FCC just cleared SpaceX's 15,000-satellite Starlink Mobile — would you actually switch carriers?
The FCC approved SpaceX's Starlink Mobile constellation on Oct 6 — 15,000 satellites, plus a waiver so they can offer wireless service without a spectrum lease with a terrestrial carrier. SpaceX is talking about up to 150Mbps to a phone and an upgrade path they claim is '100x better' than today's direct-to-cell once the EchoStar 65MHz spectrum and next-gen satellites come online, with a real consumer push against AT&T, Verizon, and T-Mobile by late 2027. The skeptical case: 65MHz of spectrum vs roughly 300MHz per Big Three carrier, and the current T-Satellite service through T-Mobile runs around 4Mbps. My phone is paid off, I'm in a city with good coverage, and I don't have a coverage problem to fix. For people who'd actually consider switching: is there a real 'fourth carrier' argument here, or is this a rural-dead-zone product with great PR?
2 answers
- Run the spectrum math: 65MHz vs 300MHz
The 150Mbps number is a peak per-user figure for an uncongested cell, not average throughput you'll see on a Tuesday in Manhattan. Here's the part that matters: SpaceX holds about 65MHz of usable spectrum via the EchoStar deal. AT&T, Verizon, and T-Mobile each hold roughly 300MHz. Spectrum is the hard ceiling on capacity, and in cities the constraint is capacity, not coverage. So yes, this is a real product for dead zones and rural corridors — filling coverage gaps is exactly what satellites are good at. But as a reason to switch in a city where all three carriers already work? I'd want to see the actual per-cell capacity numbers from the next-gen birds before I ported my number. Check the FCC grant language: it waives the leasing requirement, it doesn't conjure spectrum.
- The FCC waiver is the actual story
The headline number everyone cites is 150Mbps, but the regulatory move is more important: the FCC waived the requirement for a spectrum-leasing agreement with a terrestrial operator. That means SpaceX can sell mobile service on its own spectrum instead of renting permission from the carriers it wants to compete with. The buildout plan is also unusual — base stations mounted on the same hardware as Starlink broadband dishes instead of traditional towers, which is how they claim capex efficiency. Whether that works at scale is unproven. I wouldn't switch now; I'd watch what they actually file for the terrestrial buildout in 2027, because that's where the capacity math either holds up or falls apart.