FCC’s 2020 Look at Broadband Speeds

The FCC recently released a Notice of Inquiry asking about the state of broadband in preparation for the upcoming 2021 report to Congress. The FCC is required to annually examine the state of broadband and this is the sixteenth NOI that is the first step towards creating the annual report.

In the NOI, the FCC provides a preview of what they are planning to tell Congress in the upcoming report. The FCC continues to pat itself on the back for closing the digital divide. Consider the following facts cited by the FCC in their opening paragraphs of the NOI:

The number of Americans lacking access to fixed terrestrial broadband service of at least 25/3 Mbps continues to decline, falling more than 14% in 2018 and more than 30% between 2016 and 2018. In addition, the number of Americans without access to 4G Long Term Evolution (LTE) mobile broadband service with a median speed of at least 10/3 Mbps fell approximately 54% between 2017 and 2018. The vast majority of Americans, surpassing 85% of the population in 2018, now have access to fixed terrestrial broadband service at 250/25 Mbps, representing a 47% increase in the number of Americans with access to this speed since 2017. Over the same period, the number of Americans living in rural areas with access to such service increased by 85%. 

These statistics all sound great, but unfortunately, we can’t believe any of these claims. The FCC continues to draw conclusions based upon the badly flawed Form 477 data reported to the agency by ISPs. In every rural county I have examined, there are overstatements by ISPs of broadband speeds and availability – and those overstated coverages are included by the FCC as places that have good broadband. I’ve written blogs about entire counties that the FCC thinks haves good broadband, but where the ISP-reported broadband doesn’t exist. If the FCC’s NOI was being truthful, the above list of statistics would open with this sentence: “This report summarizes the broadband speeds and coverage that ISPs report to us. We have no way to know if any of these claims are true”.

Anybody who digs into the FCC data knows it’s terrible, but it’s impossible to know how bad it is. We get clues every time somebody takes a stab at developing a more accurate broadband map. The State of Georgia undertook a mapping effort and in July identified 507,000 homes in the state that don’t have access to 25/3 broadband. That number was 255,000 homes higher than what was shown by the FCC. If that same ratio holds everywhere in the country, then there are twice as many homes without broadband than what the FCC cites in the NOI. I think in many western states that the FCC data is even worse than what Georgia found.

What I find most troublesome about the NOI is that the FCC is planning to stick to the definition of broadband as 25/3 Mbps. It’s easy to understand why the agency wants to keep this speed as the definition of broadband. If the agency increases the definition of broadband, then overnight a whole lot more homes would be declared to not have good broadband. That would completely kill the FCC’s narrative that they are doing great work and closing the digital divide.

The FCC’s cited statistics argue against 25/3 Mbps as the right definition of broadband. Consider the statement above which says that 85% of homes have access to 250/25 Mbps broadband. If that is true, then almost by definition, the FCC’s should define broadband at least at 250/25 Mbps. After all, the FCC’s mandate from Congress is to measure and close the gap between urban and rural broadband. If urban broadband can deliver 250/25 Mbps to everybody, then by the Congressional mandate the speeds available urban America should be the target for rural America. To keep the definition at 25/3 Mbps is ignoring market reality – that most of the people in the country now have speeds far faster than the FCC’s obsolete definition of broadband. And perhaps worse of all, the FCC is drawing this conclusion based upon 2018 data. We know that homes are using roughly 50% more broadband today today than what they used in 2018.

There They Go Again

The FCC issued the 2020 Broadband Deployment Report on April 20. It’s a self-congratulatory document that says that the state of broadband in the country is improving rapidly and that the FCC is doing a great job. I had a hard time making it past the second paragraph of the report which summarized the state of broadband in the country. Consider the following:

The number of Americans lacking access to fixed terrestrial broadband service at 25/3 Mbps continues to decline, going down by more than 14% in 2018 and more than 30% between 2016 and 2018. 

The FCC has no factual basis for this statement because they don’t know the number of households in the US that don’t have access to 25/3 Mbps broadband. The numbers cited are based upon the Form 477 data collected from ISP that everybody in the country, including the FCC, has acknowledged is full of errors. The FCC has proposed moving to a new method of data collection that will produce maps based upon drawing polygons that they hope will fix the rural broadband reporting problem. 

I’ve been working all over the country with rural counties and I have yet to encounter a rural county where the Form 477 coverage of 25/3 broadband is not overstated. In county after county, we find places where the big telcos and/or WISPs exaggerate the broadband speeds that are available and the coverage area available for faster speeds of broadband. The reporting problem is getting worse rather than improving as witnessed by a recent filing by Frontier to the FCC that claims they have improved speeds in 16,000 rural Census blocks to 25/3 Mbps broadband since June 30, 2019. This claim is made by a company that just went into bankruptcy and which the whole rural industry knows is not spending a dime on rural infrastructure. There were similar claims made by the other big telcos in a proceeding that was to determine the areas available for FCC grant funding.

The number of Americans without access to 4G Long Term Evolution (LTE) mobile broadband with a median speed of 10/3 Mbps fell approximately 54% between 2017 and 2018. 

There has been a lot of rural cell sites upgraded from 3G to 4G as the big cellular carriers want to mothball 3G technology. However, any quantification of the improvement of cellular broadband coverage is suspect due to blatantly erroneous reporting by the big cellular carriers. In 2019 when the FCC went to award grant funding to upgrade rural cellular coverage the discovered that Verizon, T-Mobile, Sprint, and US Cellular had significantly overstated rural cellular coverage in an attempt to shuttle grant funds away from smaller cellular carriers. The FCC reacted by yanking that grant program and delaying it, in what is now called the 5G Fund. It’s hard to believe that the FCC would try to quantify the improvement in 4G coverage between 2017 and 2018 without acknowledging that this was the coverage that was badly overstated by the cellular carriers.  

AT&T, Sprint, T-Mobile, and Verizon are also rapidly expanding their 5G capability, with 5G networks in aggregate now covering the majority of the country’s population, especially in urban areas, and more live launches planned for 2020.

The FCC clearly buys the 5G hype from the cellular companies which are claiming widespread 5G coverage. The cellular companies have introduced new spectrum into their 4G LTE environment, and the cellular marketers have labeled this as 5G. Much of the first wave of new spectrum being used is in lower frequency bands such as 600 MHz for T-Mobile and 850 MHz for AT&T. These lower frequency bands don’t carry as much data as higher frequencies and won’t be delivering faster broadband. However, new spectrum bands improve the chances of grabbing a channel to get the data speeds that 4G was already supposed to be delivering.

5G will not arrive until the carriers begin implemented the new features described in the 5G specifications. For now, none of the important new 5G features have yet made it to the market. So, contrary to the FCC telling the public that 5G is nearly everywhere, the truth is that it is not yet anywhere in the country. I’ll be curious in a few years to see how the annual FCC reports on broadband describe the actual introduction of 5G features. It’s likely they’ll parrot whatever language the cellular marketers spin by then.

This opening pat on the back is followed by page after page of broadband statistics that are based upon the lousy Form 477 reporting from ISPs. There is almost no statistic in this report that is entirely trustworthy.

This report is unfortunate in many ways. The FCC feels compelled to exaggerate broadband coverage so that they can’t be forced to try to fix broadband gaps. The sad aspect of this report is that this is the statistics cited in this report are used to determines which parts of rural America are eligible for broadband grants – and this report is largely a fairy tale. It would have been nice if the summary of the report had acknowledged that the FCC knows that their data is faulty – something they have openly recognized in other dockets. Instead, the FCC chose to spin this fanciful tale of rapidly improving broadband that does little more than provide cover for the FCC to not have to fix rural broadband.