SpectrumTalk

NTIA Broadband Report





On January 31, NTIA released the report, “Networked Nation: Broadband in America, 2007,” This report

highlights the dramatic growth of broadband in the United States. The report shows that the Administration’s technology, regulatory and fiscal policies have stimulated innovation and competition, and encouraged investment in the U.S. broadband market contributing to significantly increased accessibility of broadband services.

So what's the problem? OECD statistics show that the US is lagging in broadband compared to other countries and ranks 15th among OECD members. (While my wife worked for OECD for 3 years, she was not at all involved with this issue.) The State Department has formally complained to OECD about the methodology used and NTIA head John Kneuer has also stated that the OECD statistics are misleading.

Surprise, the new NTIA report says things aren't so bad after all if you analyze the data the right way. The report states,

Recognizing this transformative power, four years ago President Bush articulated a national vision: universal, affordable access to broadband technology. From its first days, the Administration has implemented a comprehensive and integrated package of technology, regulatory, and fiscal policies designed to lower barriers and create an environment in which broadband innovation and
competition can flourish.

Apparently in this election year the Commission's two democrats aren't buying this logic. Commissioner Copps wrote today,

Networked Nation? If the United States were a networked nation consumers would be paying half as much for broadband connections 20 times as fast. That’s what many consumers around the globe get.

Commissioner Adelstein added,

With only half of adult Americans participating in the broadband age and U.S. consumers paying far more than citizens in other countries for less bandwidth, this report appears to be missing some key chapters. Noticeably absent is any coherent strategy going forward. This report relies on widely-discredited data in a strained effort that only distracts us from the real work ahead.

Finally at the end of the day I got my weekly ARRL Letter . (I am a card-carrying ARRL member, but do not agree with them on many aspects of the BPL issue as I was also heavily involved in the FCC BPL decision.) ARRL points out that there are some problems with the reports numbers on BPL deployment:

At one time the FCC's semi-annual reports, "High-Speed Services for
Internet Access," lumped BPL in with fiber optic lines. The FCC
eventually recognized that this was inappropriate, since the two
technologies have absolutely nothing in common, and stopped doing so
after 2004. For some reason, the NTIA's report continues to treat the
two together. Even so, reading the executive summary offered a glimmer
of hope that the report would be realistic with regard to BPL; it notes
that while "the total number of high speed lines delivered over fiber
and power line connections grew 789 percent from December 2003 to
December 2006...[f]iber optic lines...appear to be almost entirely
responsible for this expansion." (The latest FCC report that is
available is for December 31, 2006; it was released in October 2007.)
Less encouraging was the fact that the term "BPL" occurs no fewer than
45 times in the 60-page report, an emphasis that is hardly justified by
"Reliable BPL subscribership figures are difficult to find. The FCC's
most recent data identify fewer than 5,000 BPL customers as of year end
2006. That figure appears low, however. TIA [Telecommunications Industry
Association] estimates 200,000 current BPL subscribers, increasing to
700,000 by 2010. Another source forecast about 400,000 customers by the
end of 2007, growing to 2.5 million by year end 2011. "

WHAT IS GOING ON HERE? The FCC's data showing fewer than 5000 BPL
customers -- a number that dropped in the six-month period covered by
the report -- are taken from forms that service providers are required
to submit. Why does the NTIA not regard this figure as reliable? The
only way that it "appears low" is by comparison with the excessive
industry and government hype.

Further distorting the picture, at the bottom of page 26 is an
out-of-date map taken directly from the United Power Line Council
(UPLC), an industry source with a vested interest in BPL. It purports to
show BPL deployments "updated as of July 10, 2007." However, a number of
those shown had already been decommissioned by that date and others have
been taken out of service since then.

So it looks like NTIA's attempt to clarify the broadband situation may have just created more confusion.

======
10/9/08 Update
I was amused to get the following additional information in the latest ARRL Letter:

The National Telecommunications and Information Administration (NTIA) report Networked Nation: Broadband in America 2007 that was released on January 31, 2008 includes the following: "Reliable BPL [broadband over power lines] subscribership figures are difficult to find. The FCC's most recent data identify fewer than 5,000 BPL customers as of yearend 2006. That figure appears low, however. TIA [The Telecommunications Industry Association] estimates 200,000 current BPL subscribers..."

Five years of experience in dealing with BPL systems as a radio interference source have given the ARRL, the national association for Amateur Radio, considerable insight into the BPL industry. Based on that experience, the ARRL has concluded that the FCC's figure of fewer than 5000 BPL customers is entirely credible. Therefore, the ARRL set out to determine the source of the "estimate" of 200,000 current BPL subscribers.

We contacted TIA and were advised that the figure came from a market study prepared by Wilkofsky Gruen Associates Inc and based on research conducted by In-Stat, a unit of Reed Business Information.

So we contacted In-Stat and asked how the figure was derived. They responded: "The 200,000 number for BPL subs did not come from In-Stat. In our US broadband forecast, we estimate about 231,000 broadband subscribers in the 'other' category besides DSL, cable, satellite. Other includes BPL, but is not solely BPL."

We then contacted Wilkofsky Gruen Associates. They responded: "Our source for the BPL figures was In-Stat." When In-Stat's denial was shared with them, they responded, "It was our understanding that BPL was the principal component as it was the first item listed by In-Stat."

TIA was invited to comment but declined to do so.

In other words, here is what we have learned: In-Stat does not claim to know how many BPL subscribers there are, but provides an estimate of 231,000 broadband subscribers who receive service via delivery systems other than DSL, cable, and satellite. Wilkofsky Gruen Associates, on the basis of nothing more than that BPL is listed first, assumes that the bulk of these 231,000 are BPL subscribers and arbitrarily attributes 200,000 of them to BPL. In turn, NTIA -- not satisfied with an FCC figure that is derived from required reports from service providers -- cites this arbitrarily chosen figure -- a figure that is entirely unsupported by any data whatsoever -- as evidence that the FCC's figure -- which is fully supported by data -- "appears low."

As I said before, I don't fully agree with ARRL on their BPL posture. But this incident shows something on the state of communications policy in Washington now.