Single Point of Failure: When All Launches Lead to SpaceX

I’ve written a few analyses that noted that relying on SpaceX for spacecraft launches is a problem. A few folks have asked me to clarify that observation.

Before starting this analysis, I must be clear–neither Arianespace nor ULA are nonoperational. And indeed, other nations launch rockets. But for global space economy purposes, China’s space sector and rocket launchers don’t interact meaningfully with the rest of the world. Russia just volunteered to remove itself from space sector relevance altogether. The remaining space nations don’t launch near as frequently and may not have the mass lift capability of Arianespace, SpaceX, and ULA.

Bottom Line Up Front: no redundancies to SpaceX and its rockets. None.

That statement does not indicate I have a problem with SpaceX itself. I may lean, in fact, toward hoping SpaceX succeeds in any endeavor. That hope is constantly fueled because it appears to be the only company and group willing to see risk as an opportunity and not a disadvantage. The company’s behavior, failures, and successes make it the most interesting space company to observe and write about. And that is, frankly, easy for SpaceX to do when its industry competitors are committed to politics and processes and not production. 

My sentiment also doesn’t mean there aren’t other available launch services, and that I hate them. Arianespace and the United Launch Alliance (ULA) still exist, but both companies have adopted extremely limiting and shortsighted strategies. Again, I don’t hate them. In fact, I would like to see either or both of them suddenly produce viable launch systems that cost much less to use than a Falcon 9 or a Starship. However, both companies have squandered more resources and opportunities than SpaceX has ever had access to. And neither company is poised in any way to compete with SpaceX.

The result is what we are witnessing currently–both ULA and Arianespace have rocket inventories, but, for example, the remaining inventory of Ariane 5s is already allocated to contracts. ULA is in a similar situation with its Atlas 5 and Delta IV rockets. The companies’ replacement rocket development delays are exacerbating their inability to launch. Current operational rockets are already spoken for, a circumstance that already makes potential customers dismiss ULA altogether. With Arianespace’s sudden loss of the Russian Soyuz, it can offer only its Vega–a rocket with dubious reliability characteristics and limited spacecraft lift capability. Both companies can’t field new rockets, much less compete with SpaceX.

On the other hand, SpaceX seems to be able to ratchet up its launch cadence at will. The company’s first-stage reusable rockets are an asset, as it doesn’t need to manufacture rockets as more customers step forward. That SpaceX is supplying the least expensive and most reliable launch service makes it desirable to many commercial customers, to begin with. That the company can launch its rocket not just more quickly but predictably is a capability unheard of in the “bespoke” rocket launch industry. Combined, SpaceX’s Falcon 9 has obvious virtues attractive to companies exploring launch options. During the past six years, we have witnessed the partial implementation of a new business model in the industry thanks to SpaceX–a launch SERVICE.

In the past, I’ve labeled most rocket launch manufacturers as launch service providers. However, the reason for that is that even though they manufactured the rockets, they also provided the service to launch them. Some deviations exist, but Arianespace (except for Soyuz), ULA, SpaceX, Rocket Lab, Virgin Orbit, etc., all build and launch their rockets. The upshot is that each company must have customers not just willing to use their service–those customers must also pony up the money to manufacture the rockets. 

With SpaceX’s Falcon 9 partial reusability, that gets a little muddled as the second stage is expendable. Still, SpaceX’s customers aren’t buying the whole rocket, which should come to significant savings–more than what SpaceX is offering today. If Starship is fully reusable (as the company believes it will be), SpaceX will be offering the first true launch service. 

Generally, this should all be seen as good news (at least for SpaceX), so what’s the problem?

Lack of Choice

The problem is that the commercial free world industry is coming to rely on SpaceX for launch services. Worse, companies are forced to use SpaceX in some cases because what were thought to be viable competitors aren’t. OneWeb, for example, is finding itself in the awkward position of needing to use a competitor’s service to finish its satellite deployments. In a previous analysis, I wondered whether the company’s owners might be willing to move the remaining satellite launches to SpaceX:

OneWeb’s owners may see using SpaceX’s Falcon 9 as unacceptable–but at the same time, they may come to terms with the value proposition that company offers. If not SpaceX, there are potentially other options, but they impact OneWeb’s bottom line much more significantly and can’t launch 30+ of its satellites at a time. India’s Polar Satellite Launch Vehicle could be an option–maybe Arianespace’s Vega–but they and other competitors (who aren’t SpaceX) don’t launch as frequently as Soyuz and certainly don’t have the mass capacity. Adding those factors up may put OneWeb behind deployment schedule.

This week, OneWeb answered my pondering as its CEO announced it would go with SpaceX for launching the remainder of its satellites:

“We thank SpaceX for their support, which reflects our shared vision for the boundless potential of space. With these launch plans in place, we’re on track to finish building out our full fleet of satellites and deliver robust, fast, secure connectivity around the globe.”

The decision to go with SpaceX is as it should be, especially for those seeking a return on investment. It’s about conducting business, not being a fanatic. It’s also worth pointing out that SpaceX will launch the OneWeb satellites beginning sometime this year and not one or two years down the road. Again, there’s that unmatched launch cadence flexibility that saves OneWeb money. It’s unlikely any other launch provider can do that. Still, it’s troubling that SpaceX was the only realistic option once Soyuz was taken away.

Those troubles also potentially impact European government space projects that relied on Soyuz. German company OHB, for example, is attempting to sway the European Commission to move Galileo satellite launches to SpaceX from Arianespace. Arianespace received the contract to launch Galileo over four years earlier this year. There may not be any point in the EC entertaining OHB’s request generally since the program is so far behind already. But if it truly wants to deploy more European navigation satellites before the end of the decade, then choosing SpaceX to launch them would make sense. However, the decision OHB asks of the EC is political, which indicates no common sense will come into play for the decision. Other European space programs will be confronting similar decisions.

Murphy’s Law and Monopoly

To be clear, these current scenarios are not SpaceX’s fault. And existing customers with slow-moving projects such as ESA, NASA, and the DoD, will likely remain happy customers of ULA and Arianespace, as both have allocated rockets to government contracts. But, as we see, SpaceX will pick off competitors’ customers desiring to spend money on missions instead of satellite storage for unknown durations. 

But SpaceX’s opportunities don’t address the problem of relying on a single launch provider, as the company certainly cannot launch forever without any issues. Life being what it is, SpaceX will eventually encounter a problem, no matter how well the company is performing. It may be that the problem isn’t one of SpaceX’s making. Another, worse pandemic could take out company personnel. Politicians may decide that SpaceX is for U.S.-based customers and attempt to control it (improbable, but unlikelier things have happened). 

Another concerning aspect of relying on the company is that SpaceX–run by humans–will eventually succumb to the temptation of monopoly. The way it succumbs to that temptation may come in small steps. It may decide to offer its own upper stage rideshare tug while noting that other companies with similar technology aren’t as well built and will diminish the customer’s experience. The consumer products company, Apple, uses that kind of technique quite a bit. It may increase prices as ULA did in its heyday. 

Those are at least two decent reasons to view the current situation for commercial launch capability as a problem, exposing a possibility of a single point of failure. Those who genuinely believe that Vulcan, Ariane 6, New Glenn, etc., will eventually become competitors–that is an unlikely future. True, the companies that field them will view their rockets as competitors but won’t price them accordingly. They may have teething problems, setting their schedules back even more. New Glenn may eventually challenge SpaceX, but I’ve already provided reasons why I believe that’s not the case. 

However competition comes around, it won’t appear anytime soon (certainly not soon enough). Without competition, there is no choice, and SpaceX, a very strong rocket company, somehow becomes a single point of failure.