The Enemies of Usage-Based Broadband


Here’s a thought experiment: A two-lane toll highway suddenly becomes highly congested during peak times — because a growing number of people are now spending, say, 10 times as much time on the road.

Where will the money come from to expand capacity?

Do you raise tolls for everyone? Or do you establish a usage-based model that (a) shifts more of the costs to the heaviest users and (b) encourages more efficient use of the shared road?

The issue of usage-based consumption for broadband services has reared its ugly head once again, after FCC chairman Julius Genachowksi said allowing such models is important to “promote network investment and efficient use of networks.”

Predictably, many parties are interested in maintaining the status quo. Part of it is just human nature: People often will overpay for unlimited-usage plans, just because they don’t like the idea of a meter running in the background.

Then there are those who would like to consume — or have their users consume — oodles and oodles of video without having to pay more for it. “If people are forced to pay per kilobit it’s like they are forced to pay per word of a book,” Todd Weaver, CEO of online-TV streaming company Ivi, told the Washington Post.

Never mind that Weaver would like to have unlimited usage broadband mainly because he’s trying to build a business around streaming broadcast TV programming over the Internet without asking for permission to do so.

The germane question is, to get networks capable of letting millions of Internet users watching substantial amounts of HD video simultaneously, how will ISPs fund the necessary upgrades?

Some observers dismiss the economic considerations for ISPs in this debate as trivial, illogically taking it as an article of faith that the cost for additional infrastructure won’t outstrip revenue.’s Karl Bode, an articulate critic of usage-based pricing, claims that metered broadband models “aren’t based on real economics, given the continually dropping cost of terrestrial bandwidth and hardware.”

Really? Where is the evidence that the costs to upgrade Internet networks to handle double-digit yearly increases in bandwidth utilization into the next decade can be covered without raising rates in some way?

Arguing that ISPs have managed to accommodate increased usage without resorting to consumption-based pricing so far misses the point: The wave of video-driven Internet usage hasn’t fully crashed on the shore yet. Common sense dictates that adding hundreds of Gbps of capacity is a significant cost. (Even if Level 3, with respect to its feud with Comcast over interconnection fees, wishes otherwise.)

As I’ve said before, paying based on what you use is the fairest and most reasonable approach for operating broadband networks in the era of Internet TV — see Time Warner Cable: Three Mistakes on Usage Pricing and Why Monthly Broadband Usage Caps Won’t Really Work (But Usage-Based Billing Will).

ISPs are doing this by proxy today, with differently priced tiers based on connection speed, but longer term usage-based pricing will be a better way to address this issue.