A Nation of Have Nots
As we approach the holiday season, it may be that what people really want for Christmas is a true broadband connection. After all, a majority of U.S. broadband subscribers just found out what they surf the Web with-be it DSL, cable modem, or fixed wireless-isn't really broadband.
In a stunning declaration, the FCC's latest report states that 68 percent of connections within the United States do not reach the FCC's definition of broadband: 4 Mbps download and 1 Mbps upload speeds.
What started back in mid 2009 as a notice of inquiry (NOI) from the Federal Communications Commission (FCC) on mobile broadband, which we covered in August, 2009, has now become a widespread study on the limitations of broadband within the United States.
Because of all the attention, the FCC released a study in August 2010 that confirmed what anecdotal evidence had shown for years: advertised broadband speeds are more mirage than reality.
"Actual download speeds experienced by U.S. consumers lag [behind] advertised speeds by roughly 50 percent," the August broadband performance paper states. "This gap is similar across technologies and is due to a variety of factors, some within the span of control of providers in their network, and some due to the unpredictability of the Internet. This gap may cause confusion among consumers, as actual speeds, which largely determine the end-user experience, lag [behind] advertised speeds considerably."
Covering the August 2010 study we agreed with the findings, but one of the least defensible aspects of the paper was its methodology: the report relied heavily on comScore data from only two testing locations (Reston, Virginia, and Chicago, Illinois).
Thankfully, for its most recent study, the FCC spent a bit of time culling through its own Form 477 data. The December report proves that advertising rates are highly inaccurate, partly by comparing advertised rates to the self-reported rates each service provider notes on its FCC Form 477.
In other words, it appears there's a publicly-stated speed used on advertisements to entice consumers and a privately stated speed used for self-reporting. Another recent study showed that four out of five Internet users in the U.S. could not identify their broadband speeds, relying instead on the advertising they received from their service provider.
Noting that the Form 477 data is self-reported information from the service providers, the December report reminds readers of the disconnect between advertised speeds and actual speeds.
"The reported connection speed is typically based on the advertised speed of the purchased service, such as the advertised ‘up to' speed of a wired service or the advertised range of ‘typical speeds' of a mobile wireless service," the report states, adding, "It is possible that the purchased service will not operate at its advertised speed at all times. Elsewhere, the Commission is exploring the size and causes of differences between advertised connection speeds and the speeds experienced by consumers."
What the December report reveals, then, is a much larger disconnect than originally anticipated in the August report.
"At year-end 2009, 58 percent of reportable connections (or 76,594,000 connections) were slower than 3 Mbps in the downstream direction," the new report states. "12 percent (or 16,172,000 connections) were at least 3 Mbps in the downstream direction but slower than 6 Mbps, and 30 percent (or 40,382,000 connections) were at least 6 Mbps in the downstream direction."
What about upload speeds? According to the report, half of US "broadband" connections were less than 768 kbps, well below the FCC's definition for broadband.
"At year-end 2009, 49 percent of reportable connections (or 65,942,000 connections) were slower than 768 kbps in the upstream direction," the report states. "39 percent (or 51,536,000 connections) were at least 768 kbps in the upstream direction but slower than 1.5 Mbps, and 12 percent (or 15,670,000 connections) were at least 1.5 Mbps in the upstream direction."
Taking into account, then, the 58 percent of users who were slower than 3 Mbps download speeds, as well as the almost 60 percent of users that were slower than 1 Mbps upload speeds, it is clear that the U.S. as a whole is struggling with the digital divide, even among those who think they have broadband.
In the August report, FCC analysis showed that the average (mean) actual speed consumers received was approximately 4 Mbps, which has now been corrected in the December report. In addition, while the FCC stated in August that the median actual speed in 2009 was roughly 3 Mbps, it appears that data rate is also in question.
One hopes that the FCC will move to a broader range of consumers to track true usage over a period of time, through the use of DSLreports.com and other sites that have a significant base of users reporting their experience weekly or even daily.
What's the practical issue at stake here for streaming? In the words of the FCC, the "full multimedia" user has a need for speed above that of which a majority of U.S. broadband users are being offered:
"The full multimedia user jumps to 4 Mbps solely on the basis of watching live SD video streams," the August report states, pegging HD video content streaming speeds at 6 Mbps.
Even if we account for 720p at 2.5 Mbps, there's little or no overhead for HD streaming for more than 60 percent of US "broadband" users.
We've conjectured that this lack of true broadband could spell trouble for devices like the Apple TV with its HD streaming-only approach, and it does appear that this new FCC report corroborates recent news reports of a high number of customers having trouble maintaining their connections throughout the course of a long-form movie. The reported speeds-while proving that advertised rates were much higher than reality-are still suspect as being a bit higher than reality, a trend that will continue until true broadband comes to the majority of Americans.
This report's findings also lend credence to the often-heard statement of our industry not seeing enough interest in 1080p streaming to justify competing against BluRay quality levels. It turns out the statement is accurate, but not for the stated reason of a lack of interest on the consumer's part. Instead, it's because the majority of users-whether they're urban, multi-tenant unit (MTU) dwellers, or suburban soccer moms-don't have broadband connections to begin with.
One hopes the FCC's move to define broadband as 4 Mbps download / 1 Mbps upload will come with some teeth, to force truth in advertising. It can't come too soon: just yesterday I received a flyer in the mail from CenturyLink (the company that acquired Embarq, which itself started as Sprint) offering me blazing fast broadband for $29.95 per month.
Reading the fine print, though, CenturyLink was defining broadband as 768 kbps download speeds. And the phone companies wonder why everyone who has a choice is moving from DSL to cable, with hopes of at least watching a few YouTube videos without excessive download times.
Anyone who thought Netflix got negative publicity, wait and see what happens to Blockbuster.
The battle between broadcasters and wireless providers heats up with remarks made by FCC Chair Julius Genachowski at CES last week
Study also notes that video is the main driver behind bandwidth usage, and that most mobile data is consumed via wireless cards, not mobile devices
A newly issued FCC Notice of Inquiry is intended to protect consumers while fostering innovation, and could have significant impact on mobile video adoption as well as next-generation networks.
Mon., Aug. 31, by Tim Siglin