« Documents

Letter to FCC on AT&T Data Caps

May 6, 2011

This letter from PK and New America Foundation asking for an investigation of AT&T's broadband data caps is available, with full footnotes, as a PDF.

May 6, 2011

Sharon Gillett
Chief, Wireline Competition Bureau
Federal Communications Commission
445 12th St. SW
Washington, DC 20554

Dear Chief Gillett:

In light of the fact that 56% of American broadband subscribers’ connections are restricted by some sort of broadband cap, Public Knowledge and New America Foundation’s Open Technology Initiative urge the Bureau to exercise its statutory authority to fully investigate the nature, purpose, impact of those caps upon consumers. The need to fully understand the nature of broadband caps is made all the more urgent by the recent decision by AT&T to break with past industry practice and convert its data cap into a revenue source.

As the Commission has noted in too many fora to count, broadband internet access is critical to the continued economic and cultural prosperity of our nation. Sectors as diverse as economic opportunity, education, health care, energy and the environment, government performance, civic engagement, and public safety will all rely on widespread access to affordable, high-speed broadband to prosper in the 21st century.  Affordable, high-speed access to broadband has the ability to (in the words of the Commission) “stimulate economic growth, spur job creation, and boost our capabilities in education, healthcare, homeland security, and more.”

However, a shadow is emerging that threatens to dim the dawn of this new era. Caps on broadband usage imposed by Internet Service Providers (ISPs) can undermine the very goals that the Commission has committed itself to championing. While broadband caps are not inherently problematic, they carry the omnipresent temptation to act in anticompetitive and monopolistic ways. Unless they are clearly and transparently justified to address legitimate network capacity concerns, caps can work directly against the promise of broadband access.

Recently, comments of industry analysts have begun to raise disturbing questions about the true motivations behind the caps. These analysts, who often have access to the unvarnished opinions of industry insiders, suggest that network capacity is not driving the imposition of data caps. Stanford Bernstein senior analyst Craig Moffett observed “[t]his isn’t about protecting against the data network being swamped with excess usage . . . [t]his is about putting the business model on a stable, long-term economic model.”

Mr. Moffett’s comments parallel those made by Canadian ISPs last November. “[T]he fundamental purpose of [caps] is not to attempt to recover narrow incremental costs, but rather to influence end-user behaviour. Any assessment of incremental costs that does not take into account broader behavioural impacts would certainly result in rates that make an inadequate contribution to satisfying the carrier’s network management objectives.”

These caps, which are now a fact of life for 56% of all broadband users,7 can perniciously undermine each of the goals set out by the Commission in the National Broadband Plan while at the same time stifling the competition and innovation that has established itself as the sine qua non of the internet economy.

In the world of broadband data caps, the caps recently implemented by AT&T are particularly aggressive. Unlike competitors whose caps appear to be at least nominally linked to congestions during peak-use periods, AT&T seeks to convert caps into a profit center by charging additional fees to customers who exceed the cap. In addition to concerns raised by broadband caps generally, such a practice produces a perverse incentive for AT&T to avoid raising its cap even as its own capacity expands.

Furthermore, it remains unclear why AT&T’s recently announced caps are, at best, equal to those imposed by Comcast over two years ago.  The caps for residential DSL customers are a full 100GB lower than those Comcast saw fit to offer in mid-2008. The lower caps for DSL customers is especially worrying because one of the traditional selling points of DSL networks is that their dedicated circuit design helps to mitigate the impacts of heavy users on the rest of the network. Together, these caps suggest either that AT&T’s current network compares poorly to that of a major competitor circa 2008 or that there are non-network management motivations behind their creation.

In light of the Commission’s statutory power to obtain information critical to effectively serving as protector of the public interest and the priority that this Commission has placed on the deployment of affordable, high speed internet access, Public Knowledge and New America Foundation’s Open Technology Initiative urge the Bureau to request no less than quarterly reports about data caps from AT&T. AT&T’s especially aggressive data cap justifies special attention from the Bureau. Of course, the Bureau should also not hesitate to extend these reporting requirements to other ISPs imposing data caps on their customers. These reports would allow the Bureau to stay as informed as possible about this growing trend.

Specifically, Public Knowledge and New America Foundation’s Open Technology Initiative urge the Bureau to request customer anonymized reports from ISPs regarding the following:

•    Which ISP-offered services are excluded from the cap. This should include reporting on those services, such as voice telephony and video programming, that compete with internet-delivered non-ISP controlled offerings.

•    How often the cap is enforced. This should include the absolute number of customers who exceed the cap as well as the percentage of customers who run afoul of the limit. Additionally, it should include amount by which the customers exceeded the cap and how many of those customers are repeat cap exceeders.

•    Steps taken to warn customers. This should include the steps taken to warn customers, and when those steps are taken. Additionally, the reporting should include data on the effectiveness of these warnings in preventing overages.

•    Average penalty incurred by customers. Caps such as AT&T’s charge additional fees based on how significantly the cap is exceeded. Reporting should include data on the size of the penalties incurred by customers.

•    When and how often a penalty is waived. In addition to the publicly announced grace periods, the Bureau would be well served by understanding how often and under what circumstances reporting ISPs grant additional waivers.

•    The relationship of enforcement to times of network congestion. If used properly, data caps can a tool in easing network congestion. However, there is a constant threat that ISPs use network congestion as a pretext to act on other motives. The Bureau would benefit from a nuanced understanding of the relationship between data cap enforcement and network congestion.

•    How data caps are set. For example, why did AT&T choose 150GB for DSL and 250GB for U-Verse customers? What criteria are used to arrive at the cap? What criteria are used to determine appropriate overage fees?

•    How data caps are evaluated on an ongoing basis. Data caps are often defended as necessary to address current network congestion issues. However, network technology is constantly being modernized and made more efficient. In light of this pattern of improvement, caps that are appropriate for today’s network may be inappropriate in the future. How do ISPs evaluate their existing caps, and what are the conditions under which those caps could be raised and/or eliminated?

Public Knowledge and New America’s Open Technology Initiative hope that the Bureau recognizes the importance of a complete understanding of data caps and acts purposefully to enhance its understanding.

Respectfully Submitted,

Harold Feld
Legal Director
Public Knowledge

Sascha Meinrath
New America Foundation’s
Open Technology Initiative

cc: Chairman Julius Genachowski
Commissioner Michael Copps
Commissioner Robert McDowell
Commissioner Mignon Clyburn
Commissioner Meredith Attwell Baker