Will SpaceX Starlink displace fixed broadband in the UK?

Investment bank Morgan Stanley has just upgraded its valuation of Elon Musk’s SpaceX to $100 billion. $81 billion of that valuation is based on SpaceX’s satellite broadband venture, Starlink, which Morgan Stanley believes has the potential to attract between 235 million and 364 million subscribers globally.¹

SpaceX has achieved fantastic progress, so full credit to Elon Musk and the SpaceX team. The question I consider here, is what impact Starlink might have on the UK fixed broadband market, which accounts for three-quarters of Zen’s revenue.

In summary, I don’t think it will have much impact at all. Sorry Elon!

In the UK, I believe SpaceX Starlink will have a very niche target market – those people and businesses that are out in the middle of nowhere and so cannot get conventional high-speed broadband.

The 2020s will, without doubt, be the decade of full fibre in the UK, with FTTP (Fibre To The Premises) becoming the dominant broadband technology. Openreach has a target of 15 million premises passed with FTTP by 2025 and 20 million by the “mid- to late-twenties”. Virgin Media continues to invest in its network. And challengers like CityFibre have substantial funds and a mission to build sizeable competing full fibre networks.

Building a full-fibre network is an expensive business, but once it’s built, it has a comparatively low operating cost. At the end of the day, it’s just a load of glass in the ground – OK, that’s a bit of an over-simplification, but the point is that once the network is built, it’s relatively simple to operate and maintain. It’s also infinitely upgradable, since the fibre itself can carry faster and faster bandwidths by simply changing the equipment on each end of the fibre – a task that is much simpler than installing the fibre in the first place.

Compare that to deploying and operating a constellation of Low Earth Orbit satellites, and there is a massive cost differential. The cost of carrying the data is substantially higher using satellites, so that cost will sooner or later need to be paid for by the customers of that service. Satellites will also not have the easy upgrade path of fibre in the ground, since new and upgraded satellites would likely need to be launched.

The next issue with satellites is latency, or “ping time” – how long a data packet takes to get between the customer and the particular part of the Internet that they are accessing.  Latency has historically been a significant issue with satellite broadband, since data has taken a lot longer to make the journey into space and back down, compared to the time it takes on a conventional fixed broadband line. This is not an issue for things like video streaming, but is a disadvantage for other applications, and a showstopper for online gaming, when having a low latency connection can mean the difference between winning and losing.

Impressively, the Starlink team has demonstrated a latency of under 20 milliseconds, which is comparable to the latency of fixed broadband.² It remains to be seen whether this low latency can be maintained as users start to load up the service.

The final issue with satellite broadband is a purely practical one. Who wants to go to the trouble of bolting a satellite dish on the side of their house, when they don’t have to?

Taking all of the above together, I maintain that SpaceX Starlink will be very welcome for people in remote areas, or in developing countries that lack broadband infrastructure, but that here in the UK, its market penetration will be very limited. I believe it will be virtually no threat to FTTP, which Zen is already selling at 1 Gig speeds today.