This is clearly a shot across the bow for companies that currently bring fiber connectivity to rural cell sites. There are numerous rural middle-mile networks that mostly survive by providing backhaul to cell sites. While there has been downward pressure from the cellular carriers on transport rates – it’s likely that Starlink or other satellite providers could drop the bottom out of the market pricing for transport.
Since we hear so much about how the US is losing the 5G war, people may not realize how far the cellular networks around the world are behind those in the US and other developed countries. According to statistics from GSMA, in 2020 there were 7.9 billion cellular users in the work, 48% who were still using 2G or 3G cellular technology. The percentage of users on older technologies is expected to drop to about 23% by 2025, with a big transition to 4G.
But even then, cellular data speeds are likely to remain slow in many countries due to the lack of fiber backhaul and to the fact that in many countries the vast majority of people get almost all of their broadband from cellphones.
It’s been predicted for many years that satellites would play a big role in supporting cell sites. The worldwide consulting firm NSR predicted last year that there would be 800,000 cell sites worldwide connected via satellite by 2029. Over that same time period, NSR predicts the US market for satellite backhaul at $39 billion.
But it’s still a bit of a surprise to hear Starlink talking about providing cellular backhaul. A rural cell site is a large data user and requires far more bandwidth than the average residential or business customer. It would be a big challenge to Starlink or any satellite network to carry both cellular backhaul and residential broadband – because the cellular backhaul would suck away a lot of the capacity of the network out of any one satellite. One would think that cell sites would get priority routing, which means other broadband users would suffer.
It’s been less than five years since the new generation of satellite companies said they would be launching big constellations in the overhead sky. My first thought when I first heard of the new satellite technology is that they would be far better off financially by supporting a handful of cellular companies rather than million of residential customers. What I never expected is that somebody would try to handle both on the same network.
And perhaps that’s not Starlink’s plan. The company has been talking about launching 30,000 satellites over time (currently at 1,500). It would be possible to have different satellites for different customers with a constellation that large. But Elon Musk’s disclosure made it sound like discussions about cellular backhaul are already in the works.
I think we’re many years away from fully understanding how satellite companies will operate. It’s possible that cellular companies and big corporate users will make it worthwhile for the satellite companies to give them priority over residential broadband customers. It’s not hard envisioning satellites providing connectivity to large fleets like FedEx, UPS, or the US Postal Service. Satellite broadband could become the connectivity of choice for the large trucking companies. It’s going to be hard for a constellation owner to say no to those kinds of opportunities – but saying yes to big commercial opportunities will means diluting the broadband available for residential customers.