The Orbital Space Launch Industry Waiting Game

I've been working in the space industry now for six years, after a bit of a fight to get into it. The industry has changed in that time, and is in an interesting place, that I feel other commentators have not really addressed it. Six years ago the race was on for new orbital space launch vehicles. Mostly small launch vehicles. However that market has contracted with perceived winners like Virgin Orbit, as well as companies like Vector, going out of business. A large part of the reason for the small launch competitors going out of business is that SpaceX rideshares can cost only $5 million dollars for 800 kg to a 500 km orbit, at a price of $6,250/kg (and that's rounding up from $4.8 million to $5 million). Now, there is a lot of advantage to a rocket taking a spacecraft to exactly the orbit it wants to be in, that's worth a lot, but I don't know that it's worth 3-5 times as much for every customer. Specifically because for those extra million or millions of dollars you can take a propulsion system and extra fuel to get where you want to go.

With all of that being said, the space industry seems to be in a waiting game to see what the next generation of launch vehicles in development can deliver. Those include the Blue Origin New Glenn, Rocket Lab Neutron, United Launch Alliance Vulcan, and of course, the most interesting, first fully reusable orbital rocket, SpaceX Starship. To cut to the chase, my personal opinion is that it will cut the cost to orbit for payloads around half to three quarters. There are predictions of Starship cutting the cost by 90% or even more. Falcon 9 can take 22,800 kg to orbit for roughly $60 million at roughly $2,600/kg, and Starship is predicted to take 100,000 to 150,000 kg to orbit and again be fully reusable.  A recent Payload report suggested that current SpaceX costs for a Falcon 9 are under $20 million for each Starlink launch. 

Some differences, a Falcon 9 has a total of 10 main engines, and a Starship has a total of 39 main engines. Engines are the largest subsystem cost in most launch vehicles. So having four times as many engines means four times as many inspections. Plus, the Merlin engine is very simple as far as rockets go, while the Raptor is a more complicated cycle (a full flow staged combustion versus a fuel rich gas generator). This suggested to me, that the engines are going to need a lot of work. That's not necessarily true, but the Raptor has more moving parts and operates at higher pressures which can create a lot of wear and tear. Additionally, Starship has a large number of thermal tiles on the upper stage for reentry, and at least initially, those are going to need a lot of inspection and likely refurbishment between flights, like the Space Shuttle had. 

To estimate that Falcon 9 $20 million dollar launch cost estimate breakdown on paper here are my estimates:

• Launch Fees (insurance, FAA fees, etc.) - $1 million
• Expending upper stage hardware - $5 million
• Inspections between flights - $1 million
• Amortized cost of the reusable first stage hardware - $4 million
• Fuel - $500,000
• Mission Operations cost (people, ground support equipment, communications, ground transportation, launch pad, etc.) - $3.5 million
• Research and Development recouped costs to make it possible - $5 million

Of course these could be wildly off. Specifically the mission operations cost, it's probably a lot less than that, but then again, I don't know. I know that for the airlines those operations costs are substantial, and a lot more goes into a rocket launch than an individual commercial flight from Denver to Chicago.

The reason I don't think that Starship will bring down costs 90% from the existing Falcon 9 costs, is that to do that you have to quintuple the payload to orbit, and half the cost to do so. Using the above numbers, I don't think the launch fees, inspections, fuel, or mission operations costs are going to budge, in fact, in the short term for the first 100 flights them are probably going to go up. In the long term, my estimate is they are about the same, which due to inflation will be less than they are today, but a small percentage lower, not half the price. That leaves the $10 million to come out of difference between expending the second stage, the amortized cost of the first stage hardware, and R&D. Taking $4 million out of the second stage cost, halving the cost of the first stage amortization, and halving the cost of the R&D only takes out $8.5 million, and based on the complexity of reusing an orbital vehicle (which we learned on the Space Shuttle program) that is likely a very optimistic estimate for the second stage costs and R&D. 

One glaring oversight in this analysis is that I looked at SpaceX's estimated costs, and not the current estimated customer costs. Which is to say, instead of charging roughly $60 million for 22,800 kg, SpaceX needs to charge roughly $30 million for 110,000 kg to orbit. While that may be possible, there doesn't seem to be much incentive to actually do that when charging $100 million for such a launch would still be a 2/3rds reduction in cost, and leave presumably a large profit margin and get the cost to orbit under $1,000/kg. That would be well under any of the current competitors, and it remains to be seen where that would put it in relation to the next generation of launch vehicles.

So we seem to be in a waiting game as these test flights and engine tests are conducted. The entire industry is waiting to see where the costs end up in 2026, 2028 and beyond because it radically changes how things are designed and what sort of missions make sense. It would be great to send an orbiter to both Uranus and Neptune but with current launch prices it's still very cost prohibitive to do that. Similarly, mining in space will likely become an industry in the future, in particular mining water on the moon, but it's wildly easier at $260/kg than $1,000/kg to orbit, and at current prices around $3,000/kg, the industry doesn't actually exist. NASA struggled to drill a three foot deep hole on Mars just to measure the temperature.

A risk of this whole article is that I basically just talked about one launch company, since they already have the cost per kilogram lead and look to reduce that cost even further. SpaceX has been landing orbital rockets since December 2015 and still no one else has been able to fully replicate that feat. Looking ahead another nine years, I do expect at least two other companies to replicate that feat, and probably more, but then again I would have expected someone else to do it by now. Which is to say again, we're in a waiting game this year and probably all of next year to see where launch costs shake out and what new space missions become economically feasible.