After rumors leaked a few weeks ago, that SpaceX was attempting to acquire Globalstar, Amazon has made its move – agreeing to acquire the satellite operator for around $11 billion. Hoping to ramp up Amazon Leo, and inheriting Apple as a flagship customer, the deal is an acknowledgement that Amazon is behind schedule.
What most (perhaps all) coverage is missing is that this is a deal for spectrum. Globalstar has a global license for the 3GPP-approved n53 band, running as a narrowband 11.5 MHz mid-band channel (2,483.4 to 2,495 MHz), as we explored at MWC.
Satellite spectrum is hard to come by. Months, if not years, of FCC lobbying saw SpaceX seize EchoStar’s AWS-4 (around 40 MHz total) and H-block spectrum (some 10 MHz) back in September, for some $19 billion. With that tied off, it leaves a few companies with global FCC-approved MSS (Mobile Satellite Services) licenses.
Globalstar is one of them, alongside Iridium and Viasat. In 2024, the FCC added the Supplemental Coverage from Space (SCS) extension to its Part 25 ruleset, which essentially enabled terrestrial operators to combine satellite-based networks with their own terrestrial networks too – opening the door to D2D, via MSS spectrum.
To this end, in the US, AST SpaceMobile took a strange 80-year license to distressed Ligado’s L-Band MSS spectrum (a total of around 40 MHz), for a lump-sum of $550 million and then $80 million per year. This is what AT&T and Verizon are banking on, with T-Mobile opting for SpaceX.
Now, on the D2D front, Apple’s ~$1.5 billion 2024 deal (which included a 20% equity stake) with Globalstar was always an uncomfortable topic for the MNOs, as any iPhone user would not need the MNO’s D2D deal to provide emergency messaging capabilities – the main draw of D2D, as terrestrial network capabilities are generally pretty good. In the US, where the iPhone makes up the majority of smartphones in the installed base, this means that most D2D-connected smartphones could be carried by Globalstar, rather than AST SpaceMobile or Starlink.
So, from a US perspective, this is Amazon buying into the D2D market, as a main provider to the largest smartphone platform, with Amazon able to provide D2D services to its vast array of in-house devices that power the Amazon retail operation.
Outside of the FCC jurisdiction, that trio (Globalstar, Iridium, ViaSat) are able to provide global services, meaning Amazon has scope to targeting other D2D deals besides Apple. According to the terms of the acquisition, Apple can either take cash or Amazon stock, for its 20%. The acquisition announcement did confirm that Amazon Leo will connect iPhones and Apple Watches going forward.
Amazon Leo is obviously still a work in progress, but the Globalstar platform deftly adds D2D to its bow. Amazon notes that it will begin to construct a “next-generation D2D satellite system” in 2028, presumably based entirely on the in-house Amazon Leo satellite designs. As such, the spectrum is the key justification of this deal, rather than the satellite operations, but inheriting a customer like Apple is not to be sniffed at.
Amazon says the next-gen system will “offer substantially higher spectrum use and efficiency than legacy direct-to-cell systems, which translates into faster speeds and better performance for customers.
It will also integrate seamlessly with Amazon’s first- and second-generation Leo systems, forming a powerful, unified network that combines fixed and mobile satellite services to support a wide range of customers and use cases.”
Still, it looks like the Apple-funded next generation of Globalstar satellites will be deployed. These are being manufactured by MDA Space, with Apple thought to be very influential in the design.
“Apple and Amazon have a long and proven track record of working together through Amazon’s core infrastructure services, and we look forward to building on that collaboration with Amazon Leo. This ensures our users will continue to have access to the vital satellite features they have come to rely on, including Emergency SOS, Messages, Find My, and Roadside Assistance via satellite, so they can stay safe and connected while off the grid,” said Apple’s SVP Worldwide Product Marketing, Greg Joswiak.
Blue Origin interplay
Back in January, the Jeff Bezos rocketry venture Blue Origin decided it was going to launch its own satellite broadband constellation (called TeraWave), which somewhat stepped on the toes of Amazon’s recently rebranded Amazon Leo (formerly known as Project Kuiper).
Blue Origin was meant to be a key launch provider for Amazon, enabling a way to get relatively cheap satellite launches that might help it remain competitive with SpaceX and its Starlink constellation.
With SpaceX’s long-rumored IPO meant to be filed shortly, we will finally get a look at the financials of Starlink, and see how much cash the remote broadband venture generates. Starlink reportedly has over 10 million subscribers, and at an average monthly ARPU of $70, this does imply annual revenues of around $8.4 billion annually. So, what are its costs?
Amazon Leo has around 240 satellites in orbit, with plans to eventually reach around 3,200. SpaceX already has around 10,200 satellites in operation but has launched around 12,000 for its own use since 2019.
Blue Origin’s New Shepard is the space tourism vehicle that launches celebrities to a very underwhelming ~100 km sub-orbital height. New Glenn is the heavy-lift launch vehicle. After years of delays, it carried out its first successful launch in January 2025. In November, it performed the second launch, successfully recovering the reusable first-stage booster, having deployed NASA’s Escape and Plasma Acceleration and Dynamics Explorers (ESCAPADE) project.
The third launch took place this past week, reusing the booster for the first time but replacing all of the engines. Unfortunately, the second-stage vehicle placed the satellite payload in an “off-nominal orbit,” which is a polite way of saying ‘in entirely the wrong place’ – at an elliptical orbit that ranges between 154-494 kilometers, instead of a neat circular orbit at a height of 460 km. Each time the satellite reached the bottom of its orbital height, gravity would pull its inevitable downward spiral that little bit closer. The first-stage was recovered successfully, at least.
Incredibly, the payload for this launch was Blue Origin and Amazon Leo rival AST SpaceMobile’s BlueBird 7 satellite – key to expanding its D2D capabilities. The stock market reception saw AST SpaceMobile’s share price slump 15%, but the company is adamant it will have up to 45 satellites in orbit by the end of the year.
The satellite evidently does not have enough fuel to correct this orbit, but Blue Origin appears to have thoroughly missed the mark. Speculation suggests an engine failure in the second stage vehicle has caused this problem, making it miss its target trajectory, but Blue Origin killed its livestream webcast when the first stage successfully separated, so we do not currently know what happened in the control room.
The satellite must now be de-orbited safely, burnt to a crisp in the atmosphere. Notably, perhaps even negligently, AST SpaceMobile has stated in an SEC filing that its insurance coverage only covers between 3-20% of the cost of each satellite and its launch.
So, is Blue Origin the vehicle that Amazon Leo must hitch its wagon to? The short answer is no; it must continue to rely on arch-rival SpaceX, which it has used three times to launch the initial batch of ~70 satellites. It now has around 240 satellites in orbit.
Amazon has already bought capacity from the United Launch Alliance (ULA, a joint venture between Boeing and Lockheed Martin) and ArianeGroup. It likely regrets signing contracts with Blue Origin. ULA and ArianeGroup have planned to rely on Blue Origin engines, for use in their own rockets.
Both have faced major scheduling delays due to these delayed engines, forcing them to continue to use old Russian designs.
Blue Origin planned for 6-8 launches last year, but managed only two. A best-case scenario might see it make 10 launches this year. SpaceX managed around 160 last year, for context.
Amazon documents show that it paid $1.8 billion to Blue Origin in 2025, out of a total of $2.2 billion paid for satellite launch capacity. There is an Amazon shareholder proposal that calls for a new mandatory independent board chair, intended as a counter to Bezos’ external conflicts of interest ($1.8 billion!), as he remains Amazon’s executive chairman, as well as being Blue Origin’s sole owner.
The MVNO question
An interesting point is the matter of why Amazon has not launched its own MVNO to date. It went as far as designing the Fire Phone, and as we explored recently, it is reportedly due to re-enter this hardware market.
This might provide the impetus to launch an MVNO, and post-Globalstar, it would be in a strong position to use its D2D capabilities to reduce its reliance on MNOs substantially. As we have speculated previously with Apple, this would have to be done by heavily relying on WiFi networks – aiming to combine satellite and WiFi as the primary connection, and using the MNO wholesale capacity as the secondary option.
This is still more in the realm of quackery than reality, but it is technically possible today and would rapidly diminish the importance of the MNO. The WiFi Alliance’s Passpoint and the Wireless Broadband Alliance’s OpenRoaming are two examples of how automated login via MVNO-issued credentials could allow WiFi to carry the bulk of traffic, with satellite there to (eventually) serve data traffic when the user is out of range of a compatible WiFi network – with the device ready and able to finally roam onto an MNO network when the other two options have not worked.
We made the case for the MVNO reverse merger ahead of MWC 2025. Industry opinions have shifted somewhat, but most feel it is a far-fetched scenario.
The arrival of eSIM in devices has opened the door to MVNO competition, and this toothpaste is rather hard to put back in the tube.
With Amazon LEO due to provide home broadband via the Prime subscription package, as well as augmenting all manner of operational Amazon equipment and facilities, these D2D capabilities almost mandate that Amazon launch a Prime mobile service.
While it might be reliant on the MVNO agreement with an MNO in the early days, for a company with such deep pockets, moving from MVNO to MNO via terrestrial network construction or acquisition is a likely outcome.
There is no smoke without fire, and in 2023, Amazon was rumored to be preparing an entry into the US mobile market via Dish. The Boost Mobile debacle suggests that it is better to go your own way and acquire a satellite constellation than rely on the US’ federally-mandated fourth carrier to become competitive, even if parent company EchoStar also had LEO ambitions. That feels like a fairly strong condemnation of the competitiveness of the industry.