SpaceX & the Space Economy
How reusable rockets collapsed the cost of orbit, why Starlink is the real business, and what a trillion-dollar space economy looks like.
SpaceX didn't just build a better rocket, it made getting to space cheap, and cheap changes everything. When the cost of orbit falls 20-fold, businesses that were impossible become inevitable.
What this argues
- Reusability is the whole revolutionRefly rockets cut launch cost by ~95%.
- The business is internet, not rocketsStarlink was 61% of SpaceX's 2025 revenue; its share fell to about 36% in Q3 2026 as the Super Intelligence segment scaled.
- It's a flywheelCheap launch funds Starlink, which funds Starship, which cuts cost further.
- The moat is cadenceLaunching every other day is a lead rivals can't buy.
- Starship is the trillion-dollar betIf it works, orbit gets 20x cheaper again.
What has changed
Updated November 2026
SpaceX went public at about $1.77 trillion
- The IPO priced at $135 per share on 11 June 2026, raising $75 billion, and began trading on Nasdaq as SPCX on 12 June (The Next Web, 2026).
- Market value was reported near $2.0 trillion around 20 October 2026; the earlier $2.3T figure in this piece is no longer supported.
2025 was a net loss year despite $18.7B of revenue
Per the filing as summarised by The Next Web (2026), the connectivity (Starlink) segment earned $11.39B revenue and $4.42B operating profit in 2025, 61% of revenue, but SpaceX reported a $4.94B net loss versus a $791M profit in 2024.
Starlink passed 13 million subscribers, with falling revenue per user
- Subscribers: 10.3M at end of Q1 2026, 12M in June, 13.6M in Q3 (The Next Web; AP, 4 Aug 2026; Quiver Quantitative, 20 Oct 2026).
- Revenue per user fell to $66 a month in Q1 2026 from $86 a year earlier; plans were raised by $5-$10 a month in May 2026 (The Next Web, 2026).
- Connectivity revenue was $5.25B in Q3, up 69%, with $2.37B operating income (Quiver Quantitative, 20 Oct 2026).
A Super Intelligence segment now outsells Starlink
Quiver Quantitative (20 Oct 2026) reports Super Intelligence revenue of $7.56B in Q3 against $5.25B for connectivity, so Starlink was about 36% of Q3 revenue. Net income was $4.4B, adjusted EBITDA $8.8B, and computing capacity 1.7 gigawatts. Q2 revenue was $7.8B with a $541M net loss (AP, 4 Aug 2026).
Starship began flying V3 and made its first revenue mission
- Flight 12 on 22 May 2026 was the first V3 vehicle with Raptor 3 engines (The Next Web, 2026).
- A July 2026 test deployed satellites, and NASA is considering Starship for lunar missions; management targets a Moon landing in 2028 (AP, 4 Aug 2026).
- SpaceX completed its first revenue-generating Starship mission, deploying 26 Starlink V3 satellites (Quiver Quantitative, 20 Oct 2026; date not given).
Spectrum and regulation moved in SpaceX's favour
- FCC approved 7,500 more Gen2 satellites in January 2026.
- FCC approved the purchase of about 65 MHz of EchoStar spectrum in May 2026.
- Amazon Leo's deadline for its 3,232 Gen 1 satellites was extended to 2029 (all The Next Web, 2026).
Not refreshed: 2026 launch count and share of global launches, Falcon 9 totals and reuse records, $/kg figures, the global space economy forecast, and satellite counts.
The Cost Revolution
One number changed everything
For 50 years, reaching orbit cost about $54,000 per kilogram and barely moved. SpaceX cut it by more than 90%.
Every satellite, constellation and space business is downstream of that single price collapse.
The cost of a kilogram to orbit has collapsed
View the data
| Cost per kg | |
|---|---|
| Space Shuttle | 54000 |
| Falcon 9 | 2700 |
| Starship (target) | 100 |
Source: Public estimates; Starship figure is a target, log scale
A different philosophy entirely
| Old space | New space (SpaceX) | |
|---|---|---|
| Rockets | expendable, thrown away each flight | reusable, reflown dozens of times |
| Contracts | cost-plus, government-led | fixed-price, commercial |
| Cadence | a few launches a year | a launch every other day |
| Cost | $54,000/kg, static | $2,700/kg, still falling |
From startup to orbital monopoly in two decades
- Founded: Musk starts SpaceX.
- First landing: a booster returns intact.
- Crew Dragon: first private human spaceflight.
- 165 launches: one every other day.
- IPO: priced at $135 (~$1.77T), $75B raised, largest ever.
How Reuse Works
Why it was so hard, and so valuable
Imagine throwing away a Boeing 747 after a single flight. That is how every rocket worked, until reusability.
The rocket is the expensive part; the fuel is cheap. Reuse the rocket and the economics invert.
How a Falcon 9 flies, and comes home
- Launch: the booster lifts the payload toward orbit.
- Separate: the first stage detaches and flips around.
- Land: it reignites and lands upright, on a ship or pad.
- Refly: days later, refurbished, it launches again.
Reusability is no longer experimental
What rivals still can't replicate
- Landing: guiding a 14-storey booster to a pinpoint landing, every time.
- Refurbishment: turning a flown rocket around in days, not months.
- Cadence: the factories, pads and teams to launch constantly.
- Engines: reliable, reusable, mass-produced engines (Raptor, Merlin).
- Vertical integration: building ~85% of the rocket in-house.
- Iteration: failing fast and improving, a software culture in hardware.
Starlink
The real business
The rockets grab the headlines, but SpaceX's real business is selling internet from a constellation of thousands of satellites.
Cheap launch made a global satellite-internet network economically possible for the first time.
Starlink went from thousands to 13.6 million users in five years
View the data
| Subscribers | |
|---|---|
| 2021 | 10000 |
| 2022 | 1000000 |
| 2023 | 2500000 |
| 2024 | 5000000 |
| 2025 | 9000000 |
| 2026 | 13600000 |
Source: SpaceX disclosures; log scale · values read from the chart, approximate
The largest satellite operator in history
Connectivity where cables can't reach
- Rural & remote: high-speed internet far beyond any cable or tower.
- Maritime & aviation: ships and planes connected mid-ocean, mid-flight.
- Disaster response: instant connectivity when ground networks fail.
- Military: resilient, hard-to-jam comms, a strategic asset.
- Emerging markets: leapfrogging fixed infrastructure entirely.
- Direct-to-cell: ordinary phones connecting straight to satellites.
The Flywheel
Why it compounds
SpaceX's compounding flywheel
- Cheaper launch
- Deploy Starlink
- Subscription cash
- Fund Starship
- Cheaper launch again
Revenue is compounding fast
View the data
| Revenue | |
|---|---|
| 2021 | 2.3 |
| 2022 | 4.6 |
| 2023 | 8.7 |
| 2024 | 13.1 |
| 2025 | 18.7 |
Source: SpaceX S-1
Why the lead widens, not narrows
| What SpaceX has | What rivals lack | |
|---|---|---|
| Scale / cadence | launches every other day | can't match launch volume |
| Cash / funding | Starlink funds R&D | no Starlink cash engine |
| Data / iteration | every flight improves the next | slower, costlier cycles |
| Talent / time | the destination for space engineers | a decade behind on reuse |
The Space Economy
What cheap orbit unlocks
A $630 billion economy on its way to $1.8 trillion
View the data
| Space economy | |
|---|---|
| 2024 | 630 |
| 2028e | 900 |
| 2032e | 1300 |
| 2035e | 1800 |
Source: McKinsey / World Economic Forum (2024)
Where the $1.8 trillion comes from
The space economy is a stack, not a rocket
Layers of the stack, top to bottom
| Layer | What it is |
|---|---|
| 'Reach' applications | GPS-enabled apps, logistics, finance: the biggest layer |
| Space services | connectivity, Earth observation, weather, navigation |
| Satellites | the hardware in orbit: constellations and sensors |
| Launch | getting to orbit: SpaceX's domain, the enabler |
You use it every day without noticing
- Navigation: GPS underpins maps, logistics, banking timestamps.
- Connectivity: satellite internet, TV and phone backhaul.
- Earth observation: weather, agriculture, climate, intelligence.
- Finance: GPS timing synchronises markets and payments.
- Defense: reconnaissance, missile warning, secure comms.
- Science: climate monitoring and planetary research.
The Competition
Who can catch up?
SpaceX launches most of the world's payloads
View the data
| Share of global launches | |
|---|---|
| SpaceX | 63 |
| China | 22 |
| Rocket Lab | 5 |
| Others | 10 |
Source: Global launch records (2025)
Challengers, at very different stages
- Blue Origin: Bezos-funded; New Glenn flying, years behind on cadence.
- Rocket Lab: the credible #2 in the West; smaller vehicles, scaling up.
- China: a fast-moving state program; SpaceX's only volume rival.
- ULA / Arianespace: the incumbents; reliable, expensive, expendable.
- Amazon (Kuiper): a Starlink rival, but must buy launches, often from rivals.
- Startups: dozens chasing niches; most will not survive.
Cost versus cadence: the two axes that matter
A positioning map plots launch cost (lower is better, horizontal axis) against cadence (higher is better, vertical axis). SpaceX sits alone in the top-right as the leader (fast and cheap). China sits in the middle, Rocket Lab below it, and Blue Origin and ULA / Ariane are clustered at the bottom-left among the laggards (slow and costly).
SpaceX's moat versus the threats
| The moat | The threats | |
|---|---|---|
| 1 | Cost: no one is close on $/kg | China: state-backed, scaling fast |
| 2 | Cadence: a launch every other day | Key-person risk: heavily tied to one founder |
| 3 | Starlink: a cash engine rivals lack | Regulation: spectrum, orbits, antitrust |
| 4 | Iteration: a decade of learning | Starship: the bet could still slip |
What to Watch
Signposts, not predictions
Everything now rides on Starship, a fully reusable rocket that, if it works, cuts the cost of orbit another 20-fold.
It is the most ambitious machine ever built, and the single biggest variable in the whole story.
The bull and bear case on SpaceX
| The bull case | The bear case | |
|---|---|---|
| 1 | Monopoly economics: 85% share, still widening | Key-person risk: tied to one founder |
| 2 | Starlink cash: funds everything else | Valuation: priced for perfection |
| 3 | Starship: the next 20x on cost | Starship: could slip years |
| 4 | Talent magnet: the best engineers | Regulation: spectrum, debris, antitrust |
Six risks a serious investor tracks
- Starship slips: the entire cost thesis depends on it.
- Key-person risk: strategy and brand tied to one person.
- Orbital debris: crowded orbits invite collisions and rules.
- Regulation: spectrum rights and antitrust scrutiny.
- China: the one rival with state-scale resources.
- Valuation: little margin for error at ~$2T.
Four dials worth watching
Five signposts for the next 24 months
- Does Starship reach orbit reliably? The whole next chapter depends on it.
- Does Starlink keep compounding? The cash that funds everything else.
- Does cadence keep climbing? The moat widening in real time.
- Can China close the gap? The only volume competitor worth watching.
- Does the IPO change the culture? Public-market pressure meets a founder's mission.
Reference
Glossary and sources
Plain-language glossary
| Term | Meaning |
|---|---|
| LEO | low Earth orbit: where Starlink and most satellites fly. |
| Reusability | reflying rocket hardware instead of discarding it. |
| Constellation | a network of many satellites working together. |
| Starship | SpaceX's fully-reusable next-generation rocket. |
| Payload | the cargo a rocket carries to orbit. |
| Cadence | how often a provider launches. |
| $/kg to orbit | the cost to launch one kilogram: the key metric. |
| Direct-to-cell | satellites connecting ordinary phones without a dish. |
SpaceX's genius was not the rocket: it was realising that if you stop throwing the rocket away, the price of everything above the atmosphere collapses, and a new economy falls out of the sky.
Sources & Further Reading
- SpaceX: Form S-1 and IPO disclosures (2026); Starlink metrics.
- McKinsey & World Economic Forum: 'Space: the $1.8 trillion opportunity' (2024).
- Morgan Stanley: global space economy outlook (2024).
- Public launch records: 2025 global orbital launch summaries.
- Our World in Data: cost of space launches to low Earth orbit.
- Company disclosures: Blue Origin, Rocket Lab, ULA, Amazon Kuiper.
- Figures marked 'target' or 'estimate' are compilations; Starship costs are aspirational.
Sources cited
SpaceX S-1 and IPO disclosures · SpaceX launch records · SpaceX disclosures · Public estimates · Global launch records (2025) · McKinsey / World Economic Forum (2024) · Morgan Stanley · Our World in Data · Company disclosures (Blue Origin, Rocket Lab, ULA, Amazon Kuiper) · The Next Web, 2026 · AP via Fox, 4 Aug 2026 · Quiver Quantitative, 20 Oct 2026 · Pluang / Robinhood market data, Oct 2026
Keep reading
Research, not investment advice. Figures as of , from the sources cited.
Legal information and disclosures
These are my own views as of the update shown on this page. They can change without notice, and I have no duty to update them. Past performance does not indicate future results, and any investment can lose money.
This is for education only. It is not investment, legal, tax, medical or accounting advice, and it is not an offer to sell or a solicitation to buy any security in any jurisdiction. It does not consider your goals or circumstances.
Figures from company filings, public agencies and other third parties are believed to be reliable, but I have not independently verified them and cannot guarantee them. Forecasts and anything marked as an estimate rest on assumptions that may prove wrong.
Please do not copy or republish this material without my written consent.