Ultracheap Satellites and the Multi-trillion SpaceX Future

SpaceX Starlink satellites already cost less than $500,000 each and could reach $250,000 or less. Starlink satellites could beat regular cellphone towers and internet fiber. The SpaceX Starlink satellites have laser communication. The lasers are 40% faster in a vacuum than traveling through fiber.

Cellphone towers cost about $150,000 to build and have a lease cost of about $45,000 per year to landowners. SpaceX has plans filed with the FCC to scale up to over 42,000 Starlink satellites. In 2020, the market for cellphone tower construction is estimated at about $25 billion globally. The total installed base of cellphone towers is about 5 million towers.

In the 2030s, it could be possible for SpaceX Starlink to completely replace most of the ground-based internet and telecommunication infrastructure.

SpaceX could eventually dominate global telecommunications. An estimate of the value of this domination would be over 1 billion customers with $600 per year in revenue. This could be combined with high margins and a large PE multiple for a multi-trillion valuation for SpaceX.

52 thoughts on “Ultracheap Satellites and the Multi-trillion SpaceX Future”

  1. I think Starlink is mainly positioned against long-haul fiber. I don’t see how they compete with cellphone towers, that is a completely different function. The reduced latency only applies if the fiber run is more than 60% of 2x the minimum distance to a Starlink bird, which is orbiting at 280 km. So a ping via fiber from New York to Boston will go about 300 km, via Starlink it will go up 280 km, across 300 km, and then down 280 km, 860 km total. I’m pretty sure fiber will be faster. New York to LA, Starlink is faster of course.

    The real business case in my opinion is not speed or latency, but reduced cost vs fiber, which I think could be huge. I have not seen any details about the earth terminals so I don’t know the throughput. If it’s enough to share a $100 – $300 receiver among many clients then it will be a huge winner. If SpaceX starts attaching 5G nodes to the Starlink boxes then you will see AT&T and Verizon (and Comcast) screaming at congress for help.

    Tinfoil hats on, depending on the total throughput capability, there is a non-zero chance that SpaceX could take over the entire Internet worldwide.

  2. > Oh, I understand that the world’s much bigger than the US

    Apparently you don’t, because you keep driveling on and on about US politics and US emissions.

    Space solar isn’t about US, except for only the fact that SpaceX is a US company. But SpaceX only needs to provide the ride, it doesn’t need to build space solar itself. And it won’t be the only cheap launcher forever.

    Btw, I don’t give a flying F*** about US politics, for the most part, other than their foreign policy. I’m not from US.

  3. Oh, I understand that the world’s much bigger than the US. So big, in fact, that the increase in CO2 output by China annually pretty much negates any amount the US could reasonably (or even unreasonably) drop our emissions by.

    Your optimism if we get a committed D in charge is refreshing. Have you listened to the drivel the Democratic candidates are spouting? Sure, they’re trying hard to persuade their leftist base that each is greener than the other, to a point where they’re practically promising to go back to horse & buggy days. They look on socialism’s record in Venezuela and the economic collapse there as GOOD things.

    Our political ‘elite’ are essentially incompetent. We just didn’t realize how bad it was until we got someone in the job that was half-assed competent…

  4. If the Starlink network ever exists then of course it can also be maintained with the same or lower launch rate.

  5. Yes, that is correct. I was commenting more about the overall number of sats, where Starlink and Kuiper seem similar. Hand phone to sat w/o tower out of my price range!

  6. I thought Project Kuiper FCC application only listed regular sat bands like ka and ku, and not 4G/5G related bands? The only people pushing actual cellular bands for ground-to-sat-and-back seriously right now seems to be only Lynk.Global from appearances…

  7. What I meant to imply that there is currently Iridium for those living on the fringe, and of course it is slow and expensive as well. But Starlink is obviously meant to capture a considerable market share also in population centres that are already served by conventional broadband providers.

  8. The kind of service and latencies Starlink and Iridum provide are really different.

    Kind of comparing dial-up vs broadband access. Superficially both do the same, but not really.

  9. Crescent Dunes and similar concentrated solar facilities have a tendency to draw insects, an attractant for birds. Birds drawn to feed upon the swarms are injured or killed in their passing through the areas of intense light. Where is the outrage over a proven killing box?

    The microwaves are dispersed over a much larger collection receiver, as they inherently are more difficult to focus by a reasonable size of satellite transmitter array. The energy deposited into a living creature exposed to the beam, will result in only a minor heating of the bodily tissues.

    The suspended microwave receiver mesh allows use of the ground underneath, which receives more than ninety percent of the typical sunlight levels for a given location. Farm activities may be carried out without great impediment under this largely transparent canopy.

  10. Nope. All we need is a Sanders or Warren type in office, and you can kiss any or all of that goodbye – and likely SpaceX too. The Dems have gone full retard on energy – they’ll ban fracking, nuclear, and the development of anything with enough energy density to actually power the lifestyle we’ve now got.

    Politicians LISTEN to the screamers. Look at California for an example of that. Every Single Bad Idea that could be implemented was – from ‘divesting’ themselves from evil nuclear and conventional power plants to destroying fresh water storage to forestry practices that were ‘green’ enough to leave the state a flippin’ firetrap waiting for a spark.

    And the ideas are never, EVER wrong. No matter how much destruction they cause.

  11. Finally, rural America will have a way to get on the Internet that is fast and doesn’t suck balls on latency. This is what we paid the big telecoms for (several times over if I remember correctly) but never received. Universal access my a$$.

  12. Janov sez that the closest thing to an Jungian archetype is the need for air at birth, repressed. Because it is so common. And usually so easily prevented, btw.

  13. As importantly as the solution to global heating, SSP is the best cash cow I see for further Space opening, in particular ISRU. This is O’Neill’s plan. Then we can throw in population relief as an added future benefit, amongst all the others yet unimagined.

  14. It’s a non-starter primarily because of cost. Once the costs go down, the environmentalists can scream all they want. Nuclear and coal gets built despite their screaming. Coal mines are commissioned. Forests cut down. As long as the demand is there and the economics and technology and politics allow it, it will get done.

    Of the above, environmentalists can only affect the politics, and only in some places. But it also depends on the PR and public education: the better the technology and engineering are explained, and the more it’s pitched as a “green alternative to evil dirty coal” and a “deliverance from the claws of climate change” etc etc, the less reason for environmentalists to scream about it.

    After the first few are built, reality will speak for itself. If the engineering is good, the fears will be proven unfounded (for the most part).

  15. Space Solar was proposed in the ’80s – the environmentalist screaming about it was deafening. The only way to get the power down (absent a hell of a long extension cord) is by focused microwave transmission. Which wouldn’t (apparently) cause any real problems with birds flying through it or weather problems – but the E-Crowd were screaming that it’d cause boiling rainstorms and fried flocks of birds near the receiver site. Not to mention the possibility of shifting the beam to a city…

    Yeah… it’s a non-starter, unfortunately.

  16. Let’s do it this way.
    Iphone market share in south-east Asia: 4.4% in 2017, let’s say it is the same today;
    Iphone sales in south-east Asia in 2019: 30.7 million;
    First conclusion: smartphone market volume in SEA is 700 million units per year;
    Second conclusion: with notional 2 years replacement cycle (at least for iphones), it is 1.4 billion users of smartphones;
    Third conclusion: Apple and Samsung together have about 40% market share, and those are not cheap models – that is the potential market for space internet, 560 million people in SEA alone.
    Final note: Japan is not SEA, plus EU; Apple sold 178 million units in 2019, all of them expensive – adjust for market share, and one billion becomes plausible.
    IF Apple is talked into including satellite modem into next iphone, it is a done deal – Samsung will follow, and CHINA will follow. The motivation for 5G is less than for satellite, but 5G iphones are coming this year or next. Musk knows how to sell, so he will sell.

  17. Aerodynamic drag will bring these down in a couple of years. This is planned but requires constant replacement.

    I think I remember reading that planned lifetime of Starlink satellites is less than a decade. Then they burn up. Truly relying on such a network will require perpetual access to cheap and abundant launch capacity far beyond anything shown so far. Therefore, it’s unrealistic it will replace everything.

  18. (Western) Europe, Japan, and Australia are less than 500 mil total. In the area literally between them (aka Asia), 600 $/yr is expensive for most of them.

    That’s not the same as saying none of them can afford it, but I’m really having trouble seeing a billion potential customers there at that price level. Maybe a failure of imagination.

  19. Oh that’s a good point!! The stock market would have a field day if trading could become more efficient. I don’t know too much about that, but it seems that perhaps this could make people who use ECN brokers quite happy. It I’m thinking about it correctly.

  20. I am sure the idea is aimed at the main market, There is already “Iridium” offering broadband to those living on the “fringe”.

  21. “I don’t care how good your satellite is, you aren’t beating fiber next to the hub”.
    One of the reasons for it mentioned was that it would beat fibre for automatic trades, the fractional time advantage over fibre is allegedly worth a lot of money.
    @ Jennifer Amb:
    fragile satellites ? maybe, but not many can bring down a satellite. On the other hand, what does it take to cut a fibre line or a wire?

  22. Even if US is excluded entirely, clients inside the area between Europe, Japan and Australia will easily top one billion. Especially if the link is 10Mbit+ per second with unlimited traffic. Many of those would be M2M terminals of all kinds.

  23. Agreed. What if cloud cover, what if storms, to start with. What if someone’s environment. I don’t care how good your satellite is, you aren’t beating fiber next to the hub. It remains to be seen how reliable Starlink will be, though. It could end up being super freakin’ awesome, or it could just suck. But, the fact that it exists at all means that it can contribute to a world in which everything is interconnected.

    Meanwhile, Carl Jung is lounging in the corner, sighing and saying, “You guys are already all connected. Why are you trying to reinvent the wheel?”

  24. That’s actually the old way with geostationary Satellites. The main problem is it adds like 75000 km to the signal travel distance, which even at the speed of light adds 0,3 seconds latancy. In reality the latancy is 1-2 seconds.

  25. They were saying in news the other day that they had a breakthrough that would allow them to service the entire globe with only 4 satellites

  26. Forget trillions, a NASA-sized budget every year will go a long way to make humankind interplanetary.

    With SpaceX prices per pound to LEO, and a few very focused efforts for building habitats and transportation, it could achieve what people were promised since the end of Apollo.

  27. Paying more for a connection which is too slow to support desired upgrades. No alternative service—no upgrade available outside building a fiber run. Will switch to StarLink the moment service becomes available.

  28. Unless newer phones are able to communicate directly to the satellites, then wouldn’t we still need the cell towers?

  29. But for those of us IN the U.S. if I can get a faster connection than Comcast and ONLY spend $600/yr its a steal. I’m also thinking of downsizing to a mobile tiny home – starlink would be perfect – I could still work (writing software) from wherever I wanted to be.

  30. Much like Space Solar, the people without *any* will pay more for *some*, per unit. Where only delivered from Space.

  31. US military is already a customer. USAF was linking starlink with tankers in latest tests. DARPA’s Blackjack comes to mind as well.

  32. I live in the US and spend around 65 for high speed broad band. Most people I know who live in the country can’t get it. They would LOVE to get something better than the satellite choices they have now for internet.

  33. They are not exclusive, in fact they feed each other. I’m for whatever starts lunar or asteroidal ISRU. Launch is a problem, but also cost of the material, even if launch were free. We have no NASA rover on the Moon. Amazing.

  34. Unrealistic to replace the super-important internet infrastructure with something as fragile as a low hanging cloud of satellites that will deteriorate within a few years. Many “what-ifs” in that equation.
    Wire is wire….

  35. Adding Japan, South Korea, Taiwan, and Australia brings the count up to around 800 million.

    Your point that not everyone in these countries is high income is valid, but then, not everyone outside of these countries is low income. There are many who can afford $600 a year, and not all of them have a better deal from their local market.

    On the other hand, even if there are 1 billion potential customers… realistically no one captures the entire potential customer base. Or even 50% of it.

  36. Only 300 million in US, and not all of them with high income. 200 million in western europe, lower income than US. Maybe another 100 mil or so from the rounding and canada. Expensive for most everyone else.

  37. 600$/y per user is widly optimistic. A US PoV of what a broadband connection cost. This is still super expensive for most of the world.

Comments are closed.