be required to wholesale their own DSL facilities at or near cost.

The way different types of CLECs have responded to the regulators' proposal is reminiscent of the rift that appeared between CLECs and interexchange carriers over the Eighth Circuit Court decision on unbundled network element platforms. The court exempted incumbent carriers from offering the lowest wholesale pricing to competitors that weren't building their own local networks. IXCs were unhappy because they weren't doing much building. CLECs were happy because they were making that investment and they didn't want their pricing to be undercut.

For the same reason I thought the Eighth Circuit Court decision was a good one, I think the FCC is moving in the right direction on DSL unbundling: Real competition depends on investment in infrastructure.

The fiber networks that competitive long-distance carriers constructed in the 1980s brought new capabilities that would not have been there if those competitors had simply resold the existing carrier's copper. And long-distance pricing came down only when new carriers started building more capacity.

Insisting that competitors have the option of reselling infrastructure provided by a single carrier might bring more players to the market-but it doesn't bring better features or lower pricing to end users. Only infrastructure investment can do that. Luckily, the FCC has opted to encourage infrastructure investment.

Of course, it's likely to be a decade or so before competitors can build a connection all the way to the majority of U.S. customers. The impracticality of running two sets of wires down the street is what created telecom monopolies in the first place. That's why it's so important for incumbents to offer unbundled loops and collocation at reasonable terms to competitors.

But it doesn't necessarily follow that incumbents also must share their DSLAMs. Deploying DSL need not present the same degree of difficulty as rewiring every block in America-if competitors can get loops and collocation.

The next debate will be over how to ensure that competitors can get a loop capable of supporting DSL. And incumbent subsidiaries will want to know if there's a way to offer voice and data over DSL without having to unbundle the DSLAM.