Customer-Owned Fiber

by on July 30, 2008 · 8 comments

I’ve got a new piece up at Ars Technica that explores the concept of customer-owned fiber. It was inspired by a post by Google’s Derek Slater, who is working with Tim Wu on a paper making the case for customer-owned fiber in more detail.

The structure of today’s telecom market is roughly analogous to a road system in which most peoples’ driveways were owned by a private company. (To make the analogy, you’d have to imagine that most homes have two driveways owned by different companies) If another company owns your driveway, that company has a significant amount of leverage over you. Regulatory proposals like “open access” and “network neutrality” are like taking this system of third-party-owned driveways for granted and trying to use regulatory levers to ensure that companies don’t abuse that power.

But a much better approach may to recognize that the whole setup is screwy: it makes more sense for the owner of a particular piece of property to also own that part of the telecommunications infrastructure that’s exclusive to that property. A world of customer-owned fiber would solve a lot of the thorny policy problems because the barriers to entry in telecommunications markets would be radically lower. Entering the ISP market in a given neighborhood would simply require running a single strand of fiber to that neighborhood’s peering point.

The question is how you get there from here. In the near future, this will no longer be just a theoretical discussion, as a private company in Ottawa recently completed construction of a 400-household fiber network that it plans to sell to local homeowners. The preliminary cost estimate, based on 10 percent take-up, is $2700 per participating home. A higher sign-up rate means a lower cost per home. We’ll know in a few months if that estimate was optimistic or pessimistic.

In my piece, I talk about the economics of fiber rollout and the likely obstacles. I’m skeptical that it can be made to work because the costs are substantial and this is an obstacle with a lot of interia. There is also a group of incumbent companies—not a cartel—who are likely to do everything they can to kill such an effort if it gained momentum. But I think it’s absolutely worth trying, because if it could be made to work, the benefits would be huge.

Previous post:

Next post: