A recent report released on November 30 by BroadbandNow reveals the FCC grossly overreported the availability of gigabit internet in the U.S. As of 2020, the FCC reported gigabit internet was available to 84% of Americans, up from just 4% in 2016. But according to BroadbandNow, those numbers are closer to 56% in 2020, up from 2.4% in 2016, and could be even less.
The disparity has to do with how the FCC has reported internet coverage in the past. Up until recently, there was a major flaw in Form 477 that ISPs use to report what kind of coverage they offer and where. The old version of the form, which all current internet data provided by the FCC is based on, allowed ISPs to mark an entire census block as “covered” by a specific service even if only one home in that census block actually had that service.
The form did not require ISPs to provide more granular data than that, and as a result, the FCC would count every single house within each census block as having that service. “This has led to widespread issues of over-reporting when it comes to where plans are actually available at the neighbourhood level,” said the report.
According to Tyler Cooper, editor-in-chief of BroadbandNow, “Even the number 56% is likely overstated. Meanwhile, the increase from single-digit availability to more than 50% is good news, but this research finds further evidence of the flaws in Form 477 deployment data, which the FCC relies on from ISPs.”
To get that 56% number, BroadbandNow randomly selected 75 addresses across 15 zip codes in Texas, Florida, and Ohio where the FCC’s data indicates that gigabit coverage is available. It manually entered those addresses into the “check availability” option on various ISPs’ websites that provide gigabit service in those areas or called the providers directly. The entire methodology is explained in the report, but the organisation discovered that none of those addresses had an active gigabit plan available as of 2020.
I did the same for where I currently live. I can get Spectrum gigabit cable broadband (940 Mbps). AT&T Fibre, or “true” gigabit internet, is not available. In fact, AT&T only provides speeds of up to 18 Mbps of cable broadband to my address, but the neighbourhood across the street gets speeds of up to 50 Mbps for the same price according to AT&T’s website. The neighbourhood across the street from that neighbourhood doesn’t get AT&T at all. Both of the neighbourhoods are a five-minute walk from where I live. AT&T Fibre is only available in my city if the homes or apartment complexes were built within the last five years because the cables were being put into the ground at the same time. Anywhere else would need to be dug up, which would be expensive.
To add insult to injury, it seems like the FCC is also aware that despite expanding gigabit coverage across the U.S., be it fibre or cable broadband, the actual adoption rate is low. According to its most recent broadband deployment report that was released on Jun. 8, 2020, the FCC has not changed its definition of what it considers high-speed internet, 25/3 Mbps, because the “Commission’s data shows that in the areas where gigabit service is available, only 4% of Americans living in those areas are in fact subscribing to it.” This is according to the FCC own fixed broadband deployment data from 477 as of December 31, 2018 — the same form that has the aforementioned flaws.
So not only is gigabit internet available to less Americans than the FCC previously reported, it seems that few have actually subscribed to it either because of cost or need. Gizmodo reached out to BroadbandNow for further clarification on this part of the FCC’s report. According to Cooper, it has to do with what information the FCC makes public:
The FCC does in fact collect subscriber rate information alongside census block-level availability data from providers as part of the 477 release. The only difference is that subscription data is not openly released in its entirety like the mainline 477 figures. It seems almost certain to me that the FCC is aware of the inconsistencies in their own methodology, as the 477 dataset has been the subject of constant criticism dating back to its inception. It’s my hope that the incoming administration will take a comprehensive look at how this data is being collected by the agency, as there is much to be gained from reporting a more accurate portrait of gigabit availability and adoption on the ground in the U.S.
Regardless of why only 4% of Americans have access to gigabit internet, the larger issue is internet access as a whole. A generously estimated 42 million households lack access to high-speed broadband internet, and the majority of those households are located in rural areas and low-income urbans areas, with Black and brown residents more likely to lack access than whites. Local municipal broadband providers help residents in those areas get around the monopoly strongholds a single ISP can have over them, but many states have laws that roadblock those types of internet providers.