If this law is enacted, the distribution of most of the BEAD grant funds to States would be delayed by at least six months, probably longer. The NTIA has already said that it intends to announce the allocation of the $42.5 billion in grants to the states on June 30. The funds are supposed to be allocated using the best count of unserved and underserved locations in each state on that date. Unserved locations are those that can’t buy broadband of at least 25/3 Mbps. Underserved locations are those unable to buy broadband with speeds of at least 100/20 Mbps.
To add to the story, FCC Commissioner Jessica Rosenworcel recently announced that the FCC has largely completed the broadband map updates. That announcement surprised the folks in the industry who have been working with the map data, since everybody I talk to is still seeing a lot of inaccuracies in the maps.
To the FCC’s credit, its vendor CostQuest has been processing thousands of individual challenges to the maps daily and has addressed 600 bulk challenges that have been filed by States, counties, and other local government entities. In making the announcement, Rosenworcel said that the new map has added over one million new locations to the broadband map – homes and businesses that were missed in the creation of the first version of the map last fall.
But the FCC map has two important components that must be correct for the overall maps to be correct. The first is the mapping fabric that is supposed to identify every location in the country that is a potential broadband customer. I view this as a nearly impossible task. The US Census spends many billions every ten years to identify the addresses of residents and businesses in the country. CostQuest tried to duplicate the same thing on a much smaller budget and with the time pressure of the maps being used to allocate these grants. It’s challenging to count potential broadband customers. I wrote a blog last year that outlined a few of the dozens of issues that must be addressed to get an accurate map. It’s hard to think that CostQuest somehow figured out all of these complicated questions in the last six months.
Even if the fabric is much improved, the more important issue is that the accuracy of the broadband map is reliant on two issues that are reported by ISPs – the coverage area where an ISP should be able to connect a new customer within ten days of a request, and the broadband speeds that are available to a home or business at each location.
ISPs are pretty much free to claim whatever they want. While there has been a lot of work done to challenge the fabric and the location of possible customers – it’s a lot harder to challenge the coverage claims of specific ISPs. A true challenge would require many millions of individual challenges about the broadband that is available at each home.
Just consider my own home. The national broadband map says there are ten ISPs available at my address. Several I’ve never heard of, and I’m willing to bet that at least a few of them can’t serve me – but since I’m already buying broadband from an ISP, I can’t think of any reason that would lead me to challenge the claims of the ISPs I’m not using. The FCC thinks that the challenge process will somehow fix the coverage issue – I can’t imagine that more than a tiny fraction of folks are ever going to care enough to go through the FCC map challenge process – or even know that the broadband map exists.
The FCC mapping has also not yet figured out how to come to grips with broadband coverage claimed by wireless ISPs. It’s not hard looking through the FCC data to find numerous WISPs that claim large coverage areas. In real life, the availability of a wireless connection is complicated. The FCC reporting is in the process of requiring wireless carriers to report using a ‘heat map’ that shows the strength of the wireless signal at various distances from each individual radio. But even these heat maps won’t tell the full story. WISPs are sometimes able to find ways to serve customers that are not within easy reach of a tower. But just like with cellphone coverage, there are usually plenty of dead zones around a radio that can’t be reached but that will still be claimed on a heat map – heat maps are nothing more than a rough approximation of actual coverage. It’s hard to imagine that wireless coverage areas will ever be fully accurate.
DSL coverage over telephone copper is equally impossible to map correctly, and there are still places where DSL is claimed but which can’t be served.
Broadband speeds are even harder to challenge. Under the FCC mapping rules, ISPs are allowed to claim marketing speeds. If an ISP markets broadband as capable of 100/20 Mbps, they can claim that speed on the broadband map. It doesn’t matter if the actual broadband delivered is only a fraction of that speed. There are so many factors that affect broadband speeds that the maps will never accurately depict the speeds folks can really buy. It’s amazingly disingenuous for the FCC to say the maps are accurate. The best we could ever hope for is that the maps will be better if, and only if ISPs scrupulously follow the reporting rules – but nobody thinks that is going to happen.
I understand the frustration of the Senators who are suggesting this legislation. But I also think that we’ll never get an accurate set of maps. Don’t forget that Congress created the requirement to use the maps to allocate the BEAD grant dollars. Grant funding could have been done in other ways that didn’t relay on the maps. I don’t think it’s going to make much difference if we delay six months, a year, or four years – the maps are going to remain consistently inconsistent.