SpaceX Will Replace Cargo Planes on Its Path to $1 Trillion per Year in Revenue

The reality of SpaceX mass production rockets is unfolding before our eyes. SpaceX Starships will cost over ten times less than current cargo planes, have over twice the range and will be thirty times faster. These massive advantages will give SpaceX dominance of the cargo business.

The global cargo airline industry generated revenue streams of $117.7 billion U.S. dollars in 2020 and is projected to make $140 billion in 2021.

The commercial rocket launch market is about $10 billion each year.

SpaceX will be able to make ten times more from point to point cargo delivery on earth than from the space launch industry.

SpaceX is rumored to be buying oil rig platforms and will convert them into rocket launch and landing facilities.

SpaceX already dominates the commercial rocket launch business with over 60% market share.

SpaceX will create a new category of rocket air delivery for same-day international deliveries and four to six-hour delivery between continents or from east coast to west coast.

The rocket cargo delivery business will be comparable to the eventual global Starlink internet revenue.

Air cargo will be a training area where SpaceX will master volume operations and improve operational safety. Eventually, SpaceX will achieve the safety needed to move people. They will prove safety by safely flying a few thousand cargo Starships hundreds of times a year.

In a normal year, commercial passenger air travel is a $600 billion a year business.

If SpaceX captures global internet, global cargo and global passenger air travel then they will have a trillion dollars per year in revenue.

Current Rocket Mass Production

SpaceX is building six Starship prototypes and two Super Heavy Boosters at the same time. This clearly shows the mass production of fully reusable orbital-class rockets will be achieved in 2021.

Brendan Lewis (brendan2908) has the status of the SpaceX prototypes.

SpaceX appears on track to building two Starships per week and two super-heavy boosters every month.

Costs Will Be Over Twenty Times Less With More Range, Capacity and Speed

The estimated salary, benefits and overhead for 3000 SpaceX employees is about $600 million per year. This is with an estimated an average annual cost $200,000 for each SpaceX employee. SpaceX will have a runrate of building 100 Starship built each year and 25 Super Heavy Boosters by the end of 2021. Each Starship would have about $3 million in labor and each Super Heavy booster would have about $12 million in labor.

The steel is about $4 per kilogram. The dry mass of the Starship will be about 120 tons and the Super Heavy Booster will be about 300 tons. This would be $0.48 million for the Starship if most of the material was the steel alloy. The Super Heavy would be $1.2 million of steel.

If the steel and salaries are half of the total cost of the rockets then the unit costs at different production levels would be:

Two Starship per week would mean $10.4 million per Starship
25 Super Heavy Boosters per year would mean $36 million SH booster

Reaching a Starship cost of $3.2 million would require 3000 employees to build about 220 Starship every year. This would be $1.3 million in labor, $0.5 million in material and $1.3 million for other costs.

Airbus and Boeing were making 800-900 Planes in normal years. Boeing was making sixty 737s per month and the starting price is $90 million each. Boeing can make ten 787s per month at a cost of $300-400 million each. Airbus has similar costs and production rates.

A Boeing 737-200 in cargo configuration can move 15 tons of cargo and costs $120 million. The range is 1500-3800 nautical miles. It has a speed of about 460 miles per hour.
A Boeing 767-300 in cargo configuration can move 56 tons of cargo and costs $220 million. The range is up to 3,765 nautical miles and the speed is 530 miles per hour.
A Mass Produced SpaceX Starship will have 80-120 tons of cargo capacity and cost $5-20 million each. The range is up to 7000 nautical miles and speed will be up to 20,000 miles per hour.

SOURCES – Brian Wang analysis, SpaceX, Brendan2908, nolinor
Written by Brian Wang,

77 thoughts on “SpaceX Will Replace Cargo Planes on Its Path to $1 Trillion per Year in Revenue”

  1. But how many tons will the Starship be able to transport from one point of the earth to another? And how many could it bring down to earth from outer space, in case of bringing mined minerals from the asteroids?

  2. "There are too many unknowns for any simple evaluation" I would suggest then to use the K.I.S.S principle in your design. An aerodynamically simple design that can evolve over time to either a large sea-based cargo drone that can fly to remote sites/small islands or that can serve as an airborne wind energy platform in an offshore environment.

    "But unless some major player in the industry is willing to back this, it is too big for any small group."
    Who knows? Given all the “Green New Deal” money about to flow from the current US administration you might get some chump change to get the ball rolling on a technical study for a sailplane wind turbine with some starving aerospace and control systems engineers from Upwork. You might get some venture capitalist to take notice of your cargo lifter ideas that way:

  3. There are too many unknowns for any simple evaluation
    I hope some team of professionals will do that. But unless some major player in the industry is willing to back this, it is too big for any small group.

  4. Unlikely to be cost effective, successful. The lowest cost, heavy lift is more likely in the ConcordLift Concept 5 times faster than container ships, direct bypassing ports, No reasons suggested that it would not work or be profitable. see website

  5. It's a good question. Starship should have plenty of thrust for that – but in terms of how much fuel needs to remain to cancel the post-belly-flop velocity, I think the proposition gets a lot more iffy.

  6. "SpaceX is rumored to be buying oil rig platforms and will convert them into rocket launch and landing facilities."

    Bit more than a rumor at this point, at least the buying part.

  7. Allegedly, when the government goes after a monopoly, it's because the monopoly/trust is using predatory practices to lock competitors out of the market.

    Just having a natural monopoly in a niche market isn't enough to warrant government intervention…in theory anyway. Though, in the US's crony CINO (capitalism in name only) system, that isn't necessarily the case.

  8. Establishing SpaceX as a major military asset does give Musk a lot of powerful new friends, but also a lot of powerful new enemies. I reserve judgement at this stage.

  9. For airlines, IIRC, fuel is about 1/3 the total price.

    1/3 fuel
    1/3 capital cost (mostly the cost of the aeroplane itself)
    1/3 personnel costs (pilots/hostesses/ground staff/the people doing the booking/administration)

  10. It seems likely to me that the cost of fuel will be considerably more for the faster delivery. It probably makes better sense to target national security deliveries (at cost plus) rather than air transport.

  11. Sagan's aphorism is a standard for economic decisions, not science. Indeed, if one accepts Popper's criterion for scientific paradigm shift, even a donkey that requires a 2×4 between the eyes to get his attention is more of a scientist.

  12. (i suppose more about Axiom Space having the harder (refined and practicable) schedule with its ISS module already planned — but there are tight networks/partnerships with many of these visionaries/ companies)

  13. I don't disagree, but…
    Weirdest Thing.
    I would consider the usable orbits around Earth to be more like International Waters, if more in practical terms rather than current strict legal definitions. And the thing about that is the huge commercial interest in developing a presence 'up there' may make it nigh impossible to actually have an exclusive and easily manageable 'military space presence'. Just try having a military base over a platinum mine or other Rare-Earth-type resource — yeah, massive commercial interests just seem to get their Way. Secure communications and 'sniper-like' satellites may be able to function in such a noisy commercial/ science zone, but come the 2030s – the traffic will be overwhelming.
    By-the-by: wow-wow-wow — Orbital Space Assembly appears to have closed a massive round of fund-raising for the first few steps of their private space station — ancillary satellites now, but a for-profit module in a few, and schematics for their gravity-research-to-quick-usage plans pre-2030. Poor Space Force generals (or would they be Admirals?), your 'theatre' is about to be overrun by civvies.

  14. I watched one of his tests a couple of years back, supposedly showing how much 'power' was put out by squishing hydrino-containing carbon compounds between two rollers with a lowish-voltage, high amperage current running between them.

    As expected, they flashed rather brightly. This was picked up by a bank of solar cells, which were used to light a bank of LEDs. See? It makes light, which can be converted into electricity via solar cells!

    That as when I realized this really was a scam, the equivalent of a perpetual motion machine. They used carbon arc rods in theater projectors – high-amperage, low voltage DC making a brilliant white arc. If you pop chunks of carbon between two rollers with a hefty amperage between them, what's going to happen?

    You'll get a massive, brief flash.

    Thing about LEDs, they've got a VERY quick 'on-off' time, much faster than an incandescent bulb. They also need much less amperage than an incandescent bulb. And with the persistence of vision of a human eye, a brief flash of the LED is going to seem longer than it actually is.

    Replace the LED panel with a bank of incandescent bulbs, and the things likely wouldn't have even glowed.

    Extraordinary claims require extraordinary evidence. That test I saw wasn't it.

  15. 100 Starships/year at 120 tons is 12,000 tons of annual demand, a million CyberTrucks at 1 ton each is two orders of magnitude more demand. That sort of economy of scale changes costs.

  16. Ocean spaceport platforms will eventually be bespoke built in shipyards like oil platforms but old oil platforms are vastly cheaper to start, less than 1% the cost new. 
    They’ll only be 20-30 miles offshore, reached at first by high speed boats, eventually by VTOL Supersonic electric jets (See Elon in Ironman2) and by Loop tunnels like those The Boring Company is building. Yes they can tunnel under the ocean.

  17. Yep. Noise, Explosion risk, mean it can’t be on land – especially near major population centers where you want to put it – so you can easily transport cargo and later passengers. Luckily many major pop centers all over the planet happen to be next to oceans. It’s not about equatorial launch.

  18. Yep, so about 120k kg x $3 = $360k not $2.4M. The cost of Starship’s bespoke SS Alloy likely kept low by Tesla building the CyberTruck out of it and ordering it in very large quantities from a new generation Texas steel mill supplier – convenient to both Austin GF and Boca Chica. Like so many Musk moves there are plans within plans. It’s a fabulous free promotion to make your bulletproof futuristic ultra tough truck out of exactly the same steel as your Mars Transport Starship – but it also mean you get the steel for the Starship cheaper and more reliably since you’re a much bigger customer.

  19. That sounded like if the methane itself were carbon-neutral (as Musk is planning), then Starship would have a net cooling effect.

    Then I googled and found that while low-altitude contrails are great, high-altitude contrails actually contribute significantly to global warming.

    On the other hand, they mention soot, and methane produces less soot than jet fuel. Starship would also shut its engines down after a few minutes and coast most of the way, instead of producing contrail over the entire distance. And if the aircraft are still using fossil-based jet fuel, net carbon emissions would be negative if Starship's fuel is neutral.

    So while it's uncertain whether Starship would have a cooling effect compared to nothing, it seems likely that it would do less damage than today's cargo aircraft.

  20. Even more important as a Big Benefit that contributes to making him untouchable is the military/strategic importance of Starship and Starlink. That’s still a bit under the radar but if Starship performs anything like Brian projects, it will hand the US and allied militaries their dream Third Offset. It will give them a global strategic dominance technology advantage that changes everything. Russia has nothing even started to keep match and China is many years behind.

    Unlike past military advantages, this one is pretty much created, paid for and controlled by Elon Musk and just handed to them without their involvement. It’s almost like Henry Ford developed nuclear weapons as a side project on his own dime and just offered them to the US military.

    Besides Starship itself, it’s used to create Starlink – another distinct global strategic advantage. Your forces able to get fast digital com anywhere on the planet via a secure planet wide independent network that’s not dependent on ground fiber cables. Just throw that in. He owns that too.

  21. First of all, "hydrino in a bottle" is available to any physical chemists that want to analyze it. Secondly, solid hydrino compounds are are available to any physical chemists that want to analyze them and indeed this has been done independently. Thirdly, do you understand the purported mechanism by which hydrinos form? If not, how can you claim to know even the order of magnitude of the equilibrium partial pressure that would obtain in Earth's atmosphere given that Helium is known to diffuse out of the atmosphere and H2 is driven out faster due to lower molecular weight shared with hydrino gas? Fourth the gas giants are, uh, gaseous and hydrino molecules are gaseous just as are He and H2.

    Laundry list persuasion is effective propaganda and propaganda is effective but does not comprise scientific discourse.

  22. I prefer to view it as my display of actually understanding some physics.

    Look, if hydrinos existed, there would be observable consequences which we don't observe.

    For instance, if you blow a thin walled glass bulb, and evacuate it to a high vacuum, then expose it to air for a long while, it accumulates helium to a measurable extent, because the helium can penetrate the glass.

    Hydrinos, in theory, would do the same thing. But hydrinos are NOT observed to accumulate in vacuum tubes, and believe me, they WOULD be observable. Almost trivially so, we'd have known of their existence for over a century now if they actually were real.

    It would have been trivially easy, by the way, for him to have placed such a bulb by his 'hydrino generator', and collected enough to prove their existence.

    There are other examples. The idea that these "hydrinos" are a lower energy state of hydrogen, and yet don't form at any appreciable rate outside his apparatus is inherently suspicious. The Earth has been around for billions of years, how low would the regular hydrogen to hydrino transition have to be, for a large fraction of Earth's hydrogen to not have ended up as hydrinos?

    Why are gas giants as large as they are? They should be much more compact objects!

    The whole idea implodes if subjected to any degree of skeptical, informed thought.

  23. …SpaceX Starship will have 80-120…

    Each flight will have a max revenue of ~800k per flight, assuming airline parity.
    Are you expecting a premium rate above airline rates? I wouldn't expect a large % of the perishable and high value segment to shed margins just to reduce a day in half.


  24. A 100 knot nuclear-powered hovercraft freighter in the 4000-10000 ton range described in the following study could have carried cargo at one-fifth the projected cost of an air freighter. They were projecting operating costs as low as $0.01/metric ton-km (1973). Maybe with all the interest in small reactors and the Arctic region the idea might just come to fruition:

  25. "There's no telling how long it would take for the water to either be ionized, and leave earth's atmosphere…" I'm tempted to say just zap them away with lasers.

  26. "…use it for cargo delivery at first, so they could test it before sending people aboard." The "Newspace" suborbital approach to cheap space access folks suggested this back in the 1980s and 1990s.

  27. Agreed in regards to Big Search or Big Oil or Big Pharma. SpaceX, on the other hand, is dancing a fine line of Big Space and Big Telecom (satellite internet) and that other not un-related -and too close for comfort- Big -> EV. All, noble enterprises with hard-to-dispute Big Benefits (cheap satellite access, good Optics on USA-based transpo access to existing space assets (ISS), and some good science for all those European Science-y Astro Projects – for SpaceX) and the never-say-die EV revolution that would be nowhere near where it is without the Musk-effect – an almost WandaV-type superpower. At this time, I consider Elon Musk as The Most Untouchable Person on Earth – utterly Senate-proof in a way that Zuck and even Mr. T.Cook is not. This is not everlasting and is not certain. I recommend magnanimous acts by Mr. Musk such as stop pissing off California and the Unions – but keep the dream alive – Musk.
    By-the-by: neato NV resolution to allow Companies (looking at you EM-tesla) their own fiefdoms:

    follow-up reports make it less than as it originally seemed -but dare-to-dream
    (i want my own fiefdom)

  28. I was going to discuss that, but figured I didn't really know enough to make more than vague generalities.

    Though, much of those differences might creep in via the fault tolerance, fatigue resistance, maintainability, and safe handling required to meet aircraft standards.
    Safe handling requirements could end up needing complex flaps and folding landing gear. After all aircraft didn't originally have those, they were added in when they were needed.
    Windows… well there isn't much need for windows in a cargo plane today, is there? I assume they leave them in because it's cheaper to reuse the passenger body than make a special cargo variant, and passengers like windows. If passengers learn to eschew windows from passenger starship experience, then everyone can do without and pocket the resulting savings.
    Seats and toilets. Needed for passengers, not cargo, in both planes and rockets.

    Doors, floors, conveyor belt… once again I'm not seeing why a cargo plane needs them, but a cargo rocket does not.

    Wings. Yep. Can't get away with that one. Wings are a fundamental difference here, and apparently one of the major components (cost wise) in aeroplanes.

    Jet engines cost more than rocket engines. Though this is only true for SpaceX right? The ULA engines were something like $100 million each or so? (I say based only on comments made in this forum…)

  29. Airplanes are much more complicated than rockets. The wings are complicated shapes with internal stiffeners and multiple moving flaps. Landing gear have to fold out and stow away. The engines have many many more blades than a rocket turbopump. Windows, doors, a floor, seats, bathrooms, conveyor belt in the cargo hold, it goes on and on.

  30. The original goal for the Space Shuttle orbiter was 100 flights each. They only reached about 25. That's the current record for reusability of space hardware.

  31. Wiki has some relevant stuff at

    The effect is surprisingly large.

    When no commercial aircraft flew across the USA following the September 11 attacks, the diurnal temperature variation was widened by 1.1 °C (2.0 °F).[21]Measured across 4,000 weather stations in the continental United States, this increase was the largest recorded in 30 years.[21] Without contrails, the local diurnal temperature range was 1 °C (1.8 °F) higher than immediately before.[28]

  32. We may also find that once you build a starship with the levels of fault tolerance, fatigue resistance, maintainability, and safe handling required to meet aircraft standards that they end up at aircraft expense levels.

    After all, I don't think the aircraft designers are stupid. Maybe there are very good reasons that their stuff ends up costing an order of magnitude more than a rocket of similar capacity?

  33. Hydrino NERVA engine with about 10% of its shielding converted to bulk up the engine to give it more reusability yields 30% to 50% dry mass SSTO. At those economics getting out of the atmosphere for cargo transport makes a lot of sense.

    Oh, sorry, I forgot. "Mills is a crackpot".

  34. Also coking.

    Clean methane was partly chosen as fuel because it burns cleanly with very little polymerization, which ends up coating the pipes.

  35. They will be safe unless major governments around the world have recently performed spectacularly badly, having made atrociously poor decisions and management resulting in economic crises and public health disasters.

    In such a case, then pointing to a "big bad monopolistic corporation to blame everything on" would become an attractive option.

  36. I am not a rocket scientist, but I would guess that making your rocket engines tolerant to different fuel qualities would cost a lot more than you'd gain.
    All sorts of little changes, from viscosity changing pumping parameters to burn properties changing soot buildup, would have to be allowed for.
    And then you'd still need to set allowable limits on the exact gas constitution, because raw natural gas contains all sorts of weird stuff from different parts of the world.
    I'm sure it could be done, and maybe when you're running a world wide cargo network it would be a valid choice for version 5 of the engine, optimised for point to point use. But not at this stage.

  37. Boeing 767-300 fuel load = 50.8 t = 16700 gallons for the standard variant
    Taking the US price of jet fuel as US$1.50/gallon that gives $25 000
    = $0.45/kg cargo
    So only 3 times more fuel cost/kg for the starship.

    Remarkable given the much higher speed and that this is comparing the FOAK rocket transport to an aircraft honed over several decades to minimise fuel consumption.

  38. Supposedly, production ships will be made from a proprietary steel very near 304, but with a seasoning of other elements. A little vanadium here, a little manganese there.

  39. At this point, the fuel is methane purified from natural gas. Particularly for sub orbital deliveries, it might be more economic to leave at least some other alkanes in the fuel, and not cool it quite so much. If you could use plain liquified natural gas, there would be a huge savings, and you wouldn't need to have as much fuel sitting around in storage. Gas in pipelines can run up to 90% methane anyway.

  40. Assuming that with time, launches, and landings become common, would safety that much more of a problem than with airliners? Launch, and landing areas would be much smaller than flight paths, and holding patterns that are generally over populated areas.
    A rocket land with tanks near empty, not so much airliners unless they dump fuel in an emergency situation. Look at the crashes of sn8 and 9. The area that would be damaged by the explosion is pretty small compared to the area cleared for launch. Airliner crashes take out buildings in a strip for good fractions of a mile, and about as wide as the wingspan. The methane, and lox sort of flash off, but an airliner is like a Molotov coattail splashing burning petroleum distillate over what ever is on the ground.

  41. That's a huge amount of methane being burned by a fleet of thousands. I'm no climate alarmist, but I do wonder what the effect of injecting so much gaseous H2O into the upper levels of our atmosphere, all the way up to orbit. It would likely increase the greenhouse effect, and increase albedo, particularly if there is ice particle formation. There's no telling how long it would take for the water to either be ionized, and leave earth's atmosphere, or come back into the troposphere where it belongs. There's no telling how the energy balance would net out.

  42. I hope that if they pursue the cargo idea, they would pack the cargo into containers on-shore to make transporting to and loading into the Starship as simple as possible.

    Maybe I'm just too skeptical, but my feeling is that using Starship for point-to-point transport on Earth won't happen, awesome as it would be.

  43. I think that Musk aims for even more reusability. Instead of designing a spaceship just for Moon, Mars missions and its related human, cargo delivery he wants to use it as cargo delivery at Earth as well. So its development expenses would be covered with Earth – Earth cargo launches. And he would use the same thing in more places and make money at home(Earth) as well. He would use it for delivery to Mars as well.

    Well things fill in the larger picture for him. Boring company experience for digging tunnels here and on Mars. Cybertruck for domestic usage and to be used on Mars as well. Cargo deliveries here and on Mars as well,.. And he would make money out of Starship, digging tunnels, Cybertruck here on Earth before actually using them on Mars.

    I have nothing against it, it will get us to Mars much faster.

  44. When using oil rig they would depend less on the bureaucrat and if he has a good or a bad day. Probably less options for competition like Bezos to stop in and say lets mess with the Musk and slow him, by delaying his launch permits.

  45. Personally, I think the chances of any space launcher for achieving airplane levels of reusability and reliability are slim, due to the higher energies and material stresses involved.

    A rocket launcher that can be safely reused a few *hundreds* of times would be an incredible achievement in itself.

    But hey, we are in uncharted territory already, it may actually happen.

  46. Perhaps it is not a bad idea to use it for cargo delivery at first, so they could test it before sending people aboard.

    To reduce delivery times and costs would be good for the economy.

    Interesting why oil rigs. To transport cargo, people to oil rig and then into rocket seems inefficient. Perhaps they want to bypass regulators and their safety rules, so launch-landing in the middle of the sea when there are almost no people around would be less problematic for bureaucrats.

    Or he wants to launch-land rockets near equator(oil rig), if it is worth (the logistics costs) is another thing.

  47. The noise problem is a bit overblown. The landing pads indeed need to be a few miles from the places cargo is meant to go at the coasts, but they still can have the pads well connected through rail and roads.

    But flying these rockets over populated land areas on the final lading stages will take some testing, so that's an actual initial limit to where they can go.

  48. About $150,000 for fuel for Staraship. $900,000 for Super heavy and starship combined. 100 tons of cargo. $1.5 per kilogram of cargo if fully loaded for Starship. $9 per kilogram of cargo going to orbit.

  49. Aircraft have a useful life of around 20,000 flights. So until Starship has a useful life of 2,000 flights, the lower hardware cost per unit is overcome by how many units you need. So far the demonstrated life of the Falcon 9 is about six (8 for the booster, 3 for the fairings, and 1 for the upper stage). The demonstrated life for Starship is zero. So they have a lot of work to do.

    Airplanes crash less than once in a million flights. If a rocket fails significantly more often, you have to factor in insuring the cargo into the price

  50. Sounds like a monstrous anti-trust liability suit/ break-up/ regulation waiting to happen – especially for a national security-type industry, especially in the EU. Too big. Too fast means favourite target. Monopoly behaviour is hard to avoid when all other competitors are so weak and dis-organized.

  51. The steel is about $200 per kilogram. The dry mass of the Starship will be about 120 tons and the Super Heavy Booster will be about 300 tons. This would be $2.4 million for the Starship if most of the material was the steel alloy. The Super Heavy would be $6 million of steel.

    Perhaps you meant 20$?

    Btw, with such costs and production capacity it seems we're a few years away from being an interplanetary species, interesting times ahead!

  52. Airports tend to be near population centers. Landing pads for rockets will have to be further away to deal with the noise and safety issues. Saving hours of flight time to end up on an oil rig is no use.

    But as a weapons platform, the possibilities are endless.


Leave a Comment