The Challenging Economics of Commercial Suborbital Flight
After last week’s post, a reader suggested that suborbital flight would be a good topic. I agree, and so here it is.
First, next week I will be at the Markets & Society conference at George Mason University. Rather than try to cram in a post while I am traveling, I will take next week off, and the next post will be in two weeks.
I find it difficult to find solid information that I can be confident in, which is to be expected about something so speculative, so please forgive all the uncertainty and rough-and-ready calculations. Here I want to look at the motivation for suborbital flight, how it might work economically, and several of the practical problems that the idea faces. Please also forgive that my assessment is rather pessimistic, but it is definitely not hopeless.
Why Suborbital Flight?
Well, first off, what is suborbital flight? According to Wikipedia, a suborbital (or sub-orbital) flight is one that reaches outer space, but on a trajectory that returns to Earth before completing a full orbit. “Outer space” is usually defined by the Kármán Line, which is 100 kilometers above Earth’s surface, so chosen because this is about where the atmosphere is too thin for traditional aircraft to operate. According to Wikipedia, there have been 16 crewed suborbital flights that have reached the Kármán Line, and another 24 that have reached 80 kilometers, a former American definition of space. Most recent suborbital flights have been tourist flights, operated by Blue Origin or Virgin Galactic.
Despite the relatively small number of suborbital flights that have occurred so far, they offer two theoretical advantages over aerodynamically lifted aircraft for commercial passenger flight. First, since most of the trajectory occurs above most of the atmosphere, there is comparatively little atmospheric drag on the craft, saving energy. Second, again because there is relatively little drag, the craft could travel from any point in the world to any other in under two hours. For these reasons, the possibility of using suborbital flight for commercial aviation is of great interest.
Commercial aviation has, according to the industry at least, been a crucial driver of world trade and prosperity. Aviation facilitates trade, generating $3.5 trillion of GDP gain per year as a trade group estimated a few years ago. Increased trade, in turns, decreases the likelihood of warfare. It stands to reason that these benefits will be magnified through a faster means of international travel.
Not everyone is on board with this vision. Some would object to an alleged homogenizing influence that interconnectedness exerts on the modern world. Some would object to a cheapening of cultural heritage that comes with ease of access; even the word “tourism” has come to take a negative connotation of consumerism. These are important subjects, but they would take us too far afield from today’s topic.
Another motivation comes specifically from SpaceX, whose under-development Starship appears to me to be the most likely realization of widespread suborbital flight on the horizon. The motivation in founding SpaceX is to “make life multiplanetary”, which means establishing self-sustaining human settlements on Mars and elsewhere in the solar system and eventually out of the solar system. To do so will require a fleet of starships and the infrastructure to mass produce them. This requires money, which requires a nearer term business model than going straight to Mars. Ideas for Starship (which was known as the Big Falcon Rocket at the time of this presentation) include satellites, supplying the International Space Station, and lunar missions. Point-to-point transportation on Earth is discussed on the final page. Here, SpaceX claims that Starship can travel between any two major cities on Earth in under 40 minutes. The SpaceX Mars vision, and where Starship progress stands now, is a topic that I would very much like to cover at another time.
At present, though, SpaceX’s business model is most associated with satellite launch, including their own Starlink satellites for Internet access (a concept that I was more skeptical about than I should have been in a post that I cannot find right now). Passenger suborbital travel does not appear to be a major focus, and I am not aware of more than sporadic bits of activity from SpaceX since 2017, the date of the aforementioned presentation.
A History of Suborbital Flight
The first serious effort at developing a reusable spaceflight system that I know of was the Boeing X-20 Dyna-Soar project, which ran from 1957 until it was cancelled in 1963. The X-20 was in turn inspired by the Silverbird program in Nazi Germany to develop a hypersonic bomber, a program that never could have realistically gone anywhere. Research from the X-20, in turn, would later be incorporated into the Space Shuttle and the Boeing X-37B. The SpaceLiner, first conceived at the German Aerospace Center in 2005, is a long-term concept for suborbital passenger flight.
In 2004, the Ansari X Prize of $10 million dollars, designed for the first non-governmental organization that could launch a crewed reusable spacecraft twice in two weeks, was awarded for Burt Rutan’s SpaceShipOne, which was then licensed to form Virgin Galactic.
As this article explains, Virgin Galactic has been a leader in space tourism, for which the VSS Unity is the workhorse of their operations. But the Unity has not been profitable, and so this year, Virgin Galactic flew their final Unity flight and is going all-in on Delta, their next-generation vehicle. This is a make-or-break time for Virgin Galactic, and judging from the fact that the company has lost more than 99% of its share value since the peak in 2021, investors seem to be expecting more of the latter.
As of 2021, a seat on Virgin Galactic’s SpaceShipTwo or Blue Origin’s New Shepard, the other leader in space tourism, runs about $250,000 to $500,000, obviously way too much to be a viable form of mass transportation.
Economics of Starship
Now let us consider Starship, and we’ll start with cargo. Last updated in 2022, the Center for Strategic and International Studies estimated the price per kilogram of delivery to low-earth orbit for various spacecraft, with the Falcon Heavy coming in the cheapest at $1500/kg. This is already an enormous achievement; by comparison, the Space Shuttle cost $65,400/kg. Several analyses forecast that Starship could reach a price of $100/kg.
Brian Wang at NextBigFuture forecasts that the price could get as low as $10/kg. He also estimates that the cost of point-to-point delivery on Earth should be about a quarter of the cost of to-LEO delivery, so that would be $2.50/kg. By comparison, international air cargo rates are around $4-8/kg. I suspect that Brian Wang’s estimates are unrealistically low, and so if we use an eventual Starship LEO price of $100/kg, that should translate to $25/kg for cargo on Earth. That’s too expensive for most uses, but I can imagine some niche purposes for which this would be a feasible option. Earlier this year, I worked briefly at a warehouse at the Portland International Airport that processes air freight. A small portion of the cargo looked time-sensitive enough that a customer might have paid a 3-6X premium for the speed advantage of a rocket. For most of the cargo, probably not.
Despite the best efforts of some budget airlines, humans cannot be packed as efficiently as cargo, and so the per-kg costs above do not directly translate into passenger flight. But if we take the same 3-6X cost premium for passenger flight, then a ride on Starship should cost about the same as a business class seat for an intercontinental flight, or maybe a bit more.
A 2017 analysis found that Starship could achieve a $1200 ticket from New York to London. A 2022 analysis put the price at $3000. The latter is based on a $3 million trip cost and 1000 seats on the rocket. Again, some people pay that for business class, and so they might pay it to get a fast trip. People paid more for seats on the Concorde, which was profitable, albeit in a small niche.
There are 36 daily flights from the London area to the New York area, with an average of 241 passengers per flight. The 2022 analysis notes that achieving the $3000 price point requires that Starship reach a cadence of multiple flights per day. Supposing that it can do four flights per day—two from London→NYC and two from NYC→London—then a single rocket could serve about 23% of the travel market. Will 23% of fliers pay business-class prices to save travel time? It’s possible but doesn’t seem likely to me. Maybe in the future, with projected growth in demand for air travel, there will be sufficient demand for a rocket.
This analysis argues that fuel costs, and thus ticket prices, will be significantly higher than what Elon Musk projects.
Another thing that is unclear to me is how much of the above-mentioned prices are contingent on economies of scale. New York to London is one of the busiest intercontinental routes in the world, and so if that route can barely support a Starship, I can’t imagine that the demand will be all that great. Mass production of Starships might have to come from some other source, such as launching Starlink satellites.
Safety and Comfort
I’m sure that we are all wondering how safe and comfortable the ride would be. There are many questions marks there, too.
As of February 2022, 681 people have been to space, and 18 people have lost their lives in spaceflight. As Ronald Reagan movingly spoke about after the Challenger disaster in 1986, we accept some risk when it comes to exploring the frontier. But for routine passenger travel, such a safety record would be unacceptable.
The International Air Transport Association found that 2023 was an exceptionally safe year for commercial aviation, with 30 accidents around the world for an accident rate of about 1/880,000. Of those 30 accidents, one entailed fatalities. Fear of flying is real, but aviation has a reputation as an exceptionally safe mode of travel.
If we suppose that an accident rate of 1/100,000 or less is required for the public and for regulators to feel comfortable with suborbital flight, then at least 100,000 flights are needed to demonstrate this (actually more because statistics, but bear with me). There are about 150-200 launches per year worldwide nowadays, up from recent years. That number could conceivably reach 1000 launches per year with the growth of the commercial satellite industry, but even then, it will take 100 years to accumulate 100,000 launches and show safety. And even then, it is not clear how directly applicable a launch for satellite deployment will be for point-to-point safety.
Obviously it can happen, because it did happen for aviation safety. As this review outlines, getting to where we are today with aviation safety was a long process that entailed many disasters. I fear that a similar process might not be possible today. First, for better or worse, ours is a more risk-averse culture than in the past. Second, passenger aviation benefited from military experience in a way that suborbital flight could not.
A recent paper (I cannot find it; if anyone happens to know where it is, I would appreciate it) shows that passengers would be subjected to forces of 4.5 G, or multiples of Earth’s gravity. I believe the comparable number for a conventional aircraft is 1.2 G, and for ordinary living it is 1.0 G. Such forces would be tolerable, as astronauts do endure them already, but highly uncomfortable and perhaps unacceptable to large segments of the population, such as pregnant women and people with heart conditions. I once found commercial flying to be uncomfortable and scary, and I got used to it, and so people might get used to suborbital flight as well. But this is a nontrivial problem.
Environmental Concerns
I cannot address every environmental issue that is raised in the context of spaceflight, some of which are frivolous, but I do see two serious issues.
While it is advantageous that a rocket can coast for most of its journey, the energy required for launch more than makes up for it. This analysis finds that the energy cost (in dollars, not joules) for a Los Angeles to Singapore Starship flight would be 3.9-7.7 times higher than the energy costs for a Boeing 777X on the same route.
Starship is fueled by methane, which can be produced electrolytically and in a carbon-neutral fashion via the Sabatier process, albeit at a high price. The problem of carbon-neutral rocket fuel is not much different than the problem of carbon-neutral jet fuel, and it requires large volumes of low-cost, clean electricity.
The most serious environmental issue may be ozone depletion. Those readers who are about my age or older remember a time when the ozone layer was at the top of the environmental agenda, and perhaps you thought one that was “solved”. I wrote about that earlier this year.
A few years ago, I put together a summary chart of some studies about ozone depletion from rocket launches. Since then, a few more studies have come out, such as this one, which fall into the range of prior results.
Based on the numbers here, I guess that there are about 15 million international flights per year. One million rocket launches per year would result in ozone depletion that is either unacceptable or at the edge of acceptability according to the figures in my table. So, while there is room for a substantial suborbital flight industry, ozone concerns would preclude it from entirely replacing international aviation unless there is some kind of active ozone restoration.
A wrinkle in this analysis is that the impact of methane fuel in particular is unclear. This analysis finds solid fuels and kerosene would cause significant ozone depletion, while liquid oxygen/liquid nitrogen (what StarLiner would use) and hypergolic fuels would not. The analysis does not mention methane. This 2018 report from the Aerospace Corporation notes the lack of research on ozone depletion of methane rocket fuel, and I cannot tell that the situation has improved since then, though methane fuel could lead to emissions of hydrogen oxides, which have a known ozone depletion potential.
At present, the spaceflight industry is too small to worry greatly about ozone, but as the industry grows, the ozone issue will have to be better understood and addressed.
Siting Spaceports
Unlike a supersonic plane, a rocket cannot travel at low speeds near the start and end of its journey, which means that sonic booms near the spaceport are unavoidable. This will inevitably cause siting difficulties for spaceports that serve multiple flights per day, and they may be forced dozens of miles at sea or to remote areas. To some extent, this defeats the purpose of fast travel, though siting inconveniences might be mitigated by connecting the spaceports to population centers with, I don’t know, a hyperloop.
I foresee a nightmare with NIMBYism. The expansion of Heathrow Airport to include a third runway has, for instance, been snarled by opposition and litigation.
Conclusion
There are things we know that are possible, such as existing subsonic commercial aviation. And there are things we know that are impossible, such as Star Trek-style transporters. Widespread suborbital flight for passenger transportation falls in between.
I’ve discussed many formidable challenges to making suborbital flight a reality. They include getting Starship, or some other system, down to an acceptable price; whether demand for fast travel will justify a cost premium; economies of scale; demonstration of safety; passenger comfort; energy consumption; ozone depletion; and siting of spaceports.
Although they are formidable challenges, and the time to resolve some of them will be measured in generations rather than years, none of these issues are show-stoppers. I think that it is within the realm of possibility that, by the end of the century, we will see suborbital flight serving some high-value routes.
However, to be honest, I would put my money on supersonic flight or vactrains for fast transportation.
Quick Hits
I have been watching with great concern the escalating violence in the Middle East. Recent weeks have brought the pager explosions, air strikes, and an Israeli invasion of Lebanon. Iran has responded with their largest missile barrage of the war, as Hezbollah is understood to be an Iranian proxy. I have not commented on this as much as on other major geopolitical developments of the last few years, mainly because my attention is elsewhere now, but that should not be taken for indifference.
Dock workers in the U.S. went on strike this week, with the strike resolved yesterday. A major issue has been the International Longshoremen’s Association’s opposition to any sort of port automation. At the Cato Institute explains, the opposition is ridiculous and imposes harm on the American public that is much greater than whatever benefits that workers can obtain. Nevertheless, Joe Biden, Kamala Harris, and Donald Trump have all supported the ILA’s position.
California’s SB 1047, which would severely regulate artificial intelligence, met its end with Governor Gavin Newsom’s veto, as Andrew Ng explains in The Batch. Ng has argued before, and I concur, that the bill is based on ill-founded conceptions of the risks associated with AI, including fanciful notions of a robot uprising. The veto was wise, but the legislation should have never gotten as far as the governor’s desk in the first place. Passage in the Senate was nearly unanimous. I hope that this episode is a teachable moment in basing AI regulation on actual risks and not overactive imaginations.
It is being reported that the Jeddah Tower is resuming construction after seven years of being on hold. The tower is set to be the first in the world to exceed 1000 meters in height, shattering the record now held by the Burj Khalifa. I haven’t yet seen visual confirmation of resumed construction, though.
In the Catholic calendar, yesterday was the memorial of Saint Francis of Assisi. Francis is my confirmation saint, and in 1979 he was declared by Pope St. John Paul II as the patron saint of the environment. I am pleased to have visited the town of Assisi in Italy, at a time I was living in a city named (indirectly) for Saint Francis, San Francisco.
And finally, last Tuesday, Jimmy Carter became the first U.S. president to live for a full century. Aside from Washington and Lincoln, Carter is one of my favorites, for reasons that might make a good full post someday, though the Carter administration had its share of shortcomings.