Is Starship Really Going To Revolutionize Launch Costs?

There’s a lot of talk in the space industry about how launch costs are on the verge of plummeting.

The driving mechanism for this change is said to be SpaceX’s new Starship rocket, a super-heavy lift class vehicle that is intended to be fully reusable.

Starship on the launch pad near Boca Chica. Source: SpaceX

It’s an extremely alluring proposition. Slashing launch costs would theoretically increase innovation in the space industry by allowing commercial entities and research institutions to rapidly deploy space technology in a much more affordable way.

However, I find myself skeptical of this silver bullet solution to the expensive cost of getting to space. In fact, SpaceX has previously speculated on reducing prices as a result of cost savings from reusability, and those price reductions did not materialize. In this week’s newsletter, I’ll discuss the source of my skepticism and how history may repeat itself.

What is Starship?

If you’re already familiar with SpaceX’s Starship vehicle, feel free to skip this section. For those that are new to the rocket or just want a refresher, Starship is SpaceX’s next-generation launch vehicle that, when it launches, will be the most powerful rocket ever flown. Starship intends to lift 100 tons of payload into orbit with every launch, an incredibly impressive feat. Even more extraordinary though is the fact that Starship aims to be the first-ever fully reusable launch vehicle.

Below is an animation created by SpaceX to demonstrate how this might look in the case of the rocket’s first stage landing back on Earth and the second stage traveling to, and landing on, Mars.

Theoretically, What Could This Cost?

Speculation abounds as to what the cost of a Starship launch might be when it is launching regularly. By fully reusing the vehicle and eliminating any need to discard expensive flight hardware, the cost reductions to SpaceX would likely be great. Musk himself has stated that SpaceX’s costs could be as low as $2M per launch, while analysts have suggested that something around $10 million might be more reasonable.

At a $10 million cost per launch, putting 100 tons of payload into orbit would come to about $100/kg. The current cost to SpaceX for a Falcon 9 launch is likely somewhere around $3,400/kg and so Starship’s cost per kg is potentially a full order of magnitude or more improvement.

As a result, it seems understandable why some think that Starship is going to change the game.

But What About The Price?

However, if we want to greatly expand access to space, then cost isn’t necessarily the key indicator that we should be tracking. More importantly, SpaceX will need to pass those cost reductions on to its customers in order for satellite providers to reap the benefits. This is where much of my skepticism resides, because it seems to me that there's little incentive for SpaceX to pass along those savings, or precedent that they will do so.

To understand what might happen when SpaceX is able to reduce launch costs from reusability, I think we should consider what happened the last time a similar dynamic was introduced.

Deja Vu?

On December 15th, 2015, SpaceX accomplished the seemingly impossible. They successfully landed the first stage of a rocket for the first time.

Image of the first Falcon 9 first stage landing in 2015. Source: https://www.space.com/31420-spacex-rocket-landing-success.html

This meant that SpaceX would soon start reusing its Falcon 9 first stages, an innovation that could seemingly slash the cost of getting to orbit. Just three months later, SpaceX president Gwynne Shotwell spoke at the Satellite 2016 conference in Maryland. At the time, she speculated on what Falcon 9 first stage reusability might mean for SpaceX customers, citing that it could potentially reduce prices by 30%, down to $40M per launch. At the time, a Falcon 9 launch cost $61.2 million.

That wasn't the first time SpaceX leadership openly contemplated how reusability would impact pricing. Looking even further back to the Singapore Satellite Industry Forum in 2013, Shotwell speculated that Falcon 9 reusability could mean launch prices to customers as low as $5 to $7M.

Well, it has now been 9 years since the 2013 mention of $5 to $7M prices and 6 years from the 2016 speculation of $40M prices. Since that time SpaceX has widely surpassed the industry's expectations on reusability, flying some of its first stage boosters over 10 times. So what is a Falcon 9 launch priced at today?

$67 million.

The price has actually increased about $5.8M from the pricing in 2016. If we account for inflation, the real price hasn't much increased over that time but has generally stayed the same. The below image is from SpaceX’s current website.

Falcon 9 pricing. Source: https://www.spacex.com/media/Capabilities&Services.pdf

So what happened? Did SpaceX’s costs benefits from reusability not materialize? I doubt that.

Actually, I just think SpaceX hasn’t passed any of those savings on to its customers, they’ve simply expanded their margins and made more money on each launch.

Which I think is entirely understandable. When fully expendable in 2015, the Falcon 9 was already the cheapest launch vehicle in its lift-class at $61.2 million per launch. Today at $67M, that's still the case. In fact, a ride on an Atlas V rocket is twice that price. Not that you could get a ride on an Atlas V anyway, they're all sold out through the vehicle's upcoming retirement.

Furthermore, since 2015 the Falcon 9 vehicle has become arguably the most reliable rocket on the market, with over 115 successful launches in a row.

With no competition capable of putting pricing pressure on SpaceX, why would they reduce their prices? You could make a strong argument they should be raising them. Which they are!

SpaceX is laser-focused on getting humanity to Mars, an enormously expensive ambition. It isn't in their interest to charitably pass savings to customers when they already offer the best product on the market.

Want more evidence of that? Just look at the launch frequency of their rideshare business. SpaceX conducts a few rideshare launches each year where they carry many small satellite payloads in a single Falcon 9 and deploy them all into low Earth orbit. The wait time for a customer to get on one of those missions is close to around two years. SpaceX could fill the manifests for twice as many rideshare missions as they currently perform and reduce wait times significantly. They choose not to because these rideshare missions are simply not high-value launches. It is operationally intensive to work with many customers to pack potentially over a hundred satellites into a single rocket and those customers are extremely price sensitive.

Instead, SpaceX chooses to focus on pricier services like crewed missions to the International Space Station, expensive national security and commercial satellites, or Starlink deployment which SpaceX hopes will bring significant recurring revenue for the company.

So when we look forward to the introduction of Starship, why would SpaceX be incentivized to dedicate many missions to small satellite payloads when it would have to work with hundreds, or even thousands of customers to fill a single Starship with CubeSats?

While Musk states that the company could potentially conduct up to three Starship launches per day, it will take years to get to that launch cadence. In the early days and years of Starship, SpaceX is still going to prioritize deploying Starlink, delivering high-value Human Landing System missions for NASA, and maturing the technology required to put humans on Mars.

Furthermore, what competition is going to pressure SpaceX to bring its pricing anywhere close to its marginal cost of $10M per launch? If Starship can put 100,000kg in low Earth orbit, its nearest competitor by payload class is NASA's Space Launch System (SLS), which was most recently approximated to cost over $2B per launch.

If I were to guess (and to be clear, this is just a guess), I think Starship is likely to be priced early on somewhere around $150M-$250M per launch. At that price, Starship will still be a great deal for customers at only about 1.5x the price of a Falcon Heavy while carrying much more than 2x mass and volume to LEO than a Falcon Heavy. The cost per kg in that price range would be somewhere around $1,500/kg to $2,500/kg to LEO. That would be on a full Starship, but keep in mind that a rideshare Starship for smallsats would likely be carrying much less than its full 100-ton payload capacity and so the price per kg would be higher.

So in the middle of that range, $2,000/kg, we're talking about a 42% reduction in price below Falcon 9 cost per kg of $3,400. That's an incredible benefit to the industry. But it's not the order of magnitude or greater price reduction that I see many hoping for.

Wrapping it Up

Ultimately, I struggle to see the incentive structure that would encourage SpaceX to introduce a revolutionary (which I suppose I'm loosely defining as an order of magnitude or greater) reduction in launch prices for smallsats. If they wanted to further expand access for smaller satellites, they could already do more of that with more frequent rideshare missions. It's just not financially worth it for them.

Just like everyone in this industry, I'm rooting for SpaceX. I would argue, along with many others, that SpaceX is the primary reason we even have a NewSpace industry in the first place. However, I'm also rooting for SpaceX's competition, because I think competition is the only factor that will push SpaceX to bring its launch prices closer to its costs.