Evaluating claims about unlicensed fixed wireless

The wireless industry is out with a new paper that claims, “The bias [towards fiber to the home] ‘could increase costs by upwards of $30 to $60 billion depending on the distribution of fiber deployment costs for the unserved locations.’” It also says “[excluding unlicensed fixed wireless] ‘unambiguously adds’ at least 1.9 million new locations calling for government-funded overbuilding with BEAD funds”.

As both my analysis, and a separate analysis from ACA, have shown, there is enough money to fund FTTH in the vast majority of cases and in all but a few of the most expensive states. At this one moment in time, there’s enough money. We shouldn’t be looking for shortcuts.

As for the number of locations that are excluded by not including unlicensed fixed wireless, my numbers say 1.79 million fewer locations would be Unserved. In the chart below, reading Unserved left to right is what the current maps show, resulting in 7.8 million Unserved. Reading Unserved up-and-down is what they’d show if you included unlicensed fixed wirless — 6 million. It’s possible this difference between my number and that paper is methodological. The more important point is the Underserved.

Yes, including unlicensed fixed wireless would result is less Unserved, but more than half of the locations would be Underserved. Of the 1.79 million fewer Unserved, 975,000 became Underserved. A better metric would be how many Unserved and Underserved become Served. By that metric, unlicensed fixed wireless reduces the Digital Divide by 1.4 million.

Let’s look at some random examples (these are literally random locations where Unserved becomes Underserved or Served). First is a farm in Oklahoma that is Unserved. If the fixed wireless offering were included, this farm would move to Underserved because of a 25/3 fixed wireless offering.

Next up is a Unserved home in Nebraska, that would be Underserved with the fixed wireless offering. Much of this neighborhood is Served with a DSL offering at 100/20 or better, so it’s not clear that BEAD money could be used in this neighborhood since a project area needs to be 80% Unserved. Not only would I try to bring fiber to this home, but also all the DSL-only homes in this neighborhood.

Lastly, a lakeside home northeast of East Lansing, Michigan that would be Served because of two fixed wireless providers offering 100/20 service. This neighborhood is actually reasonably dense, and not super rural (by rural standards).

One of these providers doesn’t offer service to this address on their website — it produces a “Currently unavailable” message. The other provider doesn’t have a map or an address lookup tool. They say 100 Mbps “5G” service is available in select areas; and “4G” with throughput up to 20 Mbps “covering most of Central Michigan”.

This is probably a good time to say again that I’ve met a couple WISP CEOs and I have tremendous respect for their entrepreneurship, technical creativity, and passion for connecting people. I also try to follow technical advances in radios, and there’s been some exciting things happening on that front. I also think the distinction between fixed wireless using unlicensed spectrum and licensed spectrum is silly.

If we’re thinking about what’s best for the families that live in these homes — now and in the future — the best path forward, in my view, is clear: make the fiber go as far as possible.