The FCC just released a graph and statistics-laden broadband report that summarizes the data collected by ISPs for June 2024. Of course, the data in this report is only as good as what ISPs report to the FCC, so a lot of this information, while interesting, is to be taken with a grain of salt. Let’s look at some specifics.
Figure 1 shows the trend from June 2020 until June 2024 of total fixed (broadband) and mobile (cellphone connections). Fixed connections grew from 118.5 million in June 2020 to 132.6 million in June 2024. That’s 2.4% growth from 2023 to 2024 with an average 3% annual growth over the four years. Mobile connections grew from 361.7 million in June 2020 to 416.1 million in June 2024. That’s 2.5% growth from 2023 to 2024 and an average of 3.8% annual growth over the four years. The problem with the overall numbers is that we don’t know for sure what ISPs are reporting. For example, if an ISP sells a gigabit connection to the landlord of an MDU, is that one broadband connection for the building, a broadband connection to every unit, or both? The FCC numbers are interesting, but the number of broadband connections is not the same as the number of broadband customers. To be fair, the FCC isn’t misrepresenting the numbers, but I promise that users of this report will.
Figure 2(a) gets off the rails because it describes the percentage of connections at various download speeds. According to this chart, there are 7.5 million connections with speeds under 25 Mbps and 16.7 million between 25 Mbps and 100 Mbps. This is a very different story than BEAD, which is currently targeting around 13.8 million locations – 6 million under 25 Mbps and 7.8 million between 25 Mbps and 100 Mbps. There are expected differences in the counts due to grants that have been awarded and not yet constructed. But the real issue with the FCC data is that it represents the speeds that ISPs report to the FCC – not actual speeds. I, and others, have written a lot about ISPs that overreport speeds when you compare the FCC speeds in this graph with Ookla speed tests. There are a huge number of ISPs reporting exactly 100 Mbps download and delivering a lot less. Again, the FCC is not misrepresenting what these numbers are – but anybody who has dug into ISP reporting or paid any attention to State BEA map challenges knows these numbers are not close to accurate.
Figure 2(b) does the same for upload speeds. ISPs report that 27 million locations have download speeds under 20 Mbps, which should make them eligible for BEAD. I think the real number is a lot higher.
With all of that said, the FCC report is useful for seeing the trends of what ISPs report over time. For example, Figure 3 shows how reported gigabit connections (actually 940 Mbps ) has grown from 8.6 million in June 2020 to 34.4 million in June 2024.
After Figure 3 the data becomes muddled. Figure 4 shows the number of broadband connections in service by technology. It shows that 59% of customers are still served by cable technology, and that 24.5% served by fiber. The number that mystifies me is the fixed wireless. It says that 6.8% of broadband customers are served by this technology, which means if there are 132.6 million total connections, there are 9 million fixed wireless connections. As of June 30, 2024, AT&T, T-Mobile, and Verizon together claimed 9.7 million FWA customers, and there needs to be some millions of customers of WISPs added to the FWA number. That chart has me scratching my head about how the FCC compiled these numbers-.
After this, the report goes deeper into detail, but the more detailed reports all come from the same questionable data that builds the basic graphs and charts. Even knowing that there are questionable parts of the data, I found myself fascinated when looking through the report – only to occasionally pinch myself to remind me not to rely on the specific data from most of the many charts.