Friday, July 15, 2016

Is Special Access Market Competitive, or Not?

According to the U.S. Federal Communications Commission, the total market for U.S. special access services is roughly $40 billion annually. The FCC believes the market is insufficiently competitive.

But observers note that the FCC itself reported U.S. telcos had 92 percent market share in special access in 1980. Telco share had dropped to 39 percent in 2013. In that year, TDM-based services represented $25 billion of the total market, while incumbent telcos accounted for about $16 billion of the TDM market, according to the FCC’s own data.

In some markets, 39 percent might well represent “dominance.” Whether that is the case in the special access market is the issue.

With the caveat that usage or bandwidth is not the same thing as revenue, Ethernet access provided by a wide range of competitors now represents market growth, while the legacy T1 and DS3 market declines.  

In fact, cable TV companies are big suppliers of Ethernet access and other access services to business buyers.

AT&T has argued that “facilities-based competitors are serving 95 percent of all MSA census blocks (on average, about one seventh of a square mile in an MSA) nationally where there is demand for special access services, and, second, that 99 percent of all business establishments are in those census blocks.”

For the first time ever, the U.S. Federal Communications Commission also is seeking to extend some special access obligations to cable TV networks for the first time.

And some argue that it is impossible to fully do a data-driven analysis because the FCC is not releasing the full results of its earlier survey of facilities, an exercise intended to provide some rationale for assessing the existence of facilities-based competition.

Some might find the emphasis on extending regulation to a declining service curious. AT&T’s access lines have declined by almost 65 percent since 2009.

No comments:

"Tokens" are the New "FLOPS," "MIPS" or "Gbps"

Modern computing has some virtually-universal reference metrics. For Gemini 1.5 and other large language models, tokens are a basic measure...