Monday, May 18, 2009

Why Broadband "Penetration" Measures Often are Misleading

If you were trying to figure out how prevalent televisions, radios, digital video recorders, Slingboxes, PCs or DVD players were in people's lives, would it make more sense to measure how many Best Buy retail locations sold such products, or how many units are sold in any given time period?

Alternatively, if you were trying to measure the penetration of such devices, would you track the number of homes, businesses, or both, that have such devices in use?

Would you try to measure "personal" devices such as mobile phones or MP3 players the same way?

The questions aren't as "academic" as might first appear to be the case.

While it makes sense to measure the penetration of any mobile and personal technology on a per capita basis, because that is the way people buy and use such services and products, it arguably makes less sense to measure other products, such as T1 lines, Ethernet or other fixed broadband connections the same way, because that is not the way people buy or consume such products.

Were we to measure Ethernet connections on a per-capita basis, penetration would be quite low, for example. Most people intuitively would understand that sort of issue.

But where it comes to fixed broadband penetration, that is precisely the problem we face. Agencies are used to measuring fixed broadband in just about that fashion: per capita, even though people do not buy such services that way.

The point simply is that we need to measure things in a way that reflects the way people actually use a given product or technology.

People do not buy fixed broadband subscriptions the same way they buy mobile phones.

So per capita indexes are more suited to some products than others. Per-capita fixed broadband indexes are affected by mundane things such as household size, business adoption and consumer preferences.

"Consider Portugal, in which there are approximately three persons per household," says George Ford, Phoenix Center for Advanced Legal and Economic Public Policy Studies chief economist. "If every household had a broadband connection, then the per capita subscription rate in Portugal would be 33 percent"

"In Sweden, alternately, there are approximately two persons per household," says Ford. "So, if every home had a
connection, then the per-capita subscription rate is 50 percent."

"The number of fixed broadband connections per person is a flawed measure because it will vary based on the average size of a household or business establishment," Ford notes.

"In the United States, nearly every business and household had a fixed line telephone when the 1996 Telecom Act was passed," Ford notes. "Yet, telephone subscriptions per capita were only 49 percent at the time."

"In Sweden, which also had near ubiquitous telephone adoption, the telephone per-capita subscription rate was 69 percent.

The point, says Ford, is that per-capita measures are not meaningful tests of fixed broadband adoption, especially when comparing different regions or nations.

No comments:

Many Winners and Losers from Generative AI

Perhaps there is no contradiction between low historical total factor annual productivity gains and high expected generative artificial inte...