π Moore's Law meet Musk's Law: the underappreciated story of SpaceX and the stunning decline in launch costs
Cheap access to orbit is the key enabler of the New Space Age
Quote of the Issue
βThe greatest technological advances come from combining the resources of a visionary government with the scrappiness of risk-taking entrepreneurs.β - Lori Garver, Escaping Gravity: My Quest to Transform NASA and Launch a New Space Age
The Conservative Futurist: How To Create the Sci-Fi World We Were Promised
βWith groundbreaking ideas and sharp analysis, James Pethokoukis provides a detailed roadmap to a fantastic future filled with incredible progress and prosperity that is both optimistic and realistic.β
The Essay
π Moore's Law meet Musk's Law: the underappreciated story of SpaceX and the stunning decline in launch costs
I canβt stop thinking about this recent headline and deck from the Financial Times:
I mean, the headline is totally accurate β it correctly states the basic facts of the launch β other than missing the substantive point. The issue here is one of misplaced emphasis, a problem with the piece that continues beyond the headline. The FT story goes on to note that the March 14 launch ended with βthe third loss of a Starship in less than 12 months after the previous two test flights ended in explosionsβ and that the FAA tweeted it is overseeing an investigation into the βmishap.β
Contrast the FTβs coverage with that of Ars Technica reporter Stephen Clark. First, the headline and deck:Β
In the piece, Clark hits the ground running with several key points that better capture the significance of the launch than does the FTβs approach. He notes some key milestones:
The most powerful and largest rocket ever built successfully launched and then flew halfway around the world, demonstrating its ability to carry heavyweight payloads into low-Earth orbit.
The successful launch builds on two previous Starship test flights last year and puts the privately funded rocket program on course to begin launching satellites, thus accelerating the pace of Starlink deployments.
Starshipβs Raptor engines have now performed flawlessly on two consecutive flights, demonstrating that the design of the Raptor engine is maturing and becoming more dependable.
During the test flight, SpaceX tested Starship's payload bay door, which will open and shut on future flights to release satellites into orbit, while also transferring super-cold liquid oxygen propellant between two tanks inside the rocket, a precursor test for future in-orbit refueling tests.
Ars Technicaβs Clark and fellow space journalist Eric Berger understand that SpaceX isnβt NASA, and all that reality implies. While not ignoring how Starshipβs Super Heavy booster plummeted into the ocean rather than making a soft splashdown, the rockerβs failure to tick that box wasnβt the newsiest bit of the launch.
This remain unfortunate: Many reporters, as well as the general public, fail to grasp that SpaceX's approach to rocket development differs fundamentally from NASA's. Instead of extensive ground testing and a cautious pace to minimize launch failures β which can lead to longer development timelines and higher costs β SpaceX embraces rapid iteration and learning from failure. Elon Muskβs company views each launch, whether completely successful or not, as an opportunity to gather valuable data and improve future designs. Starship is no different. Its "failures" are stepping stones to success, at least so far.Β
That iterative approach, combined with their vertical integration and in-house manufacturing capabilities, allows SpaceX to move more quickly and cost-effectively than traditional aerospace players β with amazing results. Berger writes eloquently about the launch and how it happened:
The moment of true amazement came about 45 minutes into the flight, as Starship descended an altitude of 100 km and began entering a thicker atmosphere. For a couple of minutes, we were treated to unprecedented views of atmospheric heating acting on a spacecraft. It's one thing to know about the perils of plasma and compression as a spacecraft falls back to Earth at 27,000 km/hour into thickening air. It's another thing to see it.
Let's step back for just a moment to realize how these unprecedented views were possible. Starlink terminals on the ship were sending signals to satellites in low-Earth orbit, which then sent them back to Earth. This is not a new idea. For the last 40 years, NASA has used a small constellation of Tracking and Data Relay Satellites to communicate with spacecraft, beginning with the Space Shuttle. Starship was able to communicate with these satellites upon its reentry, but it was only at a low data rate, and it dropped out as the plasma thickened. The Starlink connection remained longer and is what enabled the stunning video of reentry.
To accomplish this, SpaceX had to build a reusable rocket, the Falcon 9, which is capable of reflying many times. This enabled the company to launch more than 5,500 Starlink satellites and create a global network. (SpaceX operates, by a factor of 10, more satellites than any other company or country in the world). Because of this, it was able to produce unprecedented data and video of Starship's turbulent reentry. β¦ These SpaceX moments feel like a portal opening into the future. That is their power. The first booster landings hinted at the possibility of reusing first stages. The dual booster landing suggested it could be done at scale. Today, we're seeing this promised future as some Falcon rockets fly 20 times, and SpaceX is likely to approach a truly unprecedented 150 launches this year. This high launch cadence enabled Starlink, through which SpaceX has delivered high-speed broadband around the world and in space.
What Thursday's revelatory reentry footage promises is a world in which launch is cheap and abundant. No longer will we need to worry so much about mass or volume, which have been tyrannical overlords to mission planners since the inception of spaceflight nearly seven decades ago.
Bergerβs key phrase: βa world in which launch is cheap and abundant.β Itβs worth spending a moment on the revolutionary decline in launch costs, the key enabler of the emerging new space age. As Citigroup outlined in a 2022 research note, NASA Launch costs dropped significantly from over $100,000/kg in the mid-1960s to around $5,400/kg for the Saturn V used in the Apollo launches starting in 1967. After the Apollo 11 lunar landing, the average launch cost remained relatively stable for decades, averaging about $16,000/kg for medium/heavy payloads and about $30,000/kg for light payloads. This was due to factors such as the use of existing launch systems, reduced number of launches, high reliability requirements for human spaceflight, and a government-funded spending culture. Bottom line: There simply wasnβt much innovation or financial motivation to be innovative.Β
Then came SpaceX, which pioneered lower launch costs with the Falcon 9 in 2010 ($2,500/kg) and Falcon Heavy in 2018 ($1,500/kg) that are 30 times lower than NASA's Space Shuttle in 1981 and 11 times lower than the average launch costs from 1970 to 2010, according to Citi.Β
SpaceX boss Musk has set an ambitious target of achieving a launch cost of $10/kg for Starship, with propellant costs accounting for roughly one-third. This would result in a total launch cost of $1.5 million for delivering 150 tons to orbit, or $1 million if fuel costs are eliminated through advancements in propulsion technology, according to Citi. Achieving this level of launch cost, however, would require significant improvements in materials design, propulsion technology, and operating costs, the bank adds.
So more of an aspirational goal, then. At last for now. Nonetheless, radically cheaper launches will increase access to space and support new business models. With Starship, full reusability is expected to drive launch costs down to roughly $1,600/kg to low Earth orbit, with the potential for further reduction to about $100/kg to $150/kg over time. In Citiβs best-case scenario, launch costs could fall to about $30/kg by 2040, while in a bear-case scenario, costs could be $300/kg if rockets are only reused around 10 times.
In a report earlier this year, the consultancy Bain envisioned Starship reducing the cost per kilogram to low Earth orbit (LEO) by 50 to 80 times:
This marks the commoditization of space launches, and it will put substantial pressure on other active launch providers and those that hope to compete but have not yet successfully launched. It will also allow for more business cases to close for companies that hope to offer services in spaceβeverything from communications and remote sensing satellite companies to commercial space stations, on-orbit manufacturing, and asteroid mining operations.
Musk wants SpaceXβs Starlink satellite business to financially fuel his Mars dreams. How thatβs going? In a recent analysis, JPMorgan notes that satellite internet adoption is on the rise, and traditional wireless communications companies may be underestimating its potential. As bandwidth costs decline by up to 90 percent and capacity increases by some 100 times, according to the bank, satellite internet could become a disruptive technology in many emerging markets. And Starlink, with its growing constellation of satellites and impressive download speeds, has already reached 2 million customers and claims to have achieved cash flow break-even. βResistance is futile,β the analysis says.)
Something wonderful is happening to humanityβs dream of becoming a spacefaring species and spreading across the Solar System. Too much of the media is missing it. But you shouldnβt. The stars our destination.
Micro Reads
Business and Economics
CFOs Tackle Thorny Calculus on Gen AI: Whatβs the Return on Investment? - WSJ
Software automation and teleworkers as complements and substitutes - VoxEU
How Big Tech is winning the AI talent war - FT
How Adobeβs bet on non-exploitative AI is paying off - MIT Tech Review
Will A.I. Take All Our Jobs? This Economist Suggests Maybe Not. - NYT Opinion
Do large firms generate positive productivity spillovers? - LSE
The world needs reminding β governments are not good at picking winners - FT Opinion
Policy
Biden Wants to Put AI on a Leash - WSJ Opinion
Europe Blunders on AI - City Journal
The fallacious case for abolishing the rich - The Economist
US judge temporarily blocks $649 million clean-energy transmission line - Reuters
The little-known AI group that got $660 million - Politico
States are racing ahead of Congress to regulate deepfakes - Understanding AI
AI
Generative AI for designing and validating easily synthesizable and structurally novel antibiotics - Nature
This AI Startup is Trying to Make Fax Machines Work Better for Health Care - Bberg
Health
Bioengineering Is Key to Fighting Extinction - RCS
First pig kidney transplant in a person: what it means for the future - Nature
Clean Energy
AI Will Suck Up 500% More Power in UK in 10 Years, Grid CEO Says - Bberg
Geothermal is the hottest thing in clean energy. Hereβs why - Canary Media
The Advanced Nuclear Industry Would Like to Power Your Data Center - Heatmap
The world is warming faster than scientists expected - FT Opinion
Robotics
Paper planes made by a robot fly better than ones made by humans - NS
How Robots Have Become Trusted Surgical Assistants - Undark
Space/Transportation
Starlinerβs first commander: Donβt expect perfection on crew test flight - Ars
The next generation of drugs could be made in space. Hereβs why - BBC
The future of the Space Force isnβt on Earth β itβs in the solar system - SpaceNews
Up Wing/Down Wing
The rise of bleak chic - FT Opinion
Beware AI euphoria - FT Opinion
Good article and yes, the impact of cost to orbit reductions is under appreciated. There is also little to no comprehension of the engineering methods successfully utilized by Elon.
As long as certain people think that government is immune from impure incentives (as long as the right people are elected) and the private sector is not, and consequently the former deserves the benefit of the doubt and the latter does not, this kind of crap from journalists will continue.