Last week, SpaceX received permission from the FCC to use spectrum to create a satellite-based broadband service known as Starlink. I can’t help but think that the timing of the deal was quite perfect – as SpaceX talked about superfast broadband from space, China’s old space station, the Tiangong-1, was hurtling out of control towards Earth, reminding us that this space stuff is not that easy, and doesn’t stay up there forever.
It’s hard to bet against Elon Musk. The man is making quite a career out of dreaming big, and succeeding. He does, after all, throw enormous rockets up into space that (usually at least) come back to be reused, dig tunnels to circumvent traffic issues (useful for all those Tesla drivers no doubt), and even manage to sell flamethrowers to the public (and gets away with calling it “not a flamethrower” to avoid any issues). And, let’s not forget, his ultimate dream of getting mankind to Mars, which sounds almost feasible when he says it. So, why, when he takes on the seemingly impossible on a regular basis, am I less than convinced about an Earth-orbiting high-speed broadband solution?
My reluctance to believe starts with the fact that we’ve been down this road before. Twenty-some years ago two earlier pioneers, Bill Gates and Craig McCaw, launched Teledesic with a very similar goal of launching low earth orbit satellites to provide global broadband service. Craig McCaw was, in many ways, the godfather of the U.S. mobile services that we all take for granted today, creating the first national provider (McCaw cellular) that he later sold to AT&T. In other words, the man knew his mobile stuff, while Bill Gates, of course, understood the need for speed for computing. And yet, the venture failed to get off the ground due to the economics. At roughly the same time, Iridium, a Motorola-backed venture, had a similar goal that also ended poorly. Indeed, while Iridium does still exist today, the original version failed and was sold for less than the cost of decommissioning the satellites. Yes, it was cheaper to keep the aging satellites in the air than it was to close the business and bring them down in a controlled fashion.
Of course, 20 plus years is many lifetimes in terms of technology, and SpaceX has a couple of advantages over the previous failed ventures. Since they own the rockets needed to get these satellites into space, the set-up costs are less. Additionally, the satellites are now smaller and cheaper to build than they were in Teledesic’s days. That’s the good news. The bad news is that the LEO solution requires around 800 satellites, and they seem to have an average lifespan of five years, which is a pretty hefty hardware replacement cycle regardless of how cheap the actual satellites are. And while technology has helped make the venture more feasible in terms of satellite technology, it has also hindered it, as more of the globe is now covered by high-speed broadband services, meaning that the addressable market for the SpaceX venture is far harder to comprehend.
Between the wired broadband infrastructure that is available today and the promise of 5G – both mobile and fixed – there is far less opportunity for a broadband internet solution in wealthier markets. Where there is a huge opportunity is in less developed worlds that have limited, or no, Internet access today. But here’s the catch: Elon Musk is not building this solution as a philanthropic proposition. Rather this, as with his other “big bets” is supposed to ultimately fund the Mars venture and that seems to be in direct conflict with the size of the opportunity.
Even more challenging is the fact that SpaceX is not alone in its race to build a LEO-based broadband solution. OneWeb, which is backed by Softbank, has priority from the UN to use the radio spectrum needed for a broadband solution too. So with a questionable market need, unless the goal is to provide broadband to underdeveloped markets that will pay far less, and two companies chasing the same gold, the chance of success becomes even harder to imagine. And indeed, there are several other contenders, such as Inmarsat, Globalstar, and the re-born Iridium that are already in the sky (although all three use far fewer satellites, but result in greater latency).
It’s hard to bet against Elon Musk’s winning streak to date, but the entire history of satellite broadband services tells us that wishing on space hardware is not a good move. At best, we may see OneWeb and SpaceX’s LEOs combine into a single solution over time to improve the economics. At worst, we’ll come back to the topic 20 years from now when someone else decides to take on the challenge.