Broadband isn't what it used to be in Houston and DC

Th Stephenbrown

The feds decline to call any contemporary telecom service "broadband," yet the law explicitly uses that term to describe "advanced" telecom.


A Texas class-action suit accuses the incumbent local telephone company of "capping" downstream data flow and overselling capacity; meanwhile, the Federal Communications Commission "declines" to describe any current telecommunciations service as "broadband."

On Aug. 3, the FCC released its second report in CC Docket No. 98-146, "The Deployment of Advanced Telecommunications to All Americans." It mainly assesses the deployment rates of DSL and cable modem technologies, which the agency treats as "advanced telecommunications capability," and predictably concludes that deployment in the United States "is proceeding in a reasonable and timely fashion, overall." The data-laden report is prefaced by a policy statement uncharacteristic of the agency's usual ambivalence toward bidirectional communication (see Lightwave, January 2000, page 24): "We believe that Congress intended to bring to all Americans a two-way, truly interactive medium rather than one that is passive and entertainment-oriented." That's good news.

For the first time, a federal agency affirms that substantive two-way communications is intrinsic to "advanced" telecom infrastructure, a notion without precedent in any federal or state jurisdiction and one that has long been resisted by most players in the telecom business.

Now, the bad news: The FCC reaffirmed another policy established in its first report issued in January 1999, that "a service may have asymmetrical upstream and downstream paths and still be long as both paths provide speeds in excess of 200 kbits/sec." Supposedly, this definition fits the law's literal and intended meaning that advanced telecommunications is "broadband...capability that enables users to originate and receive high-quality voice, data, graphics, and video."

The agency's policy is a de facto rewrite of the law, as if it says, "originate some and receive a lot." But despite this shortcoming, the second report has a genuine surprise: "In light of its now common and imprecise usage, we decline to use the term 'broadband' to describe any of...the services...that we discuss in this report." That's another first: The federal government declines to call any contemporary telecom service "broadband," even though the law explicitly uses the term to describe "advanced" telecommunications.

The FCC's wording shouldn't be dismissed as hairsplitting. The government view of telecom infrastructure, at least for now, is that broadband services and advanced services are not equivalent. Therefore, the FCC's second report does not affirm that broadband facilities are being reasonably and timely deployed, a major shift from the first report where the Commission said, "residential consumers are purchasing broadband services and...have access to broadband capability." Perhaps, the agency's perception is changing, moving away from its embrace of DSL technology-a change that can't come too soon.

On August 16, a major class-action lawsuit was filed in Nueces County, TX, against SBC Communications and its major operating subsidiaries, South western Bell, Pacific Bell, Ameritech, Nevada Bell, and Southern New England Telecom. According to the suit, the company offers an "always on" DSL downstream speed of 384 kbits/sec for $39.95 monthly, but once a customer takes the service, the company as a matter of policy and "without the knowledge or consent of customers" caps e-mail and newsgroup speeds to 128 kbits/sec between the central office and the customer, while also "timing out" the connection, as if it were a dial-up service. Also, the incumbent's offering is described as a "bait-and-switch scheme," because once a customer discovers the true situation and complains to the company, it responds by offering to remove the cap if the customer upgrades to "enhanced" DSL service for $80 monthly. Customers who terminate their service before their two-year contract expires pay a $200 penalty and wait three weeks to be switched to another provider.

SBC denied the allegations when it filed its response with the court, but the lead attorney for the plaintiffs, Geoffery Berg of Houston, said,"We would not have proceeded if we were not confident that we could prove our case...We will ask the court for an early trial date."

The lawsuit's chief implications are that DSL service is a loss-leader product, under-priced, and severely capacity-constrained-the hidden reasons for the service provider placing arbitrary limits on data flow. Arbitrary limits should not apply to the user or the resource being accessed, whatever the medium's composition. Of course, capacity shortage is not a major issue in a fiber medium, which is far more likely than DSL to achieve Congress's intent of bringing a "two-way, truly interactive medium," to all Americans, a fact the FCC has not acknowledged.

More important, the "bait-and-switch" allegation is a symptom of a larger problem: the proper pricing of bandwidth. If DSL systems are capacity-short, then the providers are selling capacity that can't match demand, just like AOL did a few years ago. In this case, flat-rate pricing of $40 a month for "all you can eat" is inappropriate, because it gives no incentive to users to stay offline during congested hours. On the other hand, flat-rate pricing is an easy way for providers to hook subscribers, giving them an illusion that capacity is ready and waiting as needed.

This mismatch between what is expected of a service and its price is worsened by the FCC's minimal threshold for advanced services: 200 kbits/sec, a tiny limit allowing DSL service to be a shape-shifter, posing as an advanced service for marketing purposes, then reverting to its true nature during daily operations.

However, this problem is not the same as the one mentioned by Gerry McGovern of NUA Ltd., in the essay, "13 Things to Know About Broadband." McGovern asserts, "The flat-fee...pricing model is unworkable in a broadband environment where one user might want to use hundreds or thousands times more bandwidth than their neighbor." McGovern is correct if the system is capacity-constrained, but flat-fee pricing is appropriate in a fiber system, where bandwidth is a commodity.

A different perspective is offered by Bill St. Arnaud of CANARIE in his essay, "Is DSL slowing down the Internet," which says that congestion has moved from the local drop to the network. According to St. Arnaud, the "explosion of high-speed access brought on... by DSL, [means] ISPs [Internet service providers] are suddenly finding that they can no longer blame the end-user for the slowness of the 'Net. The tables have turned, and now the end-user is blaming the ISP. 'I have a high-speed connection, so why aren't things faster?'" But the Texas case shows it's too early to bless the local drop and heap all blame on the networks.

St. Arnaud also noted, "Content owners are finding they can't solve their problem simply by adding more servers and/or bandwidth. Adding more servers to deal with peak demands results in those servers sitting idle during periods of low demand." That is one of the oldest problems in economics: How to price a product so demand is met, while minimizing idle capacity. The electric and natural gas industries deal with the issue through two-part pricing: Customers pay a price for using energy when the system is at a peak or congested as well as paying another price for using energy no matter when it is consumed. This method is very similar to the two-part pricing of frame relay service, where one price is for a "guaranteed minimum" data rate and a second price is for a "burst" or peak data rate at certain times. Thus, utility-type pricing is creeping into the telecom business and looks appropriate for DSL services.

The exact way to arrive at the two prices is always a difficult issue, because they affect usage patterns by imposing more costs on users if they are online during a peak period. There's no doubt proper pricing at the user level would ripple into the networks, alleviating the "idle server" problem, giving customers more accurate expectations, and slowing the growth of DSL services to a manageable rate so "broadband" capacity is rationed through forthright pricing rather than suspect practices that provoke lawsuits.Th Stephenbrown

Stephen N. Brown writes on public policy in telecommunications. He can be contacted by e-mail at or telephone: (615) 399-1239.

More in FTTN/C