The launch-cost collapse turned space from a government cost centre into a real market, and venture capital followed it in. The market is proven; the individual company is a coin-flip. Get the vehicle right and you own the theme without betting the house on one rocket.
Part of the Alternative Fortune research library. Category pillar: Venture Capital.
Key takeaways
- The market is real and large. The global space economy hit $613bn in 2024, with the commercial sector driving 78 per cent of growth. Forecasts run from $1trn by 2032 to $1.8trn by 2035.
- The driver is a cost curve, not a story. Cost to orbit fell roughly 20-fold from the Shuttle era to a reused Falcon 9.
- Venture is the dominant investor class. VCs supplied 77 per cent of 2025 space funding, and total infrastructure investment set a record at $22.2bn.
- The dispersion is brutal. Rocket Lab returned roughly 2,138 per cent over two years; Virgin Orbit went from a $3.7bn valuation to zero in 15 months. Same sector, same window.
- Retail access is mostly indirect, and not only American. The purest exposure, SpaceX at a ~$800bn valuation, has no ticker. Most people reach the theme through US ETFs like ARKX, UFO and ROKT, a UCITS fund like VanEck’s JEDI, or the London-listed Seraphim Space Investment Trust (SSIT).
The 60-second version
For most of the last sixty years, space was a government line item. National agencies built the rockets, national budgets paid for them, and a venture capitalist writing a cheque into an orbital-launch company would have sounded like a category error. That world is gone. In 2024 the global space economy reached $613 billion, a record, and it was the commercial sector, not government, that drove 78 per cent of the growth. The thing that once funded flags and prestige is now a market with customers, revenue and, increasingly, profit.
Venture capital does not care about this out of romance. It cares because of a cost curve. The price of putting a kilogram into orbit fell from roughly $54,500/kg in the Space Shuttle era to under $2,700/kg on a reused Falcon 9, a 90 to 95 per cent reduction, and the next-generation vehicle is priced against a number near $10/kg. When the cost of reaching a market collapses by that much, things that were previously uneconomic (satellite broadband, orbital imaging, in-space manufacturing) turn into businesses someone can fund. That is the whole thesis in one sentence. The launch-cost collapse turned space from a cost centre into an addressable market, and venture capital followed the addressable market.
The money confirms it. Space companies drew $26 billion in 2024, up 30 per cent year on year, and infrastructure investment hit a record $22.2 billion in 2025, with venture capital supplying 77 per cent of the year’s funding. And the same period produced a company that ran from a $3.7 billion valuation to bankruptcy in fifteen months. That split, between a market that is proven and a company that is a coin-flip, is the thing to keep in view through everything that follows, because it decides which vehicle actually makes sense for you.
I. What it is
A space-economy startup is a privately funded company whose product or revenue depends on getting to, operating in, or using data from space. That covers a lot of different businesses, and the venture theme makes most sense as a stack of layers rather than a single kind of company.
At the bottom sits launch, the companies that build rockets and sell the ride to orbit. This is the capital-hungry, engineering-heavy base of the pyramid, and it is where most of the famous names live: SpaceX, Rocket Lab, and the smaller launch firms fighting over the small-satellite market. Above launch sits infrastructure, the satellites themselves, the ground stations, the tracking, and the buses (a bus is the standardised body of a satellite that carries the payload) that everything else bolts onto. Above that sits applications, the businesses that sell something to a customer who never thinks about space at all: satellite broadband, Earth-observation imagery, positioning and timing, and the geospatial-intelligence products built on top.
The layers behave differently, which is why the distinction matters to an investor. Launch is binary and brutal, because a single failed flight can end a company. Infrastructure is capital-intensive but more diversified. Applications look, financially, like software and subscriptions once the satellites are up. When Space Capital’s Chad Anderson talks about the sector having “transitioned from skepticism to industrial maturity and proven utility,” with GPS, geospatial intelligence and satellite communications now the “invisible backbone” that powers the world’s largest industries, he is talking mostly about that top applications layer. The launch layer gets the headlines. The applications layer is where the recurring revenue lives.
Venture capital is the asset class that funds high-growth private companies in exchange for equity, expecting most of its bets to fail and a few to return the whole fund. It treats the space stack the way it treats any deep-tech theme. It underwrites the whole distribution and needs the winners to be enormous. Whether space is big is not the constraint. The constraint is whether the specific company you back survives long enough to reach the applications layer where the economics finally work.
II. Market history and growth
The commercial space era has a clean dividing line, and it is a rocket landing rather than a launch.
On a December night in 2015, an orbital-class rocket booster came back from space, oriented itself, relit its engines and set down on a landing pad intact. That first orbital booster landing in December 2015, and the 21-day reflight turnaround eventually achieved, is the moment the economics stopped being theoretical. Reusability had been argued about for decades. Now it was hardware you could watch land. Every venture thesis in this sector traces back to that proof point, because it is what made the launch-cost curve real rather than aspirational.
The market that grew out of it is now genuinely large. The global space economy reached $613 billion in 2024, a record, per Space Foundation’s The Space Report 2025 Q2. The commercial sector drove 78 per cent of the total growth, and government budgets accounted for the remaining 22 per cent. For an industry that was overwhelmingly government-funded within living memory, that split shows how far the transition has come.
Growth forecasts vary by source and methodology, which is worth saying openly. Space Foundation projects the economy will cross $1 trillion by 2032. The World Economic Forum, working with McKinsey, projects $1.8 trillion by 2035, up from roughly $630 billion in 2023. Those are different endpoints on different timelines, and a careful reader should treat both as directional rather than precise. The direction, at least, is not in dispute.
The tempo has changed too. In the first half of 2025 there was a launch to orbit every 28 hours, a record cadence and six hours faster than the 2024 annual record. When a rocket goes up roughly once a day, launch has stopped being an event and become infrastructure, which is a practical signal that the base of the pyramid is maturing.
III. Demand drivers
Behind the market size sit four forces pulling capital into the sector. They are worth separating, because they respond to different things and carry different risks.
The first is the cost collapse itself, the launch-cost curve set out in detail later. It is the demand driver beneath all the others, because cheaper access to orbit makes more applications viable, and more viable applications attract more funding. The other three forces all sit downstream of it.
The second is private investment momentum. Space companies drew $26 billion in 2024, up 30 per cent year on year, per BryceTech’s Start-Up Space 2025. Infrastructure investment specifically reached a record $22.2 billion in 2025, past the previous 2021 peak, per Space Capital’s Chad Anderson. Those two figures are not directly comparable. The BryceTech number counts all space companies, while the Space Capital number counts infrastructure only. They use different scopes and different methodologies, so read them as two lenses on a rising trend rather than a contradiction. What they agree on is that venture is the engine, with VCs supplying 77 per cent of 2025’s funding as the most active investor class in the sector.
The third driver, and increasingly the loudest, is government defence demand. This is where the theme stops behaving like pure risk-on venture and starts behaving like a defence contractor’s order book. The United States’ Golden Dome missile-defence programme secured $25 billion from Congress and, in Bloomberg’s reporting, “drove hundreds of millions in venture funding”, the single largest public catalyst pulling private capital into the sector. It is not the only one. The Space Development Agency’s Tranche 3 award of $3.5 billion for 72 missile-warning and tracking satellites was split across the traditional primes, Lockheed Martin, Northrop Grumman and L3Harris, and, tellingly, Rocket Lab, a startup-era name now winning prime defence work. And the demand recurs: the US Space Force’s FY2027 request includes $6.4 billion for research, development, testing and evaluation on three missile-warning and tracking programmes alone. The same pull is now visible in Europe, where governments are funding sovereign launch and satellite capability to reduce their dependence on American providers.
The fourth is the pull of the applications layer. Once satellites are up, they throw off the kind of recurring, high-margin revenue that venture capital prizes, and Starlink is the clearest worked example. The structural point is that the applications layer is what turns a launch startup’s balance sheet from a cash furnace into a subscription business.
VCs supplied 77 per cent of 2025’s space funding. Venture is the main source of capital driving the sector, not a bystander in it.
IV. The players
Names matter in venture, because in a sector this binary the operator is most of the underwriting. The people and firms below run the market.
SpaceX is the gravity well. Privately held, no ticker, valued at roughly $800 billion by December 2025, up from $350 billion a year earlier. It owns the cost curve, it owns the launch cadence, and through Starlink it owns the sector’s proof that a launch company can bootstrap a downstream subscription business. Founder Elon Musk’s stated ambition is Mars, but the practical business is that SpaceX sets the price the rest of the launch market has to compete against and runs the vertical-integration flywheel everyone else is trying to copy.
Rocket Lab, led by founder and chief executive Sir Peter Beck, is the clearest example of a startup-era launch company that reached the public markets and kept climbing. Founded in New Zealand and now headquartered in Long Beach, California, it trades on the Nasdaq as RKLB. It flies the Electron small-launch vehicle, is developing the larger Neutron rocket, runs a growing space-systems (satellite manufacturing) business, and now wins prime defence contracts. Beck is unusually specific in public about the thing that actually kills companies in this sector, which is budget discipline:
“When we started off this program with Neutron, we said we’re going to spend somewhere between $300 million and $350 million, and we are bang on budget.”
He is equally blunt about the execution risk that comes with it, telling investors “there’s no fat in the schedule, so everything has to go according to plan.” Budget discipline like that is what tends to separate the survivors from the companies that run out of money.
AST SpaceMobile is the high-volatility applications bet, a company building satellite-to-phone connectivity, partnered with more than 60 mobile operators including AT&T, Verizon and T-Mobile. Its financials show how much faith the market extends to pre-revenue space names.
On the capital side, Chad Anderson, founder and chief executive of Space Capital, is one of the sector’s most-cited investors and data trackers. His firm runs institutional venture funds and publishes the quarterly data much of this analysis draws on. His framing, that the market has moved “from skepticism to industrial maturity,” is the bull case put by someone with money on the table.
Europe has its own roster, and it looks nothing like the US startup field. The big listed names are established defence and aerospace groups: Airbus Defence and Space, Germany’s OHB SE (Frankfurt: OHB.F), Italy’s Avio and Leonardo, France’s Thales, and the satellite operators SES and Eutelsat. Launch itself runs through ArianeGroup, the Airbus and Safran joint venture behind the Ariane rockets, and its commercial arm Arianespace, with the European Space Agency as anchor customer and funder. According to fund manager GAM, several of these trade at steep discounts to their US peers: Italy’s Avio at around 2.6x forward EV/sales against Rocket Lab’s far higher multiple. The European route is less a pure-play startup bet and more space exposure bought inside profitable industrial companies.
Behind all of them sit the traditional primes, Lockheed Martin, Northrop Grumman and L3Harris, who still take the largest slices of the defence awards and set the competitive bar the startups have to clear. The venture theme is not the primes. But the primes are the incumbents the theme is trying to displace, and increasingly the acquirers the venture-backed names sell into or compete against.
V. Geography
Space is a global market, but the capital and the contracts are concentrated, and where a company is domiciled shapes what it can win.
The United States is the centre of gravity by a wide margin. US firms took $2.2 billion of Q2 2025 space-tech funding alone, and by mid-2025 the US year-to-date total had already passed the entire full-year 2024 figure. The reason is not only capital depth. It is demand. The US government is the anchor customer for the whole defence-linked layer, through the Space Force, the Space Development Agency and the Golden Dome programme. A launch or satellite startup that can win a US defence contract has a customer whose budget does not flinch when the Nasdaq falls. That is a structural advantage a company outside the US procurement system cannot easily replicate.
China is the other pole, and its role in the theme is mainly as a driver of Western defence demand rather than as an investable venue for most readers. China’s advancing space and missile capabilities are cited directly as the strategic rationale behind the Golden Dome programme, whose estimated cost had reached $185 billion by March 2026. Chinese capability is part of why the American order book is as large as it is. The geopolitical competition is itself a demand driver.
Europe and the United Kingdom are the third pole, and this is where a lot of non-US money actually gets invested. The listed exposure runs through established groups rather than pure-play startups: Airbus, OHB and Avio in continental Europe, the satellite operators SES and Eutelsat, and launch through ArianeGroup and Arianespace with the European Space Agency behind them. Britain is home to the sector’s flagship listed pure-play, the Seraphim Space Investment Trust on the London Stock Exchange, which holds a diversified international portfolio of early-stage space businesses. And European and UK investors can reach the theme through UCITS funds such as VanEck’s Space Innovators ETF, which the big US space ETFs are not built for.
That said, the gravitational pull of US capital and US defence procurement means the largest single concentration of both money and revenue still sits in North America. For an investor reaching the theme through a diversified fund, this shows up directly in the holdings, where the two largest US pure-play space ETFs carry 88.8 per cent and 71.7 per cent US exposure respectively. Buying the theme through one of those funds means mostly buying American space, whether or not that was the intention, which is one more reason a non-US investor looks at the London-listed trust or a UCITS fund first.
VI. How to actually invest
There is no single clean way to buy this theme. Your access depends almost entirely on whether you are an accredited or professional investor, how much illiquidity you can take, and which country you invest from. The vehicles below carry very different numbers to match. Note that most of the pure-play space ETFs are US-listed; if you invest from the UK or Europe, your practical routes are a UCITS fund, the London-listed investment trust, or direct shares in globally listed names.
The most accessible route is a listed exchange-traded fund, which trades like a share and holds a basket of space-exposed companies. The three biggest funds are all US-listed, and they are meaningfully different from one another. For non-US investors there is now a UCITS alternative and a listed investment trust as well.
| Vehicle | Ticker | Type | Expense ratio | AUM | Holdings / concentration | Notes |
|---|---|---|---|---|---|---|
| ARK Space & Defense Innovation ETF | ARKX | US active ETF | 0.75% | ~$1.04bn | 33 holdings; top-10 = 56.96%; 88.8% US | Broadest “space + defence innovation” mandate |
| Procure Space ETF | UFO | US index ETF | 0.75% | ~$938.6m | 68 holdings; top-10 = 47.35%; 71.7% US | Purest pure-play; index rules require ≥50% revenue from space |
| SPDR S&P Kensho Final Frontiers ETF | ROKT | US index ETF | 0.45% | ~$25m | Blends space + deep-sea | Cheapest; but >50% tracks defence budgets, not commercial launch |
| VanEck Space Innovators UCITS ETF | JEDI | UCITS ETF (Ireland) | 0.55% | Fund-level | Global space basket | The main UCITS route for UK/EU investors; launched 2022 |
| Seraphim Space Investment Trust | SSIT | LSE investment trust | Trust-level fee | ~£506m | 20 to 50 early-stage holdings, US/UK/Europe | £-denominated; the flagship listed pure-play, trades at a NAV discount |
| Direct listed equity | e.g. RKLB, ASTS, OHB.F, Avio | Single shares | Brokerage commission | n/a | Single-name risk | Concentrated exposure to one operator; includes European names |
| Venture fund (institutional) | e.g. Space Capital | Private VC fund | Typically 2/20 | Fund-level | Illiquid, multi-year lock-up | Accredited-investor-only; pre-IPO access |
| Pre-IPO secondary | e.g. SpaceX via Forge/EquityZen | Private shares | Platform fees | n/a | High minimums, illiquid | The only retail route to SpaceX |
Laid out side by side, the trade-offs are easy to read. ETFs carry standard brokerage minimums of one share and give you diversified, liquid exposure at a published expense ratio, and the differences between them are real. ROKT is cheapest at 0.45 per cent but more than half of it tracks defence budgets rather than commercial launch, so it is arguably not a clean space bet at all. UFO is the purest pure-play, with index rules demanding at least 50 per cent of a holding’s revenue come from space. ARKX is the broadest and the largest at ~$1.04bn AUM. All three are US-listed, so a UK or European investor buying them takes US-market and currency exposure, and in some pension wrappers cannot hold them at all.
For investors outside the US, two routes solve that. VanEck’s Space Innovators UCITS ETF (JEDI), Irish-domiciled and running at a 0.55 per cent total expense ratio, holds a global space basket and is built to be held inside UK and European accounts. And the London-listed Seraphim Space Investment Trust (SSIT), at roughly £506m, is the sector’s flagship listed pure-play: a £-denominated closed-ended fund holding 20 to 50 early-stage space businesses across the US, UK and Europe, which as an investment trust can trade at a discount to net asset value.
The venture-fund route, through a firm like Space Capital’s institutional funds, is accredited-investor-only, carries six- or seven-figure minimums and typical 2/20 economics (2 per cent annual management fee plus 20 per cent of profits, or “carry”). What you buy for that fee and illiquidity is pre-IPO access to companies you cannot reach any other way.
The purest exposure of all is also the hardest to get, because SpaceX has no ticker. A retail investor can reach it only through pre-IPO secondaries on platforms like Forge or EquityZen, at high minimums and with no liquidity until an exit event, and those platforms are largely restricted to US-accredited or professional investors. Its valuation ran from $350bn in December 2024, a secondary at $185 per share, to roughly $800bn by December 2025, an insider sale at $421 per share. So the company that most defines the theme is the one most investors cannot cleanly own.
VII. Unit economics
The whole theme rests on one number and its trajectory: the cost of putting a kilogram into orbit. It is worth working through slowly, because the maths is what tells you whether the thesis holds. All figures below are dollar-denominated, since that is how the industry prices, but the approximate sterling and euro equivalents follow at the exchange rates prevailing in mid-2026 (roughly $1.35 to the pound and $1.15 to the euro).
Start with the historical baseline. In the Space Shuttle era, the cost to orbit was roughly $54,500 per kilogram. On a reused Falcon 9, that fell to under $2,700 per kilogram, a roughly 20-fold, or 90 to 95 per cent, reduction. The input cost of the entire industry dropped by an order of magnitude within a single career.
Then there is the part that funds the flywheel. The internal cost of a Falcon 9 launch, what it actually costs SpaceX rather than what it charges, is estimated at roughly $300 per pound, against a sell price of about $1,200 per pound. SpaceX’s real cost is therefore around 25 per cent of what the customer pays, and that margin is the engine of everything. The gap between internal cost and sell price is what funds Starlink, funds Starship development, and funds the reinvestment loop that keeps the cost curve falling. A launch company with a 75 per cent gross margin on its own launches can build a satellite constellation with the proceeds, which is exactly what SpaceX has done.
Here is a worked example to make the leverage concrete. Suppose a customer wants to place a 500 kg satellite into orbit.
- Shuttle-era cost basis: at $54,500/kg, that is roughly $27.25 million just for the ride (about £20.2m or €23.7m).
- Reused Falcon 9 cost basis: at under $2,700/kg, the same 500 kg costs under $1.35 million (about £1.0m or €1.17m), a saving of around $25.9 million on a single payload.
- Starship target basis: at the ~$10/kg figure the next-generation vehicle is priced against, that same 500 kg would cost around $5,000 (about £3,700 or €4,350) if fully reusable, a further ~99 per cent reduction on the Falcon 9 number.
That final figure is why the venture thesis exists. A business that was uneconomic at $27 million a launch, and marginal at $1.35 million, becomes trivial at $5,000. Every satellite constellation, every Earth-observation network and every in-space idea that used to be a fantasy sits somewhere on that curve, waiting for the cost to cross its viability threshold. The Starship $10/kg number is not yet proven at scale, and is better treated as the target the industry is pricing against than as an achieved figure, but it is the number the entire startup thesis is underwritten on.
SpaceX’s internal Falcon 9 cost is about 25 per cent of its sell price. That gap is the flywheel. It pays for Starlink and the next rocket.
VIII. Macro sensitivity
Space startups are, financially, long-duration, rate-sensitive, sentiment-driven growth assets. They do best when money is cheap and risk appetite is high, and worst when neither is true. The defence layer is what stops the picture being a simple high-beta growth story, and the table below shows how the theme has behaved, and should be expected to behave, across four macro regimes.
| Macro regime | What happens to the theme | Evidence |
|---|---|---|
| High rates / risk-off | Venture funding contracts hardest; the 2023 trough was the softest year of the cycle, lining up with peak rates | BryceTech; Space Capital |
| Rates easing / recovery | Investment rebounds sharply; funding rose 30 per cent in 2024 as rate expectations softened | BryceTech |
| Risk-on / cheap money | Records set; infrastructure investment hit $22.2bn in 2025 as rate expectations eased and defence demand surged | Space Capital |
| Geopolitical stress | Defence-linked space names turn counter-cyclical; government budgets do not flinch when equities fall | SpaceNews |
The cyclical part is textbook. The 2023 funding trough, the softest year of the cycle, lined up with peak interest rates, exactly as you would expect for a long-duration growth asset. Then investment rebounded 30 per cent in 2024 and hit records in 2025 as rate expectations eased. If that were the whole story, space startups would just be another high-beta growth trade.
The defence layer changes the shape, because geopolitics is now a floor under part of the sector. China’s advancing capabilities and the US Golden Dome programme, $185 billion estimated by March 2026, give the defence-linked names a demand source that runs counter to consumer risk appetite. Government budgets do not contract when the Nasdaq falls, and the same logic now applies to European defence spending, which is rising for its own strategic reasons. For an investor, a diversified space fund blends a rate-sensitive growth component with a defence component that behaves differently, and the mix in your chosen vehicle decides how much of each you own. It is why ROKT, more than half tracking defence budgets, will behave differently from a pure-play like UFO in a risk-off shock.
IX. Tax
This section is general information, not tax advice. Space is a global theme, tax treatment is jurisdiction-specific, and the figures below are US-specific illustrations. Verify your own position with a qualified adviser in your own country before acting.
The tax dimension has one genuinely notable feature, and it applies only to direct equity in early-stage companies held by US-taxable investors.
In the United States, direct equity in a qualifying early-stage space startup can qualify for QSBS, Qualified Small Business Stock treatment under IRC §1202, a regime that can exempt some or all of the capital gain on qualifying startup shares. For stock acquired after 4 July 2025, the expanded regime allows a gain exclusion of the greater of $15 million or 10× basis, on a tiered holding period: 50 per cent exclusion at three years, 75 per cent at four years, and 100 per cent at five years. The qualifying gross-asset threshold was raised to $75 million. At the partial-exclusion tiers, excluded gains face an effective federal rate of 15.9 per cent at the 50 per cent tier and 7.95 per cent at the 75 per cent tier.
One point reshapes the whole access decision for a US-taxable investor, and it is worth saying plainly. Space ETFs held in a brokerage account do not qualify for QSBS, because only direct qualifying startup stock does. For a US investor who qualifies, the most tax-advantaged route is the least accessible one, direct early-stage equity, while the liquid, easy route through an ETF gets none of the QSBS benefit. That tension between accessibility and tax efficiency is real and worth pricing in.
For readers outside the US, none of the specific figures above apply, though the shape of the trade-off often rhymes. A UK investor, for example, can hold a UCITS space ETF or the London-listed trust inside an ISA or SIPP and shelter the gains that way, while direct early-stage equity may qualify for EIS or SEIS relief under separate rules, and other countries run their own startup-equity incentives. The principle that travels everywhere is that the vehicle you hold the theme through, whether direct equity, a fund, an ETF or a pension wrapper, usually matters more for your tax outcome than the theme itself. Get the wrapper right for your own jurisdiction first, with a qualified adviser.
X. Case studies
Case 1: Rocket Lab, the SPAC that worked
Rocket Lab went public via a SPAC, a special-purpose acquisition company (a “blank-cheque” shell that merges with a private firm to take it public), in August 2021 at a $4.1 billion valuation. For a while it did little. Then it executed. The stock traded at $4.54 on 1 July 2024 and closed at $101.65 on 30 June 2026, a run of roughly 2,138 per cent in two years, with an all-time-high close of $150.23 on 27 May 2026.
The operating numbers underneath the share price were real, not sentiment. Rocket Lab posted FY2025 record revenue of $602 million, up 38 per cent year on year, a record Q4 of $180 million, and a backlog of $1.85 billion, up 73 per cent. Then, on 29 June 2026, it announced an $8 billion cash-and-stock acquisition of Iridium, a launch startup buying an established satellite operator. What made the difference was not the SPAC vehicle but the execution. Beck told shareholders that Neutron came in “bang on budget”, and budget discipline like that is what separates Rocket Lab from Virgin Orbit.
Case 2: SpaceX and Starlink, the vertical-integration flywheel
Starlink is the clearest proof of the applications-layer thesis. Its revenue went from $7.7 billion to $11.4 billion across 2024 to 2025, up 48 per cent, and now makes up 61 per cent of SpaceX’s total revenue with $4.4 billion of operating profit. It passed 10 million active customers across 160 countries by February 2026. SpaceX used the 75 per cent margin on its own cheap launches to build a satellite-broadband business that now generates the majority of its revenue and turns a profit. A launch startup bootstrapped a downstream subscription business, using its own rockets as the distribution cost advantage, and that is the flywheel every other name in the theme is trying to build. One caveat on the figures: SpaceX is private and does not file audited accounts, so these are best-available secondary estimates from Sacra, not audited numbers.
Case 3: Virgin Orbit, the cautionary tale
Virgin Orbit is the company that lost everything. It SPAC-merged with NextGen Acquisition Corp in December 2021 at a $3.7 billion valuation, comparable on paper to Rocket Lab’s debut. But the merger raised less than half the expected ~$500 million because shareholders redeemed their stakes rather than staying in, which gutted the runway before the company had proven its economics. After a January 2023 launch failure and no rescue financing, it filed for Chapter 11 bankruptcy on 4 April 2023 and laid off nearly its entire workforce. The assets sold for scraps, with the Long Beach facility going to Rocket Lab for $16 million and the Cosmic Girl carrier aircraft to Stratolaunch for $17 million. Equity holders were wiped out, fifteen months from a $3.7 billion valuation to zero. Virgin Orbit and Rocket Lab shared the same sector, the same SPAC vehicle and the same window, and they reached opposite outcomes. The variable that separated them was runway and execution, not the theme.
Bonus, AST SpaceMobile: the volatility case
AST SpaceMobile captures the live tension of the theme. It posted FY2025 revenue of roughly $70.9 million, up 1,505 per cent off a tiny base, against a net loss of about $341.9 million, and carried a market capitalisation near $24.6 billion in June 2026 on more than $1.2 billion of contracted commitments and 60-plus mobile-operator partners. Its Q1 2026 loss of about $191 million included a write-off from the BlueBird-7 launch failure. Even the names the market values in the tens of billions still run on launch risk and pre-revenue faith.
XI. The core constraint
Put plainly, the constraint on this theme is company survival, not market size.
The market is not the question. The $613 billion market, the 30 per cent funding growth and the launch every 28 hours are not in doubt. What is in doubt, for any single company you might back, is whether it survives the gap between raising capital and reaching the applications layer where the economics finally work. Virgin Orbit did not fail because space was a bad market. It failed because it ran out of runway after a launch failure, having raised less than half what it needed. Rocket Lab succeeded not because the market was better for it, but because it stayed bang on budget and executed.
That is why the single-company version of the theme is so punishing, and why diversification is the central risk-management decision here rather than a mild preference. In a sector where one failed flight can end a company, a fund holding 68 names is a very different proposition from a single-name bet. The single name is closer to a coin-flip on one operator’s execution than to owning the theme.
XII. Inside the asset
To picture what you are actually buying, follow a single satellite from a founder’s slide deck to recurring revenue.
It starts as a business plan that only closes because launch is cheap. At Shuttle-era prices of $54,500/kg the plan does not pencil, and at reused-Falcon-9 prices of under $2,700/kg it does. That single input change is what let the venture money in. The company raises capital, builds a satellite or a constellation of them, and then hits the hardest stretch: the period between spending the capital and generating revenue, when it is a cash furnace carrying a launch-failure tail risk. That is the phase that killed Virgin Orbit, and the phase AST SpaceMobile is visibly living through, burning hundreds of millions in net losses while it builds toward scale.
If it survives, the character of the asset changes completely. Once the satellites are up and customers are paying, the business looks like a subscription company with a moat made of hardware in orbit. Starlink is the finished version of that change, with $11.4 billion of revenue, $4.4 billion of operating profit and 10 million customers. What you are buying, then, is a bet on a company crossing from cash-burning hardware developer to cash-generating applications business. The reward if it crosses is the applications-layer economics. The risk is that most companies never make the crossing.
XIII. The central dilemma
The dilemma that defines the theme has no clean resolution: the purest exposure is the hardest to own, and the easiest exposure is the least pure.
The company that best embodies the thesis, through the cost curve, the flywheel and the applications-layer proof, is SpaceX, valued near $800 billion. It has no ticker, and a retail investor can reach it only through pre-IPO secondaries at high minimums, illiquid until an exit that may be years away, and in practice mostly closed to non-US investors. The easiest exposure, a liquid ETF at one share and 0.45 to 0.75 per cent a year, dilutes the thesis into a basket, and in the case of ROKT fills more than half the fund with defence-budget exposure rather than commercial launch.
So the investor chooses along an axis with purity at one end and access at the other. Direct early-stage equity gets you the purest exposure and, for a US investor, the QSBS tax benefit, but it demands accreditation, capital, illiquidity and the stomach to eat single-company failure. The ETF gets you liquidity, diversification and a low fee, but a blunter, more diluted version of the theme with none of the tax edge. For a UK or European investor, the London-listed trust and the UCITS route sit in the middle: more accessible than pre-IPO equity, purer than a broad defence-heavy ETF, but still a basket rather than a single conviction bet. No vehicle gives you all of it, and an investor who expects one to will be disappointed with whichever route they choose.
XIV. The next frontier
Where the theme goes from here hinges largely on one number being hit at scale: $10/kg.
That is the figure the next-generation fully reusable vehicle is priced against, and it would be roughly a further 99 per cent reduction from the reused Falcon 9 cost. If it is achieved, it would do to the current generation of applications what the Falcon 9 did to the Shuttle-era ones, making a whole class of businesses viable, from in-space manufacturing to larger constellations and orbital services, that are marginal or impossible today. The venture theme’s entire second act is underwritten on that number coming true, and it has not happened yet, so it remains a target rather than an achieved reality.
The nearer-term frontier is more concrete, and it is defence. The recurring government demand is not a one-off. The Space Force’s FY2027 request of $6.4 billion for missile-warning and tracking RDT&E, the $185 billion Golden Dome estimate, rising European defence budgets, and startups like Rocket Lab now winning prime defence contracts together suggest the next few years of the theme will be shaped as much by procurement budgets as by cheap launch. So the frontier runs on two tracks: a cost target that could open new commercial markets, and a defence order book that is already funding the sector whether or not that target is hit.
XV. Lessons from history
The most useful historical parallel is not another space cycle. It is any capital-intensive technology transition where a cost collapse created a real market and a wave of funding, and where the market survived even though most of the individual companies did not.
The clearest lesson from the theme’s own short history is the dispersion. The two SPAC debuts sit close together, with Rocket Lab at a $4.1 billion valuation in August 2021 and Virgin Orbit at a $3.7 billion valuation in December 2021: the same vehicle in the same market at valuations within touching distance, four months apart. One returned ~2,138 per cent and the other went to zero in fifteen months. An investor who bought “the space SPAC theme” indiscriminately in 2021 owned both the multi-bagger and the total loss.
The funding data teaches the same lesson at the macro level. Investment fell into a 2023 trough and then rebounded 30 per cent in 2024 to records, so the aggregate survived and grew. But the companies that raised in the exuberant 2021 window and could not reach the applications layer did not survive the trough. The history rhymes with every other technology boom. The theme was right, the timing of the money mattered enormously, and diversification across the survivors mattered more than picking the single winner.
XVI. The case for it
The bull case for space-economy startups as a venture theme rests on four load-bearing facts, not on optimism.
First, the cost curve is real and demonstrated, not projected. The ~20-fold reduction from Shuttle to reused Falcon 9 already happened, and the 21-day reflight turnaround proved reusability in hardware. This is one of the few deep-tech theses where the enabling breakthrough has already been achieved rather than still awaited.
Second, the market is large and growing across every credible forecast, at $613 billion in 2024, heading for $1 trillion by 2032 or $1.8 trillion by 2035, with the commercial sector driving 78 per cent of the growth.
Third, the applications layer has a proven monetisation model. Starlink is not a projection. It is $11.4 billion of revenue and $4.4 billion of operating profit, proof that a launch company can bootstrap a profitable subscription business. At least one company has already run the flywheel to completion.
Fourth, there is now a government demand floor. The $25 billion Golden Dome, the recurring $6.4 billion Space Force RDT&E request, rising European defence budgets, and startups winning prime defence contracts give part of the sector a counter-cyclical customer whose budget does not depend on risk appetite. When Chad Anderson calls the sector’s core services the “invisible backbone” of the world’s largest industries, he is describing a market that has already been adopted.
XVII. The risks
An investor who reads only the bull case tends to get badly hurt, so the downside is worth setting out in full.
Single-company failure is the base rate, not the exception. Virgin Orbit went from a $3.7 billion valuation to Chapter 11 in fifteen months, wiping out equity holders. That is not an outlier in venture. It is the modal outcome for individual deep-tech startups, and any single-name bet on this theme carries a real, non-trivial probability of total loss.
Launch risk is permanent and physical. Even the winners run on it. AST SpaceMobile’s Q1 2026 loss of ~$191 million included a write-off from the BlueBird-7 launch failure. When a rocket fails, a quarter goes with it, and sometimes the company does too.
The theme is rate-sensitive and sentiment-driven. As long-duration growth assets, these names contracted hardest in the 2023 high-rate trough. A return to a high-rate, risk-off regime would hit the commercial-launch component hard, however good the underlying technology is.
Concentration is hard to avoid. The US pure-play funds carry 72 to 89 per cent US exposure, and even a “diversified” ETF leans heavily on a handful of names, with ARKX’s top ten making up nearly 57 per cent of the fund. The non-US routes spread the geography a little, but they are still baskets weighted to the same big operators. The label promises more diversification than the holdings deliver.
The best exposure is illiquid or inaccessible. SpaceX has no ticker, and the QSBS-eligible route requires direct early-stage equity that most investors, and nearly all non-US ones, cannot access. The $10/kg figure the second act depends on is a target rather than an achieved reality, and if it slips, the timeline for the next wave of applications slips with it.
XVIII. The Alternative Fortune verdict
Space-economy startups are a real venture theme built on a real cost collapse, and the market data backs the thesis at the aggregate level: $613 billion and growing, $22.2 billion of infrastructure investment in a single record year, a proven applications-layer business in Starlink, and a government demand floor most growth themes lack. The bull case is arithmetic on a demonstrated cost curve, not hype.
What that aggregate bull case cannot protect you from is dispersion, which is the feature that should govern how you actually buy the theme. The same two-year window that produced Rocket Lab’s ~2,138 per cent run produced Virgin Orbit’s total wipeout. The theme was right in both cases and the company outcome was opposite, which is why the aggregate can be a strong bet while any single name is close to a coin-flip.
Where the edge actually is. The edge is not in picking the winning rocket. It is in three things the data keeps rewarding. The first is matching the vehicle to your access, your country and your tax position rather than reaching for the purest name, since the QSBS benefit only reaches US direct-equity investors, a UK investor is better served by a UCITS fund or the London-listed trust inside a tax wrapper, and the liquid fund route trades purity for the diversification you actually want in a theme this binary. The second is knowing the mix inside your chosen fund, because a defence-heavy blend like ROKT behaves differently in a risk-off shock than a pure-play like UFO. The third is respecting the cash-furnace phase, since the value is created by the companies that cross from hardware burn to applications revenue, and survival through that crossing is what decides the outcome.
Questions to ask, by vehicle:
If you are looking at a fund or ETF: What is the actual space-revenue purity of the holdings, and is this commercial launch or repackaged defence? What is the top-ten concentration, and how much US exposure am I really taking? Can I even hold it in my account, given US ETFs are often blocked from UK and EU wrappers, where a UCITS fund or the London-listed trust may fit better? (For reference: UFO’s index demands ≥50 per cent space revenue; ROKT is >50 per cent defence budgets.)
If you are looking at a single listed name: Does it have the runway to survive a launch failure, or is it a Virgin Orbit waiting to happen? Is management bang on budget, and does it have a path to the applications layer, or is it still pure cash burn?
If you are looking at a venture fund or pre-IPO secondary: What are the real fees and lock-up (2/20 and multi-year illiquidity are standard), is the vehicle even open to you from where you invest, and, for a US-taxable investor, does the structure keep QSBS eligibility or forfeit it?
The theme is investable, with enormous winners and complete losers inside it. Which one you end up owning depends far more on the vehicle you choose and the survival question you ask than on whether space itself is a good market. And space, on the evidence, is a good market.
For the wider context on this asset class, see the Alternative Fortune Venture Capital category guide.
