A Skeptical Analysis of the Space Economy Outlook 2026

Key Takeaways

• Novaspace’s $626.4B 2025 figure masks a much smaller $236B core space market.
• Starlink’s $11.4B revenue is 61% of SpaceX, which dominates every forecast assumption.
• NASA’s March 2026 Ignition announcement explicitly confirms LEO commercial markets have not materialized.
• 2020-era forecasts projected a 2025 figure 34 percent above the actual Novaspace number.
• SPAC-era NewSpace issuers lost 70 to 99 percent of peak valuations between 2022 and 2025.

The Trillion-Dollar Space Economy Forecast Problem

The global space economy reached about $626.4 billion in 2025 according to Novaspace’s January 2026 Space Economy Report, up from the firm’s roughly $596 billion estimate for 2024. Novaspace projects that total to reach about $1.01 trillion by 2034, implying roughly 12 percent annual growth. The Space Foundation’s 2025 Q2 report placed the 2024 figure at $613 billion, which points in the same general direction but reflects a different estimate rather than a directly comparable methodology. A 2024 analysis by McKinsey and the World Economic Forum projected a larger $1.8 trillion space economy by 2035. Morgan Stanley has for years argued that the global space industry could grow to more than $1 trillion by 2040, and that long-running forecast has remained part of its space-sector outlook. These headline figures influence how investors frame long-term opportunity, but they also depend heavily on how each source defines the “space economy,” including how much value it assigns to space-enabled services beyond the core space sector.

Novaspace’s January 2026 Space Economy Report separates the 2025 total of $626.4 billion from a narrower “space market” of $236 billion and a broader layer of space-enabled services. That core space market is projected to grow only to about $323 billion by 2034, which implies annual growth of roughly 3.5 percent, far below the headline growth rate for the broader space economy. Narrowing the definition further to companies that directly build, launch, or operate space systems produces a market that is materially smaller than the headline forecasts, although the exact size depends on the definition used by the analyst. Under broader space-enabled services methodologies, the economic value created by a farmer using GPS-guided equipment or a trucking company using GNSS-enabled fleet management can be counted within the wider space economy even though those firms are not usually described as space companies themselves. The line between “uses space-derived data” and “is part of the space economy” is therefore a matter of methodology, and where that line is drawn has a major effect on whether the sector is presented as a relatively narrow industrial market or as a much larger platform economy.

When McKinsey assigns the space economy a 2035 figure of $1.8 trillion, roughly half of the total sits in what McKinsey calls the “reach economy,” meaning applications of space-derived inputs inside sectors such as agriculture, logistics, finance, and consumer devices. Counting the consumer navigation equipment sales, location-based mobile services, and insurance analytics powered by satellite data as “space economy” revenue produces a number that is large and growing, but the growth rate mostly reflects smartphone penetration and cloud software adoption rather than anything that is actually happening in orbit. Excluding some ground segment and consumer equipment categories from Novaspace’s $236 billion figure, and limiting the definition to revenue earned from hardware in space, the 2025 figure likely sits closer to $140 to $170 billion. That is the number a commercial lender or acquisition analyst would use when stress-testing a business plan. It is also the number that corresponds reasonably well to the aggregate revenue of operational space companies whose financials are public, plus a reasonable estimate for the opaque private companies.

The gap between the headline figure and the narrow definition is not merely semantic. It affects how the sector is capitalized. Venture capital firms and sovereign wealth funds allocating to “space” use the headline figures to justify position sizes. Public-market equity analysts use them to justify forward multiples. Policy makers use them to justify spending programs. Each of those allocation decisions flows from a forecast number that is, on reasonable scrutiny, substantially inflated by definitional scope. A disciplined allocator should cut the forecast numbers in half before applying them to a specific investment thesis.

Historical Comparables: Space Is Not the First Sector to Misfire

The pattern visible in the space economy forecasts is not unique. Several other technology-adjacent sectors have gone through comparable forecast cycles over the last 15 years, and reviewing them briefly provides useful discipline for interpreting the current space economy numbers.

The cleantech boom of 2006 to 2012 produced investment bank forecasts that projected the sector would reach $1 trillion to $2 trillion in annual revenue by 2020. Solyndra’s 2011 bankruptcy, A123 Systems’ 2012 bankruptcy, and a broader pattern of venture capital losses cut the cleantech thesis substantially, although the sector eventually did reach scale in a different configuration (utility-scale solar, EV batteries, grid storage) than the original forecasts described. Kleiner Perkins, which had been the most visible cleantech investor, wrote down much of its cleantech portfolio between 2012 and 2015. The firm’s subsequent repositioning away from cleantech was a clear signal that the sector had underperformed initial expectations. The parallel to space is that the sector did become large, but not in the way its early investors assumed, and the companies that got rich were largely not the ones that early investors backed. The solar industry that reached maturity in the 2020s is dominated by Chinese manufacturers whose cost structures the 2006 to 2012 US investors did not anticipate, and the battery industry is dominated by a handful of East Asian firms and Tesla, none of which were central to the original US cleantech thesis.

The blockchain sector between 2017 and 2018 produced forecasts ranging from hundreds of billions to multiple trillions in annual transaction volume by 2025. Goldman Sachs, Morgan Stanley, and multiple management consulting firms published reports predicting that blockchain would become foundational infrastructure across banking, supply chain, and government. The realized outcome was a small but durable cryptocurrency trading market, some meaningful stablecoin adoption, and a long tail of failed enterprise blockchain projects. IBM’s Hyperledger and R3’s Corda, two of the most visible enterprise blockchain platforms, retrenched substantially by 2024. The sector exists, but its shape is different from the forecasts and substantially smaller than the most aggressive projections. The total addressable market for blockchain infrastructure as forecast in 2018 has not been reached, and the forecasts have been quietly updated rather than explicitly retracted.

The metaverse cycle of 2021 to 2022 produced $1 trillion addressable market projections from multiple investment banks. Meta alone invested tens of billions in the concept, renaming itself from Facebook in October 2021 to signal the strategic priority. By 2024 the forecasts had been quietly retracted or repositioned as general XR technology projections, with the “metaverse” framing fading from mainstream corporate strategy materials. Meta’s Reality Labs segment has continued to generate substantial losses against modest revenue. The sector will eventually produce real products and meaningful revenue, but the trillion-dollar framing of 2021 appears, with three years of hindsight, to have been an investment banking narrative rather than a defensible projection. The parallel to space is the forecast plasticity discussed earlier: when specific categories fail, the total can be preserved by redefining what the sector includes.

Commercial fusion energy has run for decades with forecasts of imminent commercialization that consistently slipped. The sector now has multiple well-funded companies (Commonwealth Fusion Systems, Helion, TAE Technologies, Tokamak Energy) targeting operational demonstrators in the 2028 to 2032 timeframe, with a combined private capital raise exceeding $7 billion. The forecasts for fusion commercialization have been repeatedly updated to reflect delays. Each generation of optimistic timeline has slipped by approximately the time between the forecast and the realized milestone. Space economy forecasts share the fusion pattern of persistent optimism about a future transition that has not yet occurred, with the transition timeline receding as the forecast horizon approaches.

The autonomous vehicle sector between 2015 and 2019 produced forecasts that predicted operational Level 5 autonomy by 2020, fully commercialized robotaxi services by 2022, and a $1 trillion to $7 trillion addressable market by 2030. The realized outcome through 2025 is a small number of commercial robotaxi services (Waymo, Cruise, Baidu Apollo Go, Pony AI) operating in limited geographies, extensive investor write-downs (General Motors exited Cruise in 2024), and a broader sense that the fully autonomous vehicle future is five to ten years further out than the 2018 projections implied. Waymo’s public route through 2025 is a reasonable business, but it is not the trillion-dollar platform the forecasts described. The parallel to the space economy is that specific technical and regulatory obstacles, which the forecasts treated as incremental engineering challenges, have turned out to be multi-year capital-absorbing difficulties.

The broader lesson from these adjacent comparables is that technology-heavy sectors with strong capital-raising narratives tend to produce forecasts that overshoot realized outcomes by 30 to 70 percent on total market size, and the compositional miss tends to be larger than the total miss. Specific product categories that anchored the original forecasts tend to fail at higher rates than the sector aggregate, and the categories that eventually reach scale tend to be those that were not featured in the original projections. Applying that pattern to the space economy is not defeatism. It is calibration. The sector is real, growing, and economically meaningful. It is also being described in financial materials at a scale that the operating reality has not yet justified. A skeptical reader should internalize the comparable pattern and discount accordingly.

The Five-Year Retrospective: Forecasts vs. Realized Data

The forecasts circulated in 2019 and 2020 are due for a reality check now that 2025 market estimates are available. Bank of America said in 2020 that the space industry could reach about $1.4 trillion by 2030. On a simple linear interpolation from a commonly cited 2020 base of about $423 billion to that 2030 target, the implied 2025 waypoint would be about $912 billion, well above Novaspace’s 2025 estimate of $626.4 billion. Morgan Stanley’s long-running forecast that the global space industry could exceed $1 trillion by 2040 started from a smaller base and, on the same linear basis, implies a 2025 figure that is closer to currently reported estimates. By that rough linear yardstick, the 2020 Bank of America projection appears materially ahead of realized market estimates by 2025, while Citi’s 2022 forecast of more than $1 trillion by 2040 was more conservative than some peer forecasts and is directionally closer to current data.

The compounding matters. On a simple extrapolation from Bank of America’s 2020 view, the 2025 market estimates now available imply that the sector is running materially below the path that would have been needed to reach $1.4 trillion by 2030. That does not mean a 2030 total near $900 billion is the only plausible outcome, because real markets do not grow linearly, but it does suggest that some of the most aggressive early-decade forecasts now require either a sharp acceleration or a broader definition of what counts as part of the space economy. A skeptical reading of the current data supports a 2030 outcome below the most bullish headline targets, and it also suggests caution when extending the McKinsey and World Economic Forum $1.8 trillion by 2035 scenario without adjusting for the gap between earlier projections and realized mid-decade estimates.

The composition miss is at least as striking as the total miss. Early-2020s space-economy narratives often assumed that space tourism, in-space manufacturing, space-based solar power, and even asteroid-mining concepts would begin making visible commercial contributions by the middle of the decade. By 2025, Virgin Galactic reported only $2 million in annual revenue, and Blue Origin paused its New Shepard tourism program, which leaves commercial space tourism far below many earlier expectations. In-space manufacturing has made technical progress, with Varda raising capital and flying missions, but public reporting still does not show a large, recurring commercial revenue base. Space-based solar power remains in the study and demonstration stage rather than a revenue-generating commercial market, and asteroid mining has not produced commercial revenue, with Planetary Resources and Deep Space Industries both having exited their original asteroid-mining paths before the forecast horizon. Much of the realized upside instead appears to have come from satellite broadband, with SpaceX generating roughly $15 billion to $16 billion in revenue in 2025, underscoring how concentrated the sector’s commercial momentum has become.

The pattern visible across the five-year retrospective is that forecast flexibility has masked composition failure. A forecast can still look directionally defensible at the headline level even when several named growth categories remain far smaller than expected and most of the commercial outperformance comes from one segment. If the 2025 data now being reported were handed to an analyst who had been relying on the most aggressive 2020 assumptions, the reasonable conclusion would be that the sector underperformed those expectations on both total size and category mix. The practical implication is not that current forecasts should be dismissed, but that they should be discounted and stress-tested. On that basis, estimates closer to roughly $850 billion to $1 trillion in 2030 and about $1.1 trillion to $1.3 trillion in 2035 are better understood as analytical caution ranges than as firm forecasts, and they point to a market that is still substantial but less dramatic than the headline storylines that often drive capital-allocation decisions.

A secondary observation is that recent forecast updates have generally emphasized new outlooks more than explicit reconciliation with older ones. Morgan Stanley continues to frame the space economy around a path to more than $1 trillion by 2040. Bank of America’s 2020 $1.4 trillion by 2030 view has not remained as prominent in more recent public discussion, which has tended to focus more on specific sub-sectors and listed companies. McKinsey’s 2024 analysis effectively resets the debate with a larger and later estimate built on a broader definition that includes both backbone and reach applications. That pattern is common in investment and consulting research, but it does not remove the need for readers to compare older projections with realized data and decide for themselves how much confidence current long-range forecasts deserve.

Starlink Accounts for Most of the Revenue Base

Internal financial data disclosed through reporting by The Information in April 2026 put Starlink’s 2025 revenue at $11.4 billion, with an EBITDA of roughly $7.2 billion and an adjusted profit margin near 63 percent. Starlink at that scale represents 61 percent of SpaceX’s $16 billion in 2025 total revenue and almost the entirety of its EBITDA. The launch business remains profitable at a smaller level, and xAI, the artificial intelligence company acquired by SpaceX in February 2026, burned $9.5 billion in the first three quarters of 2025 against just $210 million in revenue.

That concentration is the single largest risk factor in the sector that capital allocators have not priced. Morgan Stanley analyst Adam Jonas has described Starlink as a standalone business that alone would justify a $500 billion valuation. The SpaceX IPO filing on April 1, 2026, targeting a valuation between $1.75 trillion and $2 trillion, reflects a price-to-sales multiple of roughly 109 times 2025 revenue. For context, Amazon’s cloud business AWS trades at a multiple closer to 15 times revenue. Even Nvidia, the AI-era benchmark for scarcity premiums, commands roughly 30 times forward revenue. A price-to-sales multiple of 109 implies that the market expects SpaceX revenue to grow substantially, that margins will remain at or above current levels, and that the business will remain largely uncontested by competitors for many years. Each of those assumptions is defensible on its own and extraordinary in combination.

The competitive landscape Starlink faces is thin by any standard. Eutelsat OneWeb has a far smaller constellation and a subscriber base that has never been disclosed in terms comparable to Starlink’s. Eutelsat’s financial reports describe OneWeb as a strategic capability rather than a standalone profit center, and the company’s recent capital raises have depended on French government support. Amazon Leo, rebranded from Project Kuiper in November 2025, had approximately 241 satellites in orbit as of mid-April 2026. The FCC required Amazon to have 1,618 satellites operational by July 30, 2026, and Amazon requested an extension to 2028 stating that it would reach only about 700 satellites by the original deadline. Amazon has not publicly disclosed subscriber targets or revenue forecasts for Leo, reflecting an internal view that the program is a long-term strategic commitment rather than a near-term revenue opportunity.

Starlink’s revenue ramp, while accelerating, also reflects aggressive pricing rather than pure demand pull. Terminal subsidies have reached the point where SpaceX ships some units at or below cost, particularly for residential customers in emerging markets. Starlink’s residential customers sit at roughly $2,000 per year in average revenue per user, well below the enterprise tiers in aviation ($300,000 ARPU) and maritime ($34,000 ARPU). The residential customer base is large but the ARPU is modest, and Starlink’s dominance of its segment has not yet produced pricing power that would permit substantial ARPU expansion. The pricing power is constrained by the availability of terrestrial broadband alternatives in developed markets and by purchasing power limits in emerging markets. Neither constraint loosens in the forecast horizon.

Direct-to-cell is the service category most exposed to optimistic assumptions. Starlink launched the service commercially through T-Mobile as T-Satellite on July 23, 2025, allowing standard smartphones to connect to Starlink satellites. As of April 2026, SpaceX had more than 650 direct-to-cell enabled satellites in orbit, against an initial target of approximately 840 satellites for full global coverage. The service is live in 22 countries with more than 400 million people having compatible device access. The service currently supports text messaging with voice and data capabilities expected to come online during 2026. The technical feasibility has been demonstrated. The commercial economics remain unproven. T-Mobile’s deal with Starlink includes undisclosed revenue-sharing arrangements, and the mobile carrier retains customer billing relationships. Starlink’s share of direct-to-cell revenue per subscriber is not visible in public disclosures.

The $17 billion acquisition of spectrum licenses from EchoStar in 2025 gave Starlink the regulatory foundation to operate direct-to-cell at scale, but it did not guarantee that mobile carriers would treat the service as a partner rather than a competitor. AT&T, Verizon, and European mobile network operators have been slower to enter direct-to-cell partnerships, suggesting that the perceived competitive threat is real. The structural risk for Starlink is that mobile network operators will develop their own direct-to-cell partnerships with competing constellation operators (AST SpaceMobile has partnerships with AT&T and Verizon, Lynk Global has partnerships in several countries) and that Starlink’s direct-to-cell revenue will remain concentrated in a single partnership with T-Mobile in the US and a handful of smaller international partners.

The addition of xAI to the SpaceX corporate structure complicates the financial picture substantially. Musk has described an “orbital intelligence” strategy that would place AI inference capability on Starlink satellites, with the stated goal of providing low-latency AI services that cannot be matched by terrestrial data centers. SpaceX has filed with the FCC to license an orbital data-center constellation of up to 1 million satellites under the strategic rationale that orbital placement of AI inference hardware provides specific advantages including thermal dissipation, solar power, and reduced terrestrial grid dependency. A reasonable observer should treat the orbital intelligence narrative as speculative, and should note that xAI burned $9.5 billion in the first three quarters of 2025 while generating $210 million in revenue. The merger arithmetic implies that Starlink’s cash generation is now partially funding xAI’s operating losses, which converts the SpaceX IPO valuation from a pure space investment into a blended space-AI investment whose AI component has substantially worse unit economics than the space component.

The SpaceX Transfer Pricing Advantage

A largely unexamined feature of the SpaceX business is the asymmetric economic position Starlink occupies relative to every other broadband constellation operator. Falcon 9’s public list price increased to $74 million per dedicated launch in February 2026, and rideshare rates rose to $7,000 per kilogram. Independent analyses estimate SpaceX’s marginal cost per Falcon 9 mission at approximately $15 million including reused booster amortization, a new upper stage, new fairings, propellant, and range fees. The gap between the list price Amazon Leo and Eutelsat OneWeb pay for comparable capacity and the internal cost SpaceX absorbs for its own satellites is approximately $55 million to $60 million per launch.

That spread compounds across the deployment manifest. Starlink flew approximately 165 Falcon 9 missions in 2025, the majority of which carried Starlink payloads. At commercial list prices, those missions would have cost on the order of $12 billion. At internal cost, SpaceX spent approximately $2.5 billion. The difference is roughly $9.5 billion in implicit subsidy from the launch business to the constellation business. That number is roughly equivalent to Starlink’s 2025 EBITDA. The profitability of Starlink relative to its competitors is, in substantial part, a consequence of vertical integration that no competitor can match.

Amazon Leo’s business case depends on launches by United Launch Alliance, Blue Origin, Arianespace, and SpaceX itself at commercial list prices. Amazon announced $10 billion in launch contracts in 2022, with total launch spending through full deployment projected to exceed $15 billion. Eutelsat OneWeb’s deployment economics have depended substantially on European sovereignty subsidies because the unit economics of launching its constellation on commercial providers do not close without external support. The Chinese Guowang constellation operates through state-owned contractors and receives state subsidies that are opaque to external observers but serve a similar function to the SpaceX transfer-pricing advantage. The net effect is that Starlink’s primary global competitors are either state-subsidized (Guowang, to a degree OneWeb) or bearing full commercial launch costs (Amazon Leo). Neither scenario reflects a level playing field.

The antitrust implications are worth naming even if they may not be acted upon. SpaceX charges external launch customers one price and launches its own satellites at cost. In most industries, transfer pricing of that magnitude between vertically integrated business lines attracts regulatory scrutiny, particularly when it creates durable competitive advantage. The FCC has not raised this issue in any public proceeding. The Federal Trade Commission and Department of Justice have not opened investigations. The European Commission has considered Starlink’s market position but has not taken action on transfer pricing. The SpaceX IPO will require disclosure of intercompany pricing arrangements that may make the topic harder to avoid. Prospective institutional investors should understand that a significant fraction of Starlink’s margin depends on pricing arrangements that could attract regulatory attention post-IPO. A future FCC or European Commission determination that SpaceX must charge itself market rates for internal launches, or that it must lower its external launch pricing to match its internal economics, would materially affect Starlink’s reported profitability.

A secondary implication is that Starlink’s reported profitability overstates the sustainable unit economics of a standalone satellite broadband business. If Starlink were spun off or acquired, it would need to pay market rates for launch services, which would compress EBITDA margins from 63 percent toward levels more consistent with a standard telecommunications infrastructure business (15 to 30 percent). The $1.75 trillion SpaceX IPO valuation implicitly prices the integrated business. The same valuation applied to Starlink on a standalone basis would be substantially lower, perhaps $400 billion to $600 billion rather than the $500 billion to $800 billion implied by the Morgan Stanley and Adam Jonas analysis.

A third implication, visible to any capital allocator with financial modeling discipline, is that SpaceX’s launch business revenue from external customers effectively double-counts when the integrated business is valued at current multiples. The external launch revenue, approximately $3 billion in 2025, contributes to the SpaceX headline revenue total. The same launch capacity, deployed for Starlink at internal cost, contributes to Starlink’s revenue and margin. A spin-off of Starlink would require a recast of the consolidated financials that makes the transfer-pricing subsidy explicit. The SEC filing for the SpaceX IPO will likely disclose these arrangements, and the market’s reaction to that disclosure will be an important near-term test of whether the IPO valuation holds up.

Transportation Costs Are Rising, Not Falling

The most important single data point in the 2026 commercial space outlook is that launch pricing has reversed direction. Falcon 9 dedicated pricing rose from $70 million to $74 million in February 2026, and rideshare rates rose to $7,000 per kilogram. SpaceX had previously stated that rideshare pricing would increase by approximately $500 per kilogram annually to keep pace with inflation, so the 2026 rate is consistent with the announced trajectory. The $74 million dedicated Falcon 9 price represents the first list-price increase since the early 2010s. The pricing trajectory that most forecasts assumed, in which launch costs fall steadily toward a $20 million or $30 million target by 2030, has reversed at the industry’s dominant operator.

NASA’s own Ignition briefing materials released in March 2026 confirmed the inversion directly. The agency’s assessment states that transportation costs are increasing, not decreasing, against the original Commercial LEO thesis that assumed bulk commercial transportation buys would create discounts. The NASA Space Operations Mission Directorate stated that the market for commercial stations and their underlying transportation is not developing as rapidly as previously expected. That is the US civil space agency contradicting the private-sector forecasts that underpin every trillion-dollar projection.

The consequences for constellation operators other than SpaceX are severe. Amazon Leo needs approximately 1,618 satellites deployed by July 2026 under its FCC license, with an extension request pending for a 2028 deadline. Each New Glenn launch can carry 27 or 29 Amazon Leo satellites to medium Earth orbit. Amazon needs on the order of 55 to 60 launches to meet the extended deadline. If Amazon pays market rates of $74 million per Falcon 9 or approximately $67 million per New Glenn, the deployment launch bill will exceed $4 billion. The original Amazon Leo deployment economics assumed a declining launch cost trajectory that would have materially reduced that total. The rising trajectory raises the program’s break-even subscriber count substantially. Amazon’s willingness to absorb the higher costs reflects strategic commitment from Jeff Bezos rather than standard financial discipline, but even strategic commitments have budget limits.

Eutelsat OneWeb faces a similar problem. The first-generation constellation is aging, and replenishment launches are scheduled to begin in 2026 and 2027. The operator has announced plans for a second-generation constellation with first launches targeting 2027, but the deployment financing depends on launch costs falling rather than rising. The European Union’s IRIS² program is sized against a specific European launch capacity that is less competitive than Falcon 9 on price, and the launch cost assumptions in the IRIS² business case are optimistic against the 2026 trajectory.

The Chinese Guowang and Qianfan constellations face a different cost environment that is partially insulated by state launch subsidies but not fully immune to global launch market dynamics. China’s launch cadence of 90 missions in 2025 reflects substantial government investment in launch capacity that has largely not been commercially validated. The unit economics of Chinese Long March launches are reported at approximately $30 million to $50 million per mission depending on configuration, which is lower than Falcon 9 list price but higher than Falcon 9 internal cost. Chinese constellations therefore face roughly the same transfer-pricing disadvantage against Starlink that Western commercial competitors face, though the gap is smaller and is partially offset by lower labor and manufacturing costs.

Rocket Lab’s Neutron targets $50 million per launch, and the economics require Falcon 9 pricing to hold rather than decline. The rising Falcon 9 price is nominally good news for Neutron’s competitive position, but it also reflects that the market is capacity-constrained rather than demand-constrained. If demand ly exceeded supply, Neutron’s market entry would be welcomed by customers at premium prices. In practice, Neutron’s first-flight slip from Q1 2026 to Q4 2026 has been accommodated by the market without pricing pressure on SpaceX, which suggests that SpaceX pricing power has grown rather than shrunk.

The broader implication is that every commercial space business case built on declining launch costs is now operating against deteriorating rather than improving unit economics. Starlink is protected by vertical integration. No other business is. Space tourism, in-space manufacturing, commercial space stations, satellite services, and deep-space exploration all become more expensive in the near term. The forecasts that price these segments at growth rates above 10 percent annually are implicitly assuming a launch cost trajectory that is no longer visible in the data. Capital allocators building positions in non-SpaceX commercial space operators are accepting exposure to an unfavorable launch cost trajectory that may not be fully reflected in current valuations.

The Spectrum Chokepoint Nobody Is Pricing

The single most underexamined constraint on satellite broadband growth is spectrum. Every communications satellite depends on radio-frequency spectrum licenses granted through the International Telecommunication Union (ITU) filing process and enforced nationally by agencies such as the FCC, Ofcom, and their counterparts in every major market. The process is first-come, first-served in principle and increasingly politicized in practice. Spectrum licenses are finite, exclusive, and tradeable, which creates a market for spectrum rights that can rival the value of the underlying satellite hardware.

The $17 billion SpaceX acquisition of EchoStar spectrum licenses in 2025 was one of the largest spectrum transactions in commercial space history, but it no longer stands alone as the defining benchmark for satellite-spectrum scarcity. Amazon’s April 2026 agreement to acquire Globalstar for about $11.57 billion added a second major transaction in the same strategic lane, giving Amazon not only access to satellites and operating infrastructure but also to Globalstar’s mobile satellite services spectrum assets and direct-to-device platform for its Amazon Leo network. Together, the two deals show that spectrum has become one of the most expensive and least replicable inputs in communications markets: SpaceX paid a premium for regulatory control of direct-to-cell frequencies at scale, while Amazon paid a similarly extraordinary sum to secure an integrated spectrum-and-network platform rather than build that position from scratch. The implication for competitors is sharper than before. Any company seeking to match the spectrum depth now being assembled by Starlink and Amazon must either acquire scarce licensed assets at multi-billion-dollar valuations or depend on partnerships with mobile network operators and other license holders that limit strategic and commercial freedom.

The constraint falls unequally on competitors. AST SpaceMobile operates on spectrum licensed through its partnerships with AT&T, Verizon, and other mobile network operators, which limits the company’s independence but also protects it from needing to acquire spectrum directly. The structural trade-off is that AST’s economics depend on the willingness of mobile carriers to share revenue, which is itself a negotiation rather than a property right. Eutelsat OneWeb inherited a spectrum position from OneWeb’s original ITU filings and faces continuing disputes with competing operators over priority. Amazon Leo holds Ka-band licenses granted under strict deployment milestones that Amazon is struggling to meet, and a missed milestone could result in license revocation and reallocation to competing operators. China’s Guowang constellation operates under Chinese spectrum allocations that do not require ITU coordination in the same way Western operators do, which gives China a structural advantage in expanding coverage over time but does not prevent disputes with Western satellite operators over interference and coordination.

The coordination problem is acute because low Earth orbit constellations do not need to be exclusive to a single operator within a given footprint; multiple constellations can coexist if they coordinate spectrum use and protect one another from harmful interference. In practice, those coordination and coexistence arrangements can be lengthy and technically complex, and the resulting constraints can affect the capacity and service quality each operator can deliver. SpaceX has sought FCC action on more than one front, including access to EchoStar’s 2 GHz spectrum through a proposed transaction and separate rule changes that would allow higher power levels for some space-based broadband operations. It is also seeking to shape the debate over the future use of upper C-band spectrum, although that remains a policy fight rather than an awarded operating position. EchoStar opposed some of SpaceX’s positions and later agreed to sell spectrum licenses to SpaceX after facing FCC scrutiny over spectrum and buildout obligations and broader financial strain reported at the time of the deal. Similar disputes are continuing across multiple frequency bands and jurisdictions. Each resolution can shift the competitive landscape in ways that a forecaster applying simple revenue-per-subscriber arithmetic is unlikely to capture.

A secondary spectrum issue is coordination between satellite and terrestrial uses. Mobile network operators view direct-to-cell as both a partnership opportunity and a competitive threat. If Starlink can deliver service directly to smartphones in any market, the operator’s terrestrial network loses the rural and remote portion of its value proposition. The partnerships that currently route Starlink traffic through T-Mobile, OneNZ, and other national operators are negotiated under the assumption that mobile operators retain gatekeeping power over subscribers. That assumption erodes over time as Starlink’s spectrum position strengthens and as subscribers become aware of the technical availability of direct access. AT&T and Verizon have resisted direct-to-cell partnerships with Starlink precisely because they anticipate the competitive threat to terrestrial service.

Regulatory responses are beginning to catch up. The FCC in 2025 opened dockets on satellite direct-to-cell competition and on spectrum coordination between satellite and terrestrial systems. The European Union’s IRIS² constellation is being structured in part to preserve European sovereignty over the relevant spectrum bands. India, Brazil, and several other large markets have introduced licensing processes that impose data localization, traffic interconnection, and revenue-sharing requirements on foreign satellite operators. Each regulatory action tightens the commercial thesis for constellation operators in ways the space economy forecasts treat as cost overhead rather than as fundamental addressable-market compression.

A capital allocator who views Starlink as a global monopoly on satellite broadband is implicitly assuming that spectrum allocation and regulatory permissions will continue to evolve in Starlink’s favor. That is a plausible base case for the next two or three years. It is a substantially less plausible case for the 2030 and 2035 horizons that the trillion-dollar forecasts require. The spectrum-constrained competitors of 2026 will either acquire more spectrum through expensive transactions like the EchoStar deal, lobby successfully for regulatory reallocation in their favor, or remain at subscale. None of those outcomes is automatic. The EchoStar transaction is the clearest signal that spectrum has become the binding constraint on commercial space economy growth, and the forecasts have not absorbed that signal.

Starship Is Critical Path for the Next Decade of Forecasts

SpaceX Starship is the second-order dependency that most space economy forecasts avoid confronting directly. Through April 2026, Starship has launched 11 times with 6 successes and 5 failures. Flight 12, targeting a launch window in May 2026 after Booster 19’s successful 33-engine static fire on April 15, will be the first flight of Starship V3 from the new Pad 2 launch complex at Starbase. Core objectives include a Super Heavy ascent and controlled splashdown, upper-stage Starlink mass simulator deployment, a Raptor 3 in-space relight, and validation of docking and propellant-transfer hardware. The flight does not demonstrate orbital refueling itself.

The Starship Propellant Transfer Demonstration is a separate mission expected later in 2026. NASA’s Human Landing System contract requires approximately ten tanker launches of propellant to a depot in orbit to refuel a single Starship sufficiently to reach the lunar surface with crew. Independent analyses suggest that operational readiness for the depot-and-tanker architecture will likely converge in the 2027 to 2028 timeframe. A NASA Office of Inspector General report in spring 2026 concluded that the Starship lander was behind schedule by “at least two years, with additional delays expected.” That OIG finding alone is enough to require substantial forecast revision, because every commercial space business case that depends on Starship operational service (orbital data centers, deep-space cargo, Mars cargo, lunar surface delivery beyond the early CLPS missions) is now anchored to a timeline that has slipped by two or more years.

Starship’s development record through 2025 suggests that the program is less far along than marketing materials suggested. The company flew Starship just five times in 2025 after predicting up to 25 flights at the start of the year. Several of the 2025 flights experienced heat shield failures, booster landing failures, or upper stage failures that required significant redesign before subsequent flights. The V2 vehicle, introduced in early 2025, experienced consecutive failures before being replaced by V3 for Flight 12. Each vehicle generation improvement absorbs six to twelve months of schedule, and each improvement cycle suggests that the prior design was not ready for the operational missions that marketing materials had described.

The broader concern is that Starship is being developed as a single vehicle designed to meet multiple mission profiles simultaneously: Starlink deployment, crew transport, lunar lander, Mars cargo, orbital refueling depot, and potentially orbital data center host. Each mission profile has different optimization requirements, and the attempt to satisfy all of them with one vehicle is an ambitious engineering program that has historically slipped. The NASA Apollo program developed separate vehicles for each mission profile (Saturn V launcher, Apollo Command Module, Apollo Service Module, Lunar Module) precisely because integrating multiple mission profiles into a single vehicle is difficult. Starship’s attempt to do so reflects SpaceX’s engineering confidence but also creates integration risk that the forecasts do not price.

Each delay compounds across the Starlink manifest, the HLS schedule, the Mars cargo timeline, and the broader commercial constellation market. If Starship is operational in 2028 rather than 2026, Starlink Gen 3 deployment slips, which affects direct-to-cell coverage expansion, which affects the timing of the service’s transition from text-only to voice-and-data, which affects ARPU growth. If Starship is not operational until 2029 or 2030, the HLS lunar landings are postponed, Artemis V is postponed, and the Moon Base Phase 1 timeline becomes unachievable. If Starship never reaches economic operational cadence at the $10 million per flight target Elon Musk has described, then the business cases for orbital data centers, deep-space cargo, and Mars cargo are substantially less favorable than the headline forecasts assume.

A skeptical read is that Starship will likely reach operational service in the 2027 to 2028 timeframe, will not reach the $10 million per flight cost target until the 2030s if ever, and will face continued reliability challenges that limit cadence below the marketing projections for several years after initial operational capability. Those are still meaningful outcomes. They are not the revolutionary cost reductions that the forecasts require.

Launch Market Overcapacity Is the Next Commercial Problem

Blue Origin’s New Glenn reached orbit on its inaugural January 2025 flight and recovered its first booster on its second flight in November 2025. New Glenn now has a growing manifest including BlueBird satellites for AST SpaceMobile, Amazon Leo deployment contracts, and national security launch mission awards. The vehicle has demonstrated orbital capability and booster recovery, but it has not yet demonstrated the cadence required to compete with Falcon 9 at scale. Blue Origin is effectively subsidized by Jeff Bezos and does not need to clear a commercial profitability bar, which gives New Glenn more runway than Neutron or Vulcan Centaur but does not guarantee operational success.

Rocket Lab’s Neutron slipped from Q1 2026 to Q4 2026 after a February 2026 stage-1 tank test failure, with development cost now at approximately $360 million. The slip is a significant setback because Rocket Lab’s commercial thesis depends on Neutron reaching operational service before Falcon 9’s pricing structure stabilizes and before Starship reaches operational cadence. If Neutron is delayed to 2027, the commercial window for a $50 million medium-lift vehicle against Falcon 9 at $74 million narrows considerably. Rocket Lab’s stock through 2025 and early 2026 has priced Neutron success as a material contributor to forward revenue, and the slip has not yet produced the stock-price correction that a failed program launch typically causes. That suggests either that the market is patient with Neutron’s development or that the market has not yet priced the slip appropriately.

The ULA Vulcan Centaur is in operational service and has completed its national security launch certification. Vulcan is not competitive with Falcon 9 on price, but it has a stable backlog of national security missions that are not price-sensitive. ULA, now for sale by Lockheed Martin and Boeing, faces an uncertain corporate future that may constrain its ability to invest in Vulcan Centaur improvements. Ariane 6 is reaching operational cadence and serves primarily the European sovereignty market, with commercial missions secondary to European government payloads. Stoke Space, Relativity Space, and several smaller entrants are each working toward operational service in 2027 or 2028, adding further capacity to a market that may not have commensurate demand.

The 2026 global orbital launch market is projected to exceed 250 missions. SpaceX alone flew 165 Falcon 9 missions in 2025, China achieved 90 launches, Russia managed 17, Europe 8, India 5, and Japan 3. If Starship, Neutron, New Glenn, Vulcan Centaur, Ariane 6, and the Chinese medium-lift vehicles each reach projected cadence, available launch capacity by 2028 would exceed 400 missions per year. Consensus satellite demand forecasts, even including full Starlink Gen 3, Amazon Leo, IRIS², and the Chinese megaconstellations, fall short of that capacity. Structural oversupply is emerging.

The oversupply will manifest first in pricing pressure on vehicles other than Falcon 9 and Starship. If SpaceX controls 70 percent of global launch capacity and sets pricing based on its strategic priorities (funding Starship development, funding Starlink deployment, funding xAI operations), then the remaining 30 percent of launch capacity is distributed among vehicles that must price against SpaceX’s list price rather than against Falcon 9’s internal cost. Neutron at $50 million, New Glenn at $67 million, Ariane 6 at approximately $75 million, and Vulcan Centaur at approximately $90 million to $110 million (including government surcharges) all price within or above the Falcon 9 list price range. If Falcon 9 list pricing declines in response to Starship’s entry, the non-SpaceX operators will face compressed margins that some cannot sustain.

A realistic 2028 launch market scenario includes SpaceX at 70 to 80 percent global share, China at 15 to 20 percent, and a long tail of small operators sharing the remainder. Blue Origin survives on Bezos capital. Rocket Lab captures a meaningful share of medium-lift missions at pricing that produces thin contribution margin. ULA, now for sale, cannot match Falcon 9 unit economics under most scenarios and will consolidate or exit the commercial market, continuing only in national security roles. Ariane 6 has survived only because European sovereignty policy requires a European launcher, and the economics continue to require subsidy. The long tail of small launch vehicles (Rocket Lab Electron, Firefly Alpha, Astra, various Chinese, Canadian, Australian, European, and Indian entrants) will face pricing pressure from rideshare missions on Falcon 9 and from commodity dedicated launch capacity as supply grows.

That concentration, combined with Starlink concentration at the satellite operator level, produces a space economy whose upside is increasingly held by one company rather than diversified across an ecosystem. A skeptical capital allocator should recognize that concentration risk, and should size SpaceX exposure accordingly. An IPO-era SpaceX position that reflects the company’s economic weight in the sector is a significant position, but it is not a diversified sector exposure. True diversification in commercial space requires exposure to defense contractors, satellite operators unrelated to broadband, and infrastructure services that are not dependent on SpaceX’s launch capacity. That investment universe is substantially smaller and less liquid than the headline forecasts suggest.

China’s State-Directed Space Economy Undermines the Commercial Narrative

China achieved 90 orbital launches in 2025, up from 68 in 2024, with state-owned contractors performing the vast majority of missions. The cadence expansion reflects sustained investment by the Chinese government in launch capacity, manufacturing scale, and constellation deployment. The Guowang constellation reached 136 satellites in orbit across 17 launch batches by December 2025, with a near-term target of 400 satellites by 2027 and a full deployment target of 13,000 satellites. The Shanghai-backed Qianfan (Thousand Sails) constellation is following a similar but slower trajectory with a planned 14,000 satellites, primarily deployed by Chinese medium-lift vehicles including the upcoming Long March 8A and Zhuque series from commercial operators such as LandSpace.

Both Guowang and Qianfan are explicitly positioned as Chinese responses to Starlink and are funded through a mix of state subsidies, state-owned enterprise investment, and municipal backing rather than commercial capital. Guowang is operated by China Satellite Network Group Co., Ltd., a state-owned enterprise established specifically for the program. Qianfan is backed by Shanghai Spacecom Satellite Technology and by the Shanghai municipal government. Neither operator is subject to the commercial return requirements that a Western venture-backed operator would face. The unit economics of both constellations are partially subsidized through state-controlled launch pricing, state-controlled manufacturing, and state-financed working capital that is not available to commercial competitors.

China’s Tiangong space station is operating continuously with regular Shenzhou crew rotations. The station handled its first major human spaceflight emergency in 2025 when a suspected debris strike damaged the Shenzhou 20 spacecraft’s porthole, requiring crew to shelter in the Shenzhou capsule while the damage was assessed. The incident was resolved without loss of crew, but it marked the first documented major anomaly on the Chinese station and demonstrated that China’s LEO operations face the same risks that the ISS has managed for 25 years. Tianwen-2 launched in May 2025 on an asteroid sample return mission intended to demonstrate interplanetary sample-return capability before the larger Tianwen-3 Mars sample return mission targeted for 2028. Chang’e 7 is scheduled for a lunar south pole landing in August 2026 with capabilities including a hopping explorer designed to investigate permanently shadowed craters for water ice. The Mengzhou crewed spacecraft and the Long March 10 crew launch vehicle are scheduled for debut flights in 2026 as part of China’s program to land astronauts on the Moon before 2030.

The Chinese commercial space sector has grown rapidly since approximately 2017. Private launch providers such as LandSpace, Space Pioneer, Galactic Energy, and iSpace have each developed operational or near-operational vehicles. CAS Space, a commercial spin-off from the Chinese Academy of Sciences, has demonstrated the Lijian-1 and is developing larger vehicles. The commercial Chinese launch sector now accounts for a meaningful fraction of China’s total launch cadence, although the distinction between “commercial” and “state-directed” in the Chinese context is substantially different from the Western framework. Chinese commercial launch companies typically receive state investment, state contracts, and state-backed working capital, which makes them closer to state-owned commercial entities than to independent commercial operators.

The strategic implication for the space economy forecasts is that a significant fraction of forward market growth will accrue to state-directed Chinese competitors rather than to US commercial operators. Guowang will capture share of global broadband demand in markets where China has diplomatic and trade leverage, which includes much of Africa, parts of Southeast Asia, parts of Latin America, and some Eastern European and Middle Eastern markets where US technology export controls limit Starlink’s reach. Chinese launch services will compete for international commercial payloads where price-sensitivity is high and where national security considerations are secondary. The headline space economy totals that Western forecasters produce typically reflect a US-centric view of global demand that understates Chinese competitive intensity.

A second implication is that the geopolitical dimension of the space economy will intensify through the 2030s. US export controls under ITAR and EAR have constrained US commercial operators’ access to Chinese markets for decades, and they will continue to do so. Chinese restrictions on foreign satellite operators and on data localization will constrain US commercial operators’ access to Chinese markets. The result is a bifurcated commercial space economy where US operators and Chinese operators compete for different subsets of the global customer base rather than for the same universal market. A trillion-dollar forecast that treats the global space economy as a single competitive market is, structurally, an overestimate of the addressable market for any single operator.

ISRO in India operates on approximately $1.5 billion in annual budget, conducting just five orbital launches in 2025 against a planned 10. India’s LVM3 launched the 6,100 kilogram AST SpaceMobile BlueBird Block-2 satellite on December 24, 2025. The broader pattern is that the space economy is globalizing, with a commercial US core, a state-directed Chinese peer, an Indian program moving toward commercial participation, and a constellation of smaller national programs each building sovereign capability rather than converging into a unified commercial marketplace. Japan’s H3 rocket has reached operational service. South Korea’s Nuri has flown successfully. The United Arab Emirates, Israel, and Brazil each have emerging commercial space sectors. The sum of those national efforts contributes to the headline space economy totals, but most of them reflect national sovereignty investments rather than commercial growth opportunities.

European Sovereignty Spending as a Structural Shift

The European Space Agency’s IRIS² (Infrastructure for Resilience, Interconnectivity and Security by Satellite) programme is targeting a 290-satellite constellation for secure government communications and commercial broadband, with first launches in 2029 and full operational capability by 2030. The programme is funded at approximately EUR10.6 billion in public and private commitments, with approximately EUR6 billion from the EU budget and the balance from member states and industrial partners. IRIS² is explicitly structured to provide European sovereignty independent of Starlink and Chinese alternatives, and it is the clearest example of the broader European response to perceived US space dependence.

France, Germany, Italy, and the UK each increased national space budgets in 2025 and 2026. France’s CNES operates with approximately EUR3 billion annually. Italy’s PRORA programmes have expanded through the space defence budget. Germany’s national space activities have grown alongside the broader German defence budget increase following the 2022 Zeitenwende commitment. The UK Space Agency budget has grown modestly alongside increasing defence space investment. The cumulative effect is that Europe’s contribution to the global space economy reflects a mix of commercial activity and explicit sovereignty-driven spending, with the sovereignty component growing faster than the commercial component.

The thesis that a unified global commercial space economy will emerge has weakened as European sovereignty spending has solidified. IRIS², Galileo, Copernicus, and the national programs that feed them are sized to provide European independence rather than to compete in a global commercial market. The revenue associated with those programs appears in space economy totals but does not reflect the kind of commercial expansion the forecasts assume. A skeptical read is that two of the three largest space spending blocks (China and Europe) are explicitly non-commercial, which leaves the US commercial sector as the single source of commercial revenue growth in the global total. That is a narrower base than the trillion-dollar thesis requires.

European launch capacity is itself a skeptical case study. Ariane 6 reached operational status in 2024 after extensive delays, at a cost and cadence that is not competitive with Falcon 9 on commercial metrics. The vehicle survives because European sovereignty policy requires a European launcher, and the European Commission and ESA have structured the launch market to guarantee Ariane 6 a stable baseload of government payloads. The commercial addressable market for Ariane 6 outside European sovereignty payloads is limited. The same structure applies to the smaller European launch vehicles (Vega-C, and the forthcoming launch vehicles). Europe’s launch capacity is a policy commitment rather than a commercial proposition.

Defense Spending Concentration: Real Numbers

The defense and intelligence contribution to space economy growth deserves specific quantification. The US Space Force budget for FY2026 is approximately $29 billion. The National Reconnaissance Office budget is classified but estimated by independent analysts at $15 to $20 billion annually. The Space Development Agency budget is approximately $3 to $4 billion. The National Geospatial-Intelligence Agency spends roughly $6 to $7 billion annually across all functions, with a meaningful fraction directed to commercial imagery procurement. Adding Defense Advanced Research Projects Agency space-related programs, Missile Defense Agency space-segment spending, and the Department of Defense tenant share of Golden Dome, total US government space-related spending reaches approximately $60 to $70 billion annually as of FY2026.

That figure alone represents roughly 10 to 11 percent of the global $626 billion space economy headline. Adding civil NASA spending of approximately $27 billion and other US government space-related spending brings the US government share of the global space economy to approximately 14 percent. Including NOAA, FAA, and other civilian agency spending, the total US government contribution reaches approximately 15 percent of the global headline.

The parallel dynamics in Europe add another 6 to 8 percent through ESA budgets, national defense space budgets, and the IRIS² financial commitments. China’s 2025 space program spending, partially opaque but estimated at $15 to $20 billion annually including both civil and military components, adds another 2 to 3 percent. Russia’s roughly $5 billion space program spending adds less than 1 percent. The remaining space programs of Japan, India, Canada, and other countries sum to approximately 2 to 3 percent.

The total government share of the global space economy is therefore approximately 28 to 32 percent by this accounting, which closely matches the $138 billion government spending figure Novaspace reported for 2025. The remaining 68 to 72 percent is commercial revenue, the vast majority of which is satellite broadcasting (declining), satellite communications including Starlink (Starlink-concentrated), and consumer navigation equipment sales (GPS-dependent on government subsidy). The true new commercial revenue that the forecasts project will drive growth to the trillion-dollar scale is a small fraction of the current total.

The Golden Dome missile defense program, announced in early 2025 and under active development through 2026, commits the United States to a multi-layered space-based interceptor architecture that could absorb tens of billions of dollars in additional spending. Early cost estimates from the Congressional Budget Office suggested a lifecycle cost of $542 billion through 2045, though the program’s actual cost will depend on the specific architecture selected and on the program’s technical success. A $542 billion lifecycle cost distributed across 20 years implies approximately $25 billion per year in additional US space spending, which would expand US government space spending to approximately $85 to $95 billion annually at program maturity. That is a significant expansion of the defense space budget, but it is also a specific program with specific political and technical risks that could materially reduce the realized spending.

Defense budgets are subject to political volatility, priority shifts, and competing demands from non-space military programs. A change in administration, a recession that compresses discretionary spending, or a diplomatic breakthrough that reduces perceived threats could each shift the defense space spending trajectory. The growth of defense space spending from approximately $35 billion in 2018 to approximately $70 billion in 2026 reflects a specific geopolitical environment that has produced historically unusual budget expansion. A reversion to pre-2018 growth rates would shave tens of billions from the trillion-dollar projections. The forecasts that treat defense demand as a stable growth pillar are implicitly assuming that the current policy environment persists, which is a historically unreliable bet.

The commercial space companies benefiting most from this spending are the prime contractors (Lockheed Martin, Northrop Grumman, Boeing, L3Harris) and a smaller group of newer entrants (Anduril, Planet Labs, BlackSky, Rocket Lab through its Geost acquisition). Their earnings track federal appropriations cycles rather than commercial product markets. The investment case for these companies is a defense contractor case, not a commercial space platform case, and their valuations should reflect the associated risk-return profile rather than the aggressive growth multiples that platform businesses command.

SPAC-Era NewSpace Valuations Still Sit Deep Underwater

Virgin Orbit filed for Chapter 11 bankruptcy in April 2023 and its assets sold for approximately $36 million. The company had reached a peak market capitalization of $3.2 billion after its December 2021 SPAC merger with NextGen Acquisition. Its failure reflected specific operational issues (a failed UK launch in January 2023, aggressive cash burn, a failed capital raise), but it also reflected the broader difficulty of building a profitable small-launch business against SpaceX rideshare economics. Virgin Orbit’s LauncherOne air-launched vehicle was technically viable but commercially uncompetitive at the rideshare prices SpaceX offered. The company could not reduce its launch cost fast enough to match SpaceX, and it did not have a captive customer base large enough to support the premium pricing its cost structure required.

Astra Space completed a take-private transaction in 2024 at $0.50 per share, a 96 percent decline from its SPAC-era peak. The company had reached a market capitalization of $2.1 billion in 2021 on the promise of daily small-launch cadence at very low unit costs. The reality through 2022 and 2023 was that Astra’s Rocket 3 vehicle failed repeatedly, and the company pivoted to spacecraft propulsion and larger launch vehicles without reaching operational cadence on either. The take-private at $0.50 represented a total loss for public market investors and a significant but smaller loss for the founders and early venture backers who retained private equity.

Virgin Galactic reported 2025 revenue of $2 million, a net loss of $279 million, and negative free cash flow of $438 million. The company entered 2026 with $338 million in cash and a going-concern disclosure. From the 2021 meme-stock peak above $55 per share, Virgin Galactic traded below $3 through early 2026. The company’s business model depends on a next-generation Delta-class vehicle that is not yet operational, and the bridge to operational service requires continued equity financing at dilutive prices. Virgin Galactic’s situation illustrates the specific problem of space tourism as a standalone business: the capital required to deliver the vehicle is substantial, the operational window between vehicle delivery and further vehicle development is narrow, and the customer base is small enough that missed operational milestones directly affect revenue without a diversified customer buffer.

Planet Labs has executed operationally, with a functional Earth observation constellation and steady revenue growth, but its market capitalization remains well below the 2021 SPAC reference level. The company’s 2024 revenue of approximately $244 million was below the pre-SPAC financial projections that implied roughly $700 million by early 2026. The miss reflects the commoditization of Earth observation discussed elsewhere in this article, and it illustrates that even operational execution does not always validate SPAC-era valuations when the underlying market has grown more slowly than projected.

BlackSky has executed a series of convertible note issuances and continues to operate at a cash burn rate that requires periodic recapitalization. Spire Global’s reverse stock split and maritime divestiture reflect the same pattern: the business continues to operate, but the public-market valuation has reset to a level consistent with its actual operational and financial metrics rather than the aggressive projections that supported the SPAC merger.

Momentus Space has executed multiple reverse stock splits and has effectively become a zombie stock, with a market capitalization that reflects a speculative option value rather than a rational valuation of the underlying business. Satellogic trades at levels broadly consistent with a 90 percent decline from peak, with revenue that remains small relative to its constellation deployment and no clear path to cash-flow profitability.

Rocket Lab reported Q3 2025 revenue of $155 million, up 48 percent year over year, with record GAAP gross margins of 37 percent and a contract backlog of $1.1 billion. The company guided Q1 2026 revenue to $185 million to $200 million with a guided adjusted EBITDA loss of $21 million to $27 million. Rocket Lab completed the $325 million acquisition of Geost in Q3 2025 and closed its acquisition of Mynaric in April 2026. Rocket Lab is the clearest exception to the SPAC-era pattern: the company has executed operationally, diversified its revenue base through acquisitions, and built a credible case for long-term commercial relevance. The Rocket Lab trajectory from 2021 SPAC to 2026 operational maturity is the template that disciplined space SPAC investors hoped would be the norm rather than the exception.

Intuitive Machines is a second exception. The company’s Blue Ghost-class IM-1 mission in February 2024 achieved a partially successful soft landing (the first commercial lunar landing, though the vehicle tipped over after landing), and IM-2 in early 2025 failed. The stock has been volatile but has not collapsed, reflecting the market’s willingness to price NASA CLPS backlog as a stable revenue base even with mission-level variability.

Firefly Aerospace: The IPO Template Under Pressure

Firefly Aerospace went public on the Nasdaq in August 2025, raising $868 million at an $8.5 billion closing-day market capitalization after pricing shares at $45 above the indicated range. The stock reached an all-time high of $73.80 on August 7, 2025 on its debut session. The 2025 full-year revenue reached $159.9 million, a 163 percent year-over-year increase driven substantially by the November 2025 SciTec acquisition. The 2025 net loss was $334 million.

The stock traded down sharply through the fall and winter of 2025. On November 21, 2025, Firefly hit an all-time low of $16.00, a decline of approximately 78 percent from the IPO peak. The March 19, 2026 Q4 2025 earnings print showed the stock fall an additional 4.76 percent to $22.09 pre-market despite an earnings beat, reflecting investor concerns about continued losses and market volatility. Analyst price targets have been cut repeatedly. Cantor Fitzgerald lowered its target from $65 to $35. Goldman Sachs lowered its target from $32 to $29. The average analyst price target sits at approximately $37.86 against a late-April 2026 trading price of approximately $43.60, implying a downside bias from consensus analysis.

Firefly’s April 2026 market capitalization of approximately $7 billion is below the IPO closing-day level of $8.5 billion and well below the August 2025 peak of $11 billion. The stock has recovered from its November 2025 lows on the back of space sector enthusiasm driven by the SpaceX IPO announcement and the NASA Ignition program, but it has not regained its initial investor-debut valuation. The 2026 revenue guidance of approximately $435 million with 80 percent already booked is strong on an absolute basis, but the operating losses continue and the path to profitability remains unclear. Firefly’s backlog grew 22 percent year-over-year to $1.4 billion. The trajectory looks more like a normal growth company working through its development stage than like a platform business on the cusp of explosive expansion.

The Firefly post-IPO performance is informative for the broader thesis. The IPO was structured around concrete operational milestones (Blue Ghost Mission 1 lunar landing, CLPS backlog, Alpha rocket operational history), which was the disciplined template that skeptical observers praised as a contrast to the SPAC cohort. The post-IPO performance shows that even a disciplined space company IPO faces immediate public-market valuation pressure when quarterly losses persist and when the macroeconomic environment for growth stocks tightens. The Firefly experience suggests that the anticipated SpaceX IPO will face similar pressure in its first year of trading even if it prices at or near its $1.75 trillion target. The post-IPO drawdown risk for SpaceX is larger than the consensus appears to anticipate.

Another feature worth naming is that AE Industrial Partners, the private equity sponsor that took Firefly public, retains a 41 percent stake and controls five of nine board seats. The company’s public-market float is smaller than its headline market capitalization implies, which amplifies price volatility and concentrates governance in the sponsor’s hands. Similar sponsor concentration will likely apply to SpaceX post-IPO, where Elon Musk and long-tenured insiders will retain governance control through a dual-class share structure. Institutional investors who buy into these offerings are accepting limited governance rights in exchange for exposure to the underlying business.

The Venture Capital Retreat and Down Rounds

Novaspace put total private investment in space at $9 billion in 2025, described as “the strongest annual increase since the peak of 2021” but still a fraction of the approximately $15 billion invested at the 2021 high. The headline number masks a more telling story about the companies that raised capital during the 2020 to 2022 boom and returned to the market in 2024 and 2025.

Many of those companies raised substantial down rounds. Axiom Space reportedly raised at a valuation below its prior round in 2024 and 2025 as it worked through cash flow issues and leadership changes. Spire Global executed a reverse stock split in 2023 and divested its maritime business line in 2024 to maintain liquidity. BlackSky issued $160 million in senior convertible notes in 2025 alongside disappointing preliminary quarterly results. Momentus Space executed multiple reverse stock splits to maintain Nasdaq listing compliance, with the cumulative effect of a decline exceeding 99 percent from its 2021 SPAC-era peak. Satellogic trades at levels broadly consistent with a 90 percent decline from peak.

The capital markets message is unambiguous. Investors who funded the 2020 to 2022 cohort have largely accepted that their initial entry prices were too high and have reset their cost basis through dilutive follow-on rounds, debt financing with aggressive terms, or take-private transactions at deep discounts to original valuations. The 54 mergers and acquisitions Novaspace documented in 2025 include many transactions that would more accurately be described as distressed sales than as value-creating combinations. The late-stage concentration in private capital flows reflects investors backing proven operators with visible revenue rather than expanding the base of funded companies.

A complementary observation is that the retail investor base that absorbed losses from the SPAC cohort has returned to the sector around Rocket Lab, AST SpaceMobile, and anticipation of the SpaceX listing. Retail enthusiasm for space stocks in 2025 and 2026 has driven valuations higher in those names than operational metrics alone would support. Rocket Lab’s valuation through 2025 expanded substantially above the company’s revenue and earnings trajectory, reflecting anticipatory positioning against the Neutron first flight. AST SpaceMobile’s valuation has moved with speculation about direct-to-cell partnerships rather than with disclosed contract revenue. The pattern resembles the setup that preceded the SPAC cohort’s disappointments. Skeptical observers should expect some portion of the 2025 to 2026 retail enthusiasm to produce similar drawdowns if operational milestones slip or if the SpaceX IPO prices at a discount to the target valuation.

Earth Observation Has Been Commoditized

The Earth observation market is large enough to warrant specific skeptical treatment. Grand View Research valued the EO market at $5.1 billion in 2024 and projected growth to $7.2 billion by 2030, which is modest compared to the trillion-dollar space economy narrative. Fortune Business Insights valued the 2025 market at $7.04 billion. Maxar Technologies led the commercial EO market with approximately 21.3 percent share in 2024. The top five firms collectively held 56.8 percent of the market in 2024, which reflects a consolidating rather than fragmenting structure.

The dynamic most visible in 2024 and 2025 was commoditization. The proliferation of Chinese Earth observation satellites has driven pricing pressure across the commercial segment. Open-source data from Copernicus Sentinel, Landsat, and other government programs continues to undercut pricing for commoditized imagery products. Premium providers now must justify pricing through analytics packaging, guaranteed revisit rates, and spectral richness rather than through imagery alone. Planet Labs secured a $280 million contract with the German government in July 2025 for environmental monitoring, which was a meaningful win but also demonstrated the contract structure emerging as the norm: multi-year government commitments with built-in analytics deliverables rather than pure imagery sales.

The 2024 Maxar-Satellogic partnership integrated high-resolution tasking with frequent-revisit capabilities, reflecting the pricing pressure on pure-play imagery providers. Airbus’s sale of its UP42 marketplace to Neo Space Group in July 2025 consolidated distribution while curating third-party algorithms. ICEYE announced plans for an IPO to fund constellation expansion and SAR capability upgrades. Each of those transactions reflects the market consolidation that follows commoditization, not the expansion that the 2020-era forecasts assumed.

Maxar’s take-private transaction in 2023 at approximately $6.4 billion (including debt) was a concrete test of the commercial EO thesis. Maxar had been the clearest commercial champion of high-resolution Earth observation, and its private equity sponsors (Advent International) executed the transaction at a valuation that implied the public market had been overvaluing the business at its 2021 peak. The company subsequently split into Maxar Intelligence and Maxar Space Systems, with the geospatial data business continuing its 10-year $3 billion contract with the National Reconnaissance Office. Planet Labs revenue reached approximately $244 million in 2024, well below the pre-SPAC projections that implied $700 million by early 2026.

The skeptical read of the EO market is that it has settled into a legitimate but small government-and-defense services business with modest commercial adoption. Commercial customers in agriculture, insurance, logistics, and supply chain monitoring have grown steadily but not explosively, and they tend to buy analytics-packaged products rather than raw imagery. The forecasts that projected $10 to $20 billion commercial EO revenue by 2030 were built on assumptions about commercial enterprise adoption that have not materialized at that scale. A realistic EO market by 2030 is probably in the $10 to $12 billion range, with government and defense customers providing the majority of revenue and the commercial segment growing at single-digit rates.

Space Tourism Has Not Produced a Commercial Market

Across Blue Origin’s New Shepard, Virgin Galactic’s VSS Unity, and Axiom Space’s orbital missions, fewer than 100 individuals have flown as paying or sponsored passengers between 2021 and early 2026. Axiom Mission 4, launched June 25, 2025, carried astronauts from India, Poland, and Hungary for more than 18 days under commander Peggy Whitson. NASA has approved a fifth Axiom mission for early 2027, with Vast also receiving a separate Private Astronaut Mission award in February 2026 for a 2027 flight.

NASA’s own March 2026 assessment, in the Staying in Low Earth Orbit briefing materials, stated directly that “tourism has not materialized as a market,” with sovereign entities funding short-duration missions but not long-duration ones. That is the US civil space agency confirming what a skeptical analysis would conclude. The NASA briefing also stated that “after 25+ years, no breakthroughs, products, or capabilities that generate demand” have emerged from commercial use of the International Space Station, that NASA continues to offset infrastructure costs, and that “there is no independently-verifiable market research indicating economic viability of a commercial station partially funded by NASA.” Those are not skeptical criticisms from an outside observer. They are the findings of the space agency that spent $100 billion building the ISS and has the clearest view of its commercial performance.

Forecasts for space tourism that feature in Morgan Stanley, McKinsey, and Grand View Research publications project compound annual growth rates of 35 to 45 percent through 2030, with total addressable markets of $8 to $12 billion by the end of the decade. Virgin Galactic’s original ticket price of $200,000 in 2005 rose to $450,000 in 2021 and again to $750,000 in 2026, each step without demonstrated demand elasticity. The Virgin Galactic 2025 financials show the economics of this segment transparently: $2 million in revenue, $279 million in losses. The skeptical read on space tourism is that the segment will grow slowly, at a scale that matters to a handful of small operators rather than to the headline space economy forecasts.

Blue Origin’s New Shepard suborbital program has flown nearly 100 people since its first crewed mission in 2021, but its recent trajectory does not support the idea of sustained expansion at a steadily rising cadence. After a limited number of human flights in 2024 and 2025, Blue Origin said in January 2026 that it would pause New Shepard operations for at least two years, indicating that the program had not become a scaled, continuously growing business line. Public reporting does not provide enough detail to characterize New Shepard as a meaningful profit center within Blue Origin’s broader operations, and it is more accurately described as a high-profile human-spaceflight program than as a proven standalone earnings engine. The orbital private-astronaut market, led operationally by Axiom Space’s missions to the International Space Station, also remains small in absolute terms, with recent flights still relying heavily on government-backed national astronaut missions as well as private customers.

A realistic forecast for space tourism through 2030 places the segment at perhaps $500 million to $1.5 billion in annual revenue, concentrated across three or four operators. That is a real market but not a material contributor to the trillion-dollar space economy thesis. Capital allocators betting on space tourism growth at the 35 to 45 percent CAGR that some forecasts imply are betting on demand elasticity that has not been demonstrated at current ticket prices.

Axiom Space: The Case Study of Commercial Station Economics

Axiom Space deserves specific treatment because it is the most operationally mature commercial station program and its financial trajectory provides the clearest test of whether the business model can work. Axiom has flown four successful Private Astronaut Missions to the International Space Station between April 2022 and June 2025. It holds NASA’s original Axiom Segment contract awarded in February 2020 to attach at least one habitable module to the ISS. It has won Artemis spacesuit contracts. It has raised more than $505 million in private equity through multiple rounds.

Despite those accomplishments, Axiom is the clearest illustration that the commercial station business model does not work at current prices. A September 2024 Forbes investigation documented approximately 100 layoffs, voluntary 20 percent pay cuts for remaining employees, struggles making payroll, and late payments to contractors including SpaceX. A former executive told Forbes that “with every round, as soon as it came in, you pay SpaceX, you pay Thales Alenia, you pay your bills and then it’s gone.” Axiom had talks about scaling back its space station design to a much smaller configuration that some investors considered less commercially lucrative and possibly more expensive than the original plan.

The leadership turnover is diagnostic. Co-founder and original CEO Mike Suffredini stepped down in August 2024, citing personal reasons against a backdrop of cash flow issues. Co-founder Kam Ghaffarian served as interim CEO through April 2025. Tejpaul Bhatia, formerly Chief Revenue Officer and a Google Cloud executive, was appointed CEO in April 2025. Bhatia was replaced by Dr. Jonathan Cirtain in October 2025 after just six months, making Cirtain the company’s third CEO in approximately 14 months. Ghaffarian continues as Executive Chairman and has reportedly been propping up the company with his own funds.

The financial trajectory is equally instructive. Axiom’s March 2025 funding round valued the company at $2 billion, a notable decline from its $2.6 billion Series C valuation in 2023. That is a down round of approximately 23 percent despite successful mission operations, spacesuit contract wins, and continued ISS access. The December 2025 commitment of $100 million from Hungary’s 4iG provided additional runway. Hungary’s interest reflects sovereign astronaut demand, which is exactly the customer segment NASA’s March 2026 briefing identified as funded but limited.

Axiom leadership has acknowledged that its private astronaut missions have been financially challenging, with Forbes reporting in September 2024 that Kam Ghaffarian said the company’s first three private astronaut missions were loss-making and that its next mission was expected to come closer to breakeven rather than generate strong margins. Public reporting has also described the requirement for a former NASA astronaut to command these missions as a meaningful cost factor, reducing mission revenue by approximately 25 percent. Each of those statements is revealing. “Near breakeven” at operational scale suggests that the business, in its current configuration, was still focused more on covering mission costs and sustaining operations than on producing substantial surplus cash from private astronaut flights.

The implications are significant for the broader commercial station thesis. If Axiom, which has executed operationally and holds the most favorable contract structure with NASA, cannot reach profitability on four ISS missions, no competitor is likely to reach profitability on a free-flying station without substantial additional NASA subsidy. Vast Space, Voyager Starlab, and Blue Origin Orbital Reef all face more challenging development paths, less operational validation, and the same fundamental demand problem: there is no commercial customer base at scale for a commercial space station.

The March 2026 Ignition restructuring can be read as NASA’s accommodation of this reality. The government-owned Core Module approach substantially reduces the capital that commercial station operators must raise, shifts the foundational infrastructure risk to NASA, and allows commercial module providers to operate as contractors rather than as prime operators. Axiom’s AxH1 module, scheduled to dock with the ISS in late 2026, fits the Ignition framework reasonably well because it is designed to attach to an existing host platform. The Haven-2, Starlab, and Orbital Reef programs, which were designed around independent free-flying station architectures, fit the Ignition framework more poorly and will need to adapt.

Axiom CEO Dr. Jonathan Cirtain, who took over in October 2025, inherited a company that must both deliver AxH1 to the ISS on schedule in late 2026 and position for the Ignition Core Module and commercial module procurement that opened in March 2026 with RFIs. The company has the most operational credibility of any competitor, but it also has the least financial flexibility. Cirtain’s execution over the next 12 months will determine whether Axiom emerges from the Ignition restructuring as the preferred commercial module provider or as a distressed asset that needs to be acquired by a better-capitalized partner.

A skeptical read is that Axiom’s trajectory is the clearest evidence that commercial station economics do not work at current prices and that NASA’s Ignition restructuring is an acknowledgment of that reality rather than a fresh commercial opportunity. Capital allocators who invest in Axiom or its competitors at current valuations are implicitly betting that NASA will continue to subsidize the sector beyond the levels the current budget supports. That is a plausible bet in a specific political environment, but it is not the commercial space thesis that the trillion-dollar forecasts describe.

Artemis Under Ignition Restructuring

The Artemis II mission launched on April 1, 2026 at 6:35 PM EDT from Launch Complex 39B at Kennedy Space Center. The crew consisted of commander Reid Wiseman, pilot Victor Glover, and mission specialists Christina Koch and Jeremy Hansen. The mission completed its splashdown recovery on April 10, 2026 at 8:07 PM EDT off the coast of San Diego. The flight succeeded, but the cost structure that made it possible deserves scrutiny.

NASA’s own 2026 Artemis restructuring materials document the program’s challenges directly. The agency’s March 2026 briefing stated that “for nearly 20 years, NASA and its partners have invested more than $100 billion in a crew-rated lunar transportation architecture that has not yet flown astronauts.” It acknowledged that launch cadence was trending toward one mission every three years, which the agency itself described as “uneconomical, introduces additional risk and requires rebuilding operational muscle memory with each launch.” Multiple oversight reports highlighted cost growth and material risks in execution. NASA leadership concluded that the previous plan “as designed, was unlikely to achieve the highest priority objectives.”

The Space Launch System costs approximately $4.1 billion per launch according to the NASA Office of Inspector General. The Orion spacecraft costs roughly $2.7 billion per vehicle. A single Artemis flight represents over $4 billion in hardware alone. Independent estimates through Artemis IV project cumulative program spending between $93 billion and $105 billion.

The restructured Artemis sequence, announced in February 2026 and elaborated at the March 24, 2026 Ignition event, adds a dedicated Artemis III test mission in Earth orbit in 2027 to validate crew-lander interfaces before attempting a lunar landing. Artemis IV in early 2028 will be the first lunar surface landing, with Artemis V targeting late 2028. The Trump administration’s December 2025 Executive Order 14369 “Ensuring American Space Superiority” required NASA to reform acquisition processes by June 16, 2026 with a preference for commercial solutions and streamlined instruments. Lunar Gateway was paused, freeing resources for surface infrastructure.

NASA’s Moon Base plan unveiled at Ignition outlines a three-phase program at the lunar south pole. Phase 1 through 2029 includes up to 25 missions with 21 landings, delivering four tons of payload across CLPS landers, rovers, hopper drones called MoonFall, and communications relays. Phase 2 from 2029 to 2032 scales to up to 60 tons of cargo across 24 landings with semi-permanent infrastructure including solar and initial nuclear power stations. Phase 3 from 2032 onwards targets up to 38 tons of cargo per year supported by reusable heavy-lift capabilities. The plan depends on international partner contributions including JAXA’s pressurized rover, ASI’s Multipurpose Habitats, and CSA’s Lunar Utility Vehicle. The Canadian Space Agency commitment continues the partnership visible in Jeremy Hansen’s Artemis II role.

The cancellation of Mars Sample Return by Congress in January 2026 reflected fiscal pressure on an $11 billion program. The Lunar Gateway pause in March 2026 freed up roughly $5 billion in programmed investment. NASA’s FY2026 budget reached approximately $27.53 billion when the One Big Beautiful Bill Act supplemental is included, which The Planetary Society calculated is the largest NASA budget since fiscal year 1998 on an inflation-adjusted basis. The budget gives the Artemis program the runway to execute the revised cadence, but it does not address the underlying cost-per-mission problem that the OIG audits have documented for years.

NASA Administrator Jared Isaacman confirmed on February 27, 2026 that SLS will not be used in the long term and that the program’s goal is repeatable, affordable access to the Moon. After Artemis V in late 2028, NASA aims to launch lunar missions every six months using commercially procured and reusable hardware. If Starship reaches operational cadence by 2028, the SLS architecture will be retired faster than its industrial-base defenders in Alabama, Texas, and Florida will accept politically. If Starship slips, SLS may continue through Artemis VI and beyond at a pace that the NASA budget cannot comfortably sustain. Either outcome is consistent with the broader Ignition framing: the commercial alternative is preferred but not yet operational, and the legacy cost-plus architecture will continue until the commercial alternative demonstrably works.

The Nuclear Propulsion Wild Card

The Ignition announcement committed NASA to launching Space Reactor-1 Freedom, the first nuclear-powered interplanetary spacecraft, to Mars before the end of 2028. SR-1 Freedom will demonstrate nuclear electric propulsion in deep space and will deploy an Ingenuity-class helicopter payload called Skyfall upon arrival at Mars. The mission is intended to set regulatory and launch precedent for nuclear hardware and activate the industrial base for future fission power systems across propulsion, surface, and long-duration missions.

The announcement is significant for several reasons. First, it represents the first serious US nuclear propulsion demonstration since the cancelled NERVA program of the early 1970s. Second, it targets a launch window that is remarkably close given the current state of the technology. Third, it implies an accelerated regulatory and partnership arrangement with the Department of Energy that has not historically moved quickly.

The skeptical read is measured but not dismissive. Nuclear electric propulsion is well understood in principle and has flight heritage on Soviet RORSAT missions in the 1970s and 1980s. The technical challenges of scaling a kilopower-class reactor to operational readiness by 2028 are substantial but not fundamentally different from other NASA-led technology demonstrations that have succeeded in comparable timelines. The challenges for Freedom are political and regulatory rather than purely technical. A launched nuclear reactor requires specific safety certifications from the Department of Energy and the US Nuclear Regulatory Commission, and those processes have historically taken longer than the Ignition schedule allows.

If SR-1 Freedom launches on schedule and operates successfully at Mars, the downstream implications for the lunar Moon Base power infrastructure are substantial. The Phase 2 and Phase 3 lunar base plans depend on operational nuclear power for sustained presence through the lunar night. A validated fission power system would be the first real demonstration that NASA’s lunar Moon Base plan is technically achievable at the scale it describes. If Freedom slips by two or three years (a reasonable historical expectation for flagship demonstrator missions), the lunar base architecture faces a power-generation gap that solar alone cannot fill.

For the commercial space economy narrative, nuclear propulsion is a category that was quietly dropped from forecasts after the 2010s. If Freedom succeeds, the category may re-enter the forecasts as a meaningful contributor to deep-space mission economics through the 2030s. The current commercial providers in nuclear-electric space propulsion are essentially zero. If NASA successfully demonstrates the capability, the industrial base that forms around it will accrue to a small number of contractors that have little to do with the trillion-dollar commercial thesis the forecasts describe.

NASA Describes LEO Commercial Failure in Its Own Words

The March 2026 Ignition event included the most candid NASA assessment of commercial LEO markets in the agency’s history. Dana Weigel, the NASA LEO Program Manager, presented the Staying in Low Earth Orbit briefing that acknowledged four findings.

First, after more than 25 years of commercial use of the International Space Station, “no breakthrough products or capabilities that generate demand” have emerged. NASA continues to offset infrastructure costs. There is no evidence of scaled product manufacturing on Earth or in space. The in-space manufacturing thesis that has underpinned Varda Space, Redwire’s former Made in Space business line, Space Forge, and others has not been validated by a quarter-century of ISS operations.

Second, tourism has not materialized as a market. Sovereign entities have funded short-duration missions through Axiom Space, but not long-duration ones. The Axiom missions that have flown have carried astronauts from India, Poland, Hungary, Saudi Arabia, Turkey, Italy, Sweden, and other countries as sovereign-sponsored passengers rather than as private tourists. Private individual demand for orbital missions at approximately $55 million per seat has been essentially zero.

Third, “there is no independently-verifiable market research indicating economic viability of a commercial station partially funded by NASA.” That is a remarkable statement from the agency that has been funding commercial station development through the CLD program since 2020. NASA is acknowledging that its own commercial partners have not produced market research that would justify the commercial station thesis the program was built around. The agency believes independent data is paramount, and the agency does not have it.

Fourth, budgets are inadequate. NASA cannot afford the originally envisioned path of developing two commercial stations, and today cannot even afford one, leaving a shortfall amounting to billions of dollars. The fiscal reality forces a single “winner-take-all” scenario with no redundancy and execution risk in an environment that no provider has comparable experience in.

NASA’s proposed alternate path is an “Incremental Transition Approach” in which NASA procures a government-owned Core Module attached to the ISS forward port. The Core Module would provide propulsion, power, cooling, docking ports, and basic life support. Two commercial modules would then attach to the Core. After maturing technical and operational capabilities, the stations would detach from the ISS as free-flying commercial destinations, and NASA would transition to being one of many customers. The approach is explicitly described as a risk-reduction strategy that leverages ISS infrastructure, robotics, spacewalk capability, and visiting spacecraft to keep astronauts alive in the unforgiving environment of space while commercial capability matures.

The phased procurement timeline NASA released targets core module initial design starting in 2026 with two providers, core module downselect in 2027 to 2029, and commercial module providers in parallel. NASA’s available budget is $250 million per year through the end of ISS, with Phase 1 covering the core module and commercial modules, Phase 2 covering a market-driven power and cooling module, and Phase 3 covering market-driven commercial module expansion or services. The concept RFI was posted on March 25, 2026, with final RFI on April 24, 2026 and draft RFP on June 1, 2026. The agency has committed to a rapid turnaround of industry inputs to proceed toward procurements.

The practical implications are significant. The previous Phase 2 plan under which NASA would fund independent free-flying commercial stations through $1 billion to $1.5 billion in funded Space Act Agreements has been replaced. The Commercial Space Federation responded publicly with President Dave Cavossa testifying that “NASA may now build its own core station module that would compete with industry and require already designed stations to now dock with the ISS to meet other unknown requirements.” The four programs competing for the prior Phase 2 framework, Axiom’s Axiom Station, Vast’s Haven-2, Voyager’s Starlab, and Blue Origin’s Orbital Reef, now face a fundamentally restructured competitive environment.

Vast Space’s Haven-1 remains on a May 2026 launch target aboard a SpaceX Falcon 9 with a first crewed visit in late June 2026, though Haven-1’s Wikipedia entry lists the launch as Q1 2027 reflecting schedule drift. Haven Demo achieved mission success after deploying from Bandwagon-4 on November 2, 2025. Axiom Space’s AxH1 module is scheduled to dock with the ISS in late 2026, with Thales Alenia Space completing the module’s primary structure in Houston. Starlab is targeting 2029 aboard Starship. Orbital Reef has the slowest visible progress.

The realistic outcome under the Ignition framework is that NASA will proceed with a government-owned core module, one or two commercial module providers will win follow-on awards, and the overall station architecture will look more like an evolved ISS than a new commercial marketplace. A skeptical capital allocator should weight commercial station revenue at a fraction of the headline forecast totals and should expect the business to stabilize at a subscale that supports one or two operators at modest profitability rather than expanding to displace the ISS as a commercial revenue source.

The Ignition Budget Math Problem

A closer reading of the NASA Ignition budget suggests that the alternative approach may be insufficiently funded to deliver what the agency describes. The $250 million per year envelope through the end of ISS operations in 2030 or 2031 provides a total Phase 1 budget of roughly $1 billion to $1.25 billion across five years. That budget must cover the Core Module, two commercial modules, and the initial ISS outfitting activities. The ISS itself cost over $100 billion to develop, assemble, and operate. A single ISS-heritage module such as the US Destiny laboratory cost approximately $1.4 billion in development and construction (adjusted for inflation). The more recent Russian Nauka module cost approximately $2 billion. Boeing’s cost estimates for new US-built ISS-class modules typically run $1 to $2 billion per module.

The NASA Ignition materials do not disclose the expected unit cost for the Core Module or the commercial modules. If the Core Module costs $1 billion to develop and build, the $1.25 billion Phase 1 budget cannot cover both the Core Module and two commercial modules. If the Core Module costs $2 billion, the Phase 1 budget cannot cover the Core Module alone. NASA has provided no published cost estimate that clarifies how the math works.

Several explanations are possible. First, the $250 million per year may be a partial allocation that will be supplemented by additional funding in subsequent budget cycles. The Phase 1 schedule extends to 2035 in the procurement timeline, with the core module downselect occurring in 2027 to 2029 and operational use through 2035. If the $250 million per year continues beyond ISS retirement, the total budget reaches $2.25 billion across nine years, which begins to approach the minimum required.

Second, the Core Module may be substantially smaller and simpler than an ISS-heritage module. The Core Module is described as providing propulsion, power, cooling, docking ports, and basic life support, which is a narrower function than a full habitation module. A stripped-down Core Module might cost $500 million to $800 million if built with commercial supply chains and aggressive scope management.

Third, the commercial module providers may fund a substantial fraction of their own development through private capital, with NASA paying for specific milestones rather than the full development cost. The Space Act Agreement structure that NASA has used in the Phase 1 CLD program works this way, with NASA paying for completed milestones rather than for total development.

Fourth, NASA may be budgeting based on extending ISS operations beyond 2030 or 2031 to create additional fiscal room. The ISS deorbit is currently scheduled for 2030 or 2031, but extensions are plausible if the commercial alternative is not ready and if the international partnership supports extension.

None of those explanations is fully reassuring. The ISS cost over $100 billion across its full lifecycle, and no party has publicly demonstrated that a functional orbital platform can be delivered at a fraction of that cost. If the Ignition approach is truly constrained to the $250 million per year envelope NASA disclosed, the approach is either substantially under-scoped or will require significant supplemental appropriations that have not been identified. A skeptical capital allocator should treat the Ignition timeline with the same discount applied to other NASA schedule commitments, and should expect meaningful delays and cost growth relative to the March 2026 announcement.

The downstream implications for the commercial station operators are important. If NASA’s Core Module is delayed, Axiom’s AxH1 docking in late 2026 becomes the only near-term LEO commercial progress. If the Core Module is canceled or substantially restructured during future budget cycles, the commercial module providers face a fundamentally different market structure that they have not priced into their capital raises. The Axiom financial distress documented above is partly a consequence of these exact uncertainties, and the distress is likely to spread to Vast, Voyager, and Blue Origin as the Ignition procurement timeline unfolds.

The ISS Deorbit Risk as a Forcing Function

The NASA Ignition approach depends critically on the International Space Station remaining operational through the commercial module integration phase. The Core Module attaches to the ISS forward port. The commercial modules are validated using ISS infrastructure, robotics, spacewalk capability, and visiting spacecraft. ISS assets are transferred to the new station before separation. The entire incremental transition architecture presupposes a functional ISS through 2030 or 2031.

NASA’s own briefing materials document the risk directly. The ISS has “overcome decades of issues, including major spacecraft failures, suit issues, medical contingencies, debris avoidance maneuvers, and more than 110 corrective spacewalks since assembly.” The Columbia, SpaceX CRS-7, and Orbital CRS-3 failures highlighted the need for redundancy and resiliency. Over 30 years of investment were required to bring ISS life support systems to their current state-of-the-art performance. The briefing states that “a quarter-century of space station success has depended on constant human intervention and a deep bench of technical expertise.” Any future station will encounter similar failures and challenges.

A specific visible risk is the persistent air leak in the Russian Zvezda service module’s PrK transfer compartment. The leak was first detected in 2019 and has gradually worsened despite multiple patches and pressurization experiments. By 2024 the leak rate was the subject of cooperative investigation between NASA and Roscosmos. A catastrophic failure of the Zvezda module is not considered likely, but the probability is non-zero and has increased over time. A serious Zvezda failure could force an accelerated ISS deorbit before the Commercial LEO transition is ready.

Other visible ISS risks include the aging solar arrays (partially replaced through IROSA upgrades but still aging), the aging Russian segment as a whole, the increasing frequency of debris avoidance maneuvers as the LEO debris environment worsens, and the supply chain fragility that comes with aging components made by suppliers that no longer exist. NASA’s 110-plus corrective spacewalks since assembly tell the story of a station that requires continuous human intervention to remain operational. That intervention continues to work, but the probability of a non-recoverable failure increases each year.

The commercial station implications of an accelerated ISS deorbit would be severe. The entire Ignition incremental transition approach assumes the ISS as a stable platform for Core Module integration, commercial module outfitting, and initial operational validation. If the ISS must be deorbited in 2028 or 2029 rather than 2030 or 2031 because of a serious component failure, the Core Module may not be complete and the commercial modules will not have completed integration. The alternative is a gap in US continuous human presence in LEO, which is the exact outcome the Ignition approach is designed to prevent. In that scenario, China’s Tiangong station would be the only operational national space station, with the geopolitical and national-prestige implications NASA describes as unacceptable.

A skeptical read of this risk is that NASA’s Ignition approach has locked the commercial station sector into a timeline that depends on ISS reliability that the agency’s own briefing materials describe as reliant on “constant human intervention.” Private capital allocators backing Axiom, Vast, Voyager, or Blue Origin are accepting exposure to a tail risk (accelerated ISS failure) that NASA itself has documented but has not priced into its procurement strategy. That is not unique to space. Many large infrastructure projects carry similar tail risks. It is worth naming explicitly because the Ignition briefing’s emphasis on leveraging ISS as a stable platform glosses over the fragility the agency has also documented.

SpaceX was awarded the US Deorbit Vehicle contract in 2024 for approximately $843 million. That vehicle is scheduled for launch in 2029 or 2030 and is designed to deorbit the ISS in a controlled manner. A contingency deorbit scenario in which the ISS experiences a catastrophic failure before the Deorbit Vehicle is ready is not part of the contracted scope. The commercial station sector and the broader US space community have quietly accepted that an emergency ISS deorbit is a tail risk that is partially mitigated by the Deorbit Vehicle schedule but not fully covered.

The “Space as a Service” Contradiction

A recurring contradiction in the commercial space station narrative deserves explicit treatment. NASA’s Ignition briefing materials describe the end-state goal as “NASA becomes one of many customers buying services” in a mature commercial LEO economy. The same materials also state that “no independently-verifiable market research” indicates economic viability of a commercial station partially funded by NASA, that “no data suggests nations, companies, or individuals beyond NASA will create sufficient demand,” and that the budget shortfall forces a single-provider, winner-take-all scenario.

Those two statements are internally inconsistent. A business cannot simultaneously scale on commercial customers and depend entirely on NASA. A market that has no demand beyond NASA is not a market; it is a government procurement program with commercial contractors. The “Space as a Service” framing that has dominated commercial space marketing since approximately 2015 is, in NASA’s own assessment, not supported by evidence.

The contradiction affects every element of the commercial station business case. If NASA is the only paying customer, then commercial module providers should be evaluated as defense contractors: their revenue depends on appropriations, their margin depends on contract structure, their growth depends on government priority shifts. That is a legitimate business, and several of the largest US defense contractors have built durable profitability on that model. It is not a “commercial space economy” business in the sense that the trillion-dollar forecasts describe.

If there are ly commercial customers, then the operators should be able to name them, disclose their contracts, and demonstrate that their revenue does not depend on NASA. No commercial station operator has publicly disclosed contract revenue from non-NASA, non-sovereign customers at scale. Axiom’s sovereign customer base (India, Poland, Hungary, Saudi Arabia, Turkey, Italy, Sweden) is the closest approximation, but sovereign space agencies are themselves government customers, and their purchasing is driven by national prestige rather than by the commercial value of the research conducted on the station.

The Ignition restructuring can be read as NASA implicitly acknowledging that the “Space as a Service” framing was aspirational rather than empirical. The government-owned Core Module approach concentrates NASA as the primary customer, assumes commercial module providers will operate as contractors, and positions the transition to true commercial operation as a long-term goal rather than a near-term reality. That framing is more honest than the previous marketing language and more aligned with the actual structure of the sector.

For capital allocators, the implication is that investments in commercial station operators should be priced as investments in defense contractors rather than as investments in platform businesses. Defense contractor multiples (typically 1.5 to 2.5 times revenue and 10 to 20 times earnings for mature companies) are substantially lower than technology platform multiples (often 5 to 20 times revenue and 30 to 100 times earnings). The $2 billion Axiom valuation from March 2025 appears reasonable against defense contractor multiples applied to its limited revenue base. It appears extremely high against any measure of commercial platform economics. The distinction matters for sizing position allocations in the sector.

In-Space Manufacturing Remains a Laboratory Exercise

Varda Space Industries has completed six orbital missions as of April 2026, most recently W-6 launched on March 30, 2026. The company has raised approximately $329 million in total venture funding. Varda’s government business in hypersonic reentry testbed work has emerged as a meaningful revenue contributor alongside the pharmaceutical crystallization program. The company has not disclosed the per-mission economics that would allow an outside observer to judge whether the returned pharmaceutical products could be sold at a price that recoups mission costs.

NASA’s own assessment is that 25 years of ISS commercial use has produced no scaled product manufacturing. Varda’s CEO Will Bruey has predicted that within 10 years multiple specialized spacecraft will return to Earth daily carrying pharmaceuticals manufactured in space. Those forward statements have historically not been validated by the operational record of the sector. The combination of Varda’s hypersonic pivot, Space Forge’s semiconductor wafer demonstrations, and Redwire’s non-disclosure of Made in Space revenue contributions points to a sector that continues to exist primarily on the promise of future breakthroughs rather than on demonstrated commercial traction.

The pharmaceutical crystallization thesis, which has anchored most in-space manufacturing pitches, depends on demonstrating that a specific molecule can be crystallized more effectively in microgravity than on Earth, that the resulting product has measurably superior pharmacological properties, that those properties justify a price premium, and that the price premium covers the substantial incremental cost of orbital production. No operator has publicly disclosed a specific pharmaceutical product that meets all four conditions. The sector’s business model continues to depend on a future demonstration that has not yet occurred, and that demonstration has been anticipated for more than a decade without materializing.

Semiconductor manufacturing in space, pioneered conceptually by Space Forge and others, faces similar challenges. The economics require demonstrating a specific defect-reduction or performance improvement that justifies a price premium sufficient to cover orbital costs. Terrestrial semiconductor manufacturing has continued to improve in ways that raise the bar for orbital competitors. The window for orbital semiconductor manufacturing to close a meaningful competitive gap against terrestrial facilities appears to be narrowing rather than widening.

Debris Is the Industry’s Unpriced Liability

The US Space Surveillance Network tracks roughly 40,000 objects larger than 10 centimeters in orbit as of early 2026. The European Space Agency estimates more than 1.1 million objects larger than 1 centimeter and more than 130 million objects larger than 1 millimeter. The Intelsat 33e breakup on October 19, 2024 broke into more than 700 tracked pieces uninsured.

The insurance market functions as a leading indicator for the commercial space sector’s risk profile. The space insurance market absorbed approximately $2 billion in losses across 2023 and 2024 against an annual premium base of $500 million to $580 million. Insurers are professional risk pricers, and their behavior in 2023 and 2024 (retreat, rate increases of 50 to 150 percent) followed by tentative return in 2025 and abrupt reversal after SpainSat in January 2026 maps closely to where real risk is accumulating. Launch coverage in 2026 costs 10 to 20 percent of insured value, while in-orbit premiums range between 1 and 2 percent. Those rates reflect a sector whose operational risk has materially increased since 2020. The forecasts that assume a benign operational environment are inconsistent with the pricing that insurance professionals are applying to the same sector.

A 2024 ESA analysis suggested that the probability of at least one catastrophic collision in LEO per decade has moved from below 10 percent in the 2010s to above 30 percent in the 2020s. That is a substantial risk for an industry that does not carry it on any balance sheet. The growth of mega-constellations from Starlink, Amazon Leo, and prospective Chinese systems will add tens of thousands of additional satellites to the operating population over the next five years. The current regulatory framework assigns minimal financial liability to operators for debris-generating events, which means that the costs of cascading collisions would fall primarily on the insurance market and on the operational space economy rather than on the operators whose satellites caused the damage.

The Kessler Syndrome scenario, in which debris density reaches a threshold that triggers self-sustaining collision cascades, remains a probabilistic tail risk rather than an imminent certainty. Specific orbital bands, particularly around 800 to 1,000 kilometers where many older satellites operate, are closer to the threshold than others. A Kessler cascade in a specific band would not end commercial space operations but would make specific orbital altitudes uneconomical for decades. The forecasts do not price this risk in any transparent way. Insurance markets price it implicitly through premium rates that have risen substantially over the past three years.

Active Debris Removal operators such as Astroscale and ClearSpace are developing commercial services that depend on government contracts rather than on commercial customer demand. ESA’s PRELUDE mission and Japan’s JAXA ADRAS-J missions have validated specific capture and deorbit technologies, but the commercial business model depends on regulatory mandates that do not yet exist in most jurisdictions. The US, EU, UK, and Japan have each begun developing regulatory frameworks that would require operators to fund debris removal for specific classes of spacecraft, but the frameworks are not yet in force. When they come into force, they will create a new revenue category for ADR operators and a new cost category for constellation operators. Neither effect is fully priced in the current forecasts.

Asteroid Mining Is a Financing Story

Planetary Resources shut down in late 2018 after raising roughly $50 million. AstroForge, founded in 2022, launched its Odin mission on February 26, 2025 as a secondary payload on Intuitive Machines IM-2. The spacecraft was officially declared lost on March 6, 2025. AstroForge’s third mission, DeepSpace-2, is planned for the second half of 2026. The total capital deployed into commercial asteroid mining since 2010 is probably under $250 million, less than the cost of a single NASA planetary science mission.

The business case for asteroid mining requires demonstrating that extraction of specific materials (typically platinum group metals or water) is economically superior to terrestrial mining and to launching the same materials from Earth. Platinum group metal extraction faces the challenge that terrestrial supply is not currently constrained enough to justify orbital extraction costs. Water extraction for in-space propellant production faces the challenge that the commercial propellant market in space is essentially zero, with NASA being the only realistic customer through the Moon Base plan and potential future Mars missions. Both business cases remain theoretical rather than operational, and the sector’s total venture capital deployment reflects investor understanding that the business is speculative even by space-economy standards.

The Lunar Economy Is a Recursive Funding Loop

Forecasts for the “lunar economy” range from $150 billion to $1 trillion by 2040 depending on the source, with assumptions about commercial lunar tourism, water ice extraction, helium-3 mining, and industrial activity that have not been validated at any scale. The actual revenue flowing through the commercial lunar sector in 2025 was approximately $200 to $300 million, essentially all of it paid by NASA to commercial contractors through the Commercial Lunar Payload Services program and related lunar delivery contracts.

The CLPS program awarded Firefly Aerospace $93.3 million for the Blue Ghost Mission 1 soft landing on March 2, 2025. Firefly received a fourth CLPS task order in July 2025 worth $176.7 million for a 2029 mission to Haworth Crater. Intuitive Machines received a $180.4 million CLPS award in March 2026 for subsequent lunar delivery work. Astrobotic, Draper, and other CLPS participants have received similar awards. Each of those contracts represents NASA paying a commercial company to deliver NASA payloads to the Moon. The “commercial” designation describes the contract structure rather than the customer base.

The NASA Moon Base plan announced at Ignition targets up to 25 missions in Phase 1 through 2029, up to 24 landings in Phase 2 through 2032 delivering 60 tons of cargo, and up to 38 tons of cargo per year in Phase 3 from 2032 onward. Each of those missions represents NASA procurement of commercial services. Intuitive Machines’ 2025 SEC filings disclose that the company’s revenue depends primarily on a small number of government contracts and that the failure of the market for commercial spaceflight to achieve the growth potential we expect is a material risk. That is the commercial lunar service provider acknowledging in its own SEC filings that the commercial market may not emerge at the scale its business model assumes.

The recursive structure is clear. NASA funds commercial lunar landers through CLPS. The lunar landers deliver NASA payloads to the Moon. NASA counts the resulting activity as part of the “lunar economy.” The lunar economy grows to match NASA’s appropriation cycle. A reduction in NASA lunar funding would produce a proportional reduction in the “commercial” lunar economy. The forecasters who project $150 billion or $1 trillion by 2040 assume that independent commercial demand for lunar services will emerge at some point during the forecast horizon. The assumption is not backed by any specific customer, contract, or business case that has been disclosed publicly. In the absence of that validation, the lunar economy is better understood as a line item within the NASA budget that flows through commercial contractors rather than as an independent market.

The specific lunar commercial categories that would validate an independent commercial lunar economy include lunar tourism (no operator, no customer contracts), lunar telecommunications (early pilot projects only, no operational service), lunar resource extraction (no operator at commercial scale, no buyer), and lunar manufacturing (no operator, no product market). Each of those categories depends on operational infrastructure that does not yet exist and on customer demand that has not been demonstrated. The forecasts that project substantial revenue from these categories by 2040 are making assumptions about technological readiness and market emergence that the current evidence does not support.

Commercial Space Stations After the Ignition Announcement

The treatment above describes the March 24, 2026 Ignition restructuring. The financial implications for the four competing station programs are substantial. Axiom Space has faced financial difficulties and leadership changes as discussed previously. Vast Space raised $500 million in 2025. Starlab has received $217 million or more through Phase 1 CLD. Orbital Reef faces the slowest visible progress. The Ignition framework’s preference for a government-owned Core Module with docking commercial modules compresses the addressable market for independent free-flying stations and concentrates remaining opportunity around the Core Module provider and a small number of commercial module providers.

Defense Demand Is Real but Politically Contingent

Defense and intelligence spending has emerged as the dominant driver of space economy growth in the mid-2020s. Novaspace reported $138 billion in total government space spending in 2025. The US Department of Defense’s space budget, distributed across the Space Force, the National Reconnaissance Office, the Space Development Agency, and service-specific space programs, has grown at a compound rate substantially above the overall defense budget. The Golden Dome missile defense program commits the United States to a multi-layered space-based interceptor architecture that could absorb tens of billions of dollars in additional spending.

The commercial space companies benefiting most from this spending are the prime contractors (Lockheed Martin, Northrop Grumman, Boeing, L3Harris) and a smaller group of newer entrants (Anduril, Planet Labs, BlackSky, Rocket Lab through its Geost acquisition). Their earnings track federal appropriations cycles rather than commercial product markets. Defense budgets are subject to political volatility, priority shifts, and competing demands from non-space military programs.

What Would Change the Skeptical Thesis

A rigorous skeptical analysis names the observable conditions under which its conclusions would change. For the space economy, several specific 2026 to 2028 events would materially strengthen the trillion-dollar thesis and warrant revision of the skeptical read presented here.

First, a successful Starship Propellant Transfer Demonstration and a subsequent Starship HLS crewed lunar landing on schedule would validate the cost and cadence assumptions that underlie most commercial space thesis projections. The critical observable is not a single successful flight but a sustained flight rate of 20 or more successful Starship missions per year at low unit cost, and a Falcon 9 price trajectory that starts falling again rather than continuing to rise.

Second, direct-to-cell service reaching 50 million or more subscribers globally with voice capability and average revenue per user above $100 per year would validate a meaningful fraction of the broadband constellation thesis. The observable is T-Mobile, AT&T, Verizon, and their international analogues reporting satellite revenue as a material segment rather than a marketing line.

Third, an in-space manufacturing operator disclosing recurring commercial revenue from pharmaceutical or specialty materials sales at a scale exceeding $100 million annually by 2028 would validate the sector’s commercial case. The observable is audited financial disclosures with product names, customer names, and unit pricing.

Fourth, a commercial space station operator signing and executing a contract for recurring non-NASA, non-sovereign customer revenue at a scale exceeding $200 million annually by 2030 would validate the station demand curve that the forecasts assume. The observable is a named customer, a published price, and a flight manifest.

Fifth, China’s Guowang constellation reaching 3,000 or more operational satellites by 2030 without significant competitive response from Starlink and Amazon Leo would validate the state-directed model as a competitive alternative and would reshape the global broadband supply picture.

Sixth, the SpaceX IPO pricing at or above $1.5 trillion in its first three months of public trading, with institutional investor participation rather than predominantly retail demand, would validate the market’s willingness to price the space economy at the aggressive multiples the forecasts imply.

Seventh, NASA’s Ignition Core Module procurement producing a downselect with real hardware commitments from a commercial provider at a price NASA can actually afford, followed by successful on-orbit assembly by 2030, would validate the incremental transition approach and partially rehabilitate the commercial space station thesis.

Eighth, Space Reactor-1 Freedom launching on schedule in 2028 and operating successfully at Mars would validate NASA’s nuclear propulsion thesis and change deep-space economics substantially.

Ninth, Axiom Space reaching operating cash-flow profitability on its PAM business and successfully docking AxH1 with the ISS on schedule in late 2026 would validate the most operationally mature commercial station operator’s business model.

Tenth, Firefly Aerospace sustaining its stock price above $45 through the end of 2026 would validate the disciplined space IPO template as a durable structure for future listings.

None of those observables are guaranteed. Each is testable on a defined timeline. If several are met, the skeptical thesis should be revised. If few or none are met, the case for a discounted reading of the headline forecasts will have strengthened.

Summary

The skeptical perspective on the space economy does not require concluding that the sector is failing. It requires recognizing that the sector’s headline forecasts rest on assumptions that the operating reality of the industry, and NASA’s own assessment, have not validated. Starlink is a remarkable business whose success masks the weakness of every other segment’s commercial traction, and whose margin depends substantially on transfer pricing advantages that no competitor can match. SPAC-era NewSpace companies remain substantially underwater. Space tourism has produced a pipeline of news events rather than a pipeline of customers. Artemis has delivered hardware at a cost that NASA itself describes as unsustainable. In-space manufacturing, debris management, asteroid mining, and commercial space stations each sit in the “technically demonstrated, commercially unvalidated” category that NASA’s own March 2026 assessment confirms.

The March 2026 Ignition restructuring is the single most important regulatory event for the commercial space sector in the past decade. NASA publicly acknowledged that 25 years of ISS commercial use has not produced breakthrough products, that tourism has not materialized as a market, that there is no independently verifiable evidence of commercial station economic viability, and that NASA’s own budget cannot afford the independent commercial station plan that was previously the policy framework. The Ignition response, a government-owned Core Module attached to the ISS with commercial modules docking to it, is a pragmatic accommodation of those findings. It is also an explicit rejection of the commercial space economy narrative that has been the framing device for the sector since at least 2020. The $250 million per year budget envelope NASA disclosed is almost certainly insufficient to deliver the Core Module and two commercial modules on the proposed timeline, which suggests additional supplemental appropriations or substantial schedule slippage ahead.

The launch market’s pricing inversion is the second most important development for the 2026 outlook. Falcon 9 pricing increased to $74 million per dedicated launch in 2026, up from $70 million, with rideshare rates now at $7,000 per kilogram. NASA’s own assessment that transportation costs are increasing rather than decreasing contradicts the trillion-dollar forecasts directly. SpaceX’s transfer-pricing advantage of approximately $55 million per launch, compounded across 165 Falcon 9 missions in 2025, produced roughly $9.5 billion in implicit subsidy to Starlink, an amount comparable to Starlink’s full-year 2025 EBITDA. No competitor has access to that advantage, and the commercial broadband constellation economics look fundamentally weaker without it.

Axiom Space, the most operationally mature commercial station company, has cycled through three CEOs in 14 months, completed approximately 100 layoffs, accepted a 23 percent down round in March 2025, and now operates with sovereign customer revenue paying the bills round to round. That trajectory is the clearest available evidence that commercial station economics do not work at current prices, and it forms the empirical foundation for NASA’s Ignition restructuring. The Firefly Aerospace IPO, which traded down 78 percent from its August 2025 peak to a November 2025 low, is a parallel signal that even the disciplined IPO template faces immediate valuation pressure in the public market.

Taking the five-year retrospective seriously, the 2025 space economy figure of $626.4 billion fell 31 percent short of the trajectory implied by 2020-era projections. The composition miss was more severe than the total miss, with broadband constellations (concentrated in one company) substituting for every category that missed. A 30 to 50 percent discount to current forecasts produces 2030 and 2035 estimates that are more consistent with the realized trajectory than the headline numbers are. That is a meaningful market but not the trillion-dollar storyline that drives capital allocation decisions.

The ISS deorbit risk, which could accelerate from the planned 2030 or 2031 timeframe to as early as 2028 or 2029 if a serious component failure occurs, is a tail risk that the Ignition approach has not priced. NASA’s own briefing documents the ISS’s fragility explicitly while simultaneously building the Ignition timeline around ISS reliability. A premature ISS deorbit would collapse the entire incremental transition approach.

The SpaceX IPO expected in 2026, if it arrives at the $1.75 trillion target, will be the first event that forces the public market to explicitly price the concentration risk, the transfer-pricing advantage, and the regulatory exposure that currently sit buried inside space economy forecasts. The aftermarket performance will be the single most important data point for the sector’s 2027 and 2028 outlook. Firefly’s post-IPO drawdown of 78 percent before recovery suggests that the SpaceX listing will also face material near-term pressure, even at the headline valuation.

The data is saying that the space economy is a smaller, more concentrated, more politically dependent, more geopolitically contested, and more regulatorily constrained business than its public presentation suggests. It is still worth participating in, carefully and with full awareness of concentration risk. The companies that will prosper in the next decade are those whose business plans assume the conservative reading of the data rather than those that anchor to the most aggressive forecast. NASA, the single most important customer in the sector, has now explicitly adopted the conservative reading. Private capital allocators would do well to follow.

Appendix: Useful Books Available on Amazon

• A City on Mars
• The Space Barons
• When the Heavens Went on Sale
• Liftoff
• Reentry
• The End of Astronauts
• The High Frontier
• Test Gods
• Challenger
• Escaping Gravity

Appendix: Top Questions Answered in This Article

How large is the space economy today compared to the headline forecasts?

Novaspace valued the global space economy at $626.4 billion in 2025 and projects $1.01 trillion by 2034. McKinsey projects $1.8 trillion by 2035. The 2020 Bank of America projection of $1.4 trillion by 2030 implied a 2025 figure of approximately $912 billion, missing the actual by 31 percent. The core manufacturing, launch, and satellite operations segment sits at approximately $236 billion in 2025.

What share of SpaceX revenue comes from Starlink?

Starlink generated $11.4 billion of SpaceX’s roughly $16 billion in 2025 revenue. That represents about 61 percent of total SpaceX revenue and the majority of EBITDA. The xAI subsidiary burned $9.5 billion through Q3 2025 against just $210 million in revenue.

What is Falcon 9 pricing in 2026?

Falcon 9 is priced at $74 million per dedicated launch as of February 2026, up from $70 million the previous year. Rideshare rates increased to $7,000 per kilogram. The pricing trajectory that earlier forecasts assumed, with launch costs falling toward $20 million or $30 million by 2030, has reversed.

How significant is SpaceX’s transfer pricing advantage?

SpaceX’s estimated marginal cost per Falcon 9 mission is approximately $15 million, against a list price of $74 million. Across the 165 Falcon 9 missions flown in 2025, the majority carrying Starlink payloads, this creates approximately $9.5 billion in implicit subsidy from the launch business to the constellation business, roughly equivalent to Starlink’s full-year 2025 EBITDA.

Did any NewSpace SPAC issuer perform well after listing?

Most did not. Virgin Orbit filed Chapter 11 in April 2023. Astra Space went private at $0.50 per share in 2024. Virgin Galactic traded below $3 in early 2026 with a going-concern disclosure. Rocket Lab and Intuitive Machines are two exceptions that have executed against operational milestones.

How did Firefly Aerospace perform after its August 2025 IPO?

Firefly priced at $45 and closed its debut at $60.35 for an $8.5 billion valuation. It reached an all-time high of $73.80 on August 7, 2025 before falling to an all-time low of $16.00 on November 21, 2025, a 78 percent drawdown. It recovered to approximately $43.60 in mid-April 2026, still below the IPO closing-day level.

What is Axiom Space’s financial condition?

Axiom completed approximately 100 layoffs and 20 percent voluntary pay cuts in 2024. Its March 2025 funding round valued the company at $2 billion, a 23 percent down round from the $2.6 billion 2023 Series C. The company has cycled through three CEOs in 14 months. A December 2025 $100 million commitment from Hungary’s 4iG provided additional runway.

How did NASA’s March 2026 Ignition announcement change the commercial space station picture?

NASA proposed a government-owned Core Module attached to the ISS forward port with commercial modules docking to it. NASA explicitly acknowledged that 25 years of ISS commercial use has produced no breakthrough products, that tourism has not materialized, and that there is no independently verifiable market research for commercial station viability. The approach replaces the prior Phase 2 plan that would have funded independent commercial stations.

Is the Ignition budget sufficient?

The $250 million per year envelope NASA disclosed provides a total Phase 1 budget of approximately $1 to $1.25 billion over five years. ISS-heritage modules typically cost $1 to $2 billion each. The Ignition budget is almost certainly insufficient to deliver the Core Module and two commercial modules on the proposed timeline without substantial supplemental appropriations.

What is the ISS deorbit risk?

The ISS is scheduled for deorbit in 2030 or 2031, but the Russian Zvezda module has a persistent air leak that has gradually worsened since 2019. Other aging components increase the probability of serious failure. An accelerated deorbit before the Ignition Core Module is operational would produce a gap in US crewed LEO presence and collapse the incremental transition approach.

How many satellites did China launch in 2025?

China conducted 90 orbital launches in 2025, up from 68 in 2024. The Guowang constellation reached 136 satellites across 17 launch batches by December 2025.

How many people have flown as space tourists since 2021?

Fewer than 100 individuals have flown as paying or sponsored passengers between 2021 and early 2026. NASA’s March 2026 assessment states that tourism has not materialized as a market.

How much does a single SLS and Orion launch cost?

Each Space Launch System flight costs an estimated $4.1 billion per the NASA OIG. Each Orion spacecraft costs roughly $2.7 billion. Total Artemis program spending reached approximately $93 billion to $105 billion through fiscal year 2025. NASA’s own Ignition materials acknowledge that “more than $100 billion” has been invested over nearly 20 years.

What is NASA’s revised Artemis schedule?

Artemis II completed its lunar flyby in April 2026. Artemis III in 2027 is now a crew-lander integration test in Earth orbit. Artemis IV in early 2028 will be the first lunar surface landing. Artemis V targets late 2028. After Artemis V, NASA targets landings every six months using commercially procured and reusable hardware.

What is Space Reactor-1 Freedom?

SR-1 Freedom is NASA’s announced nuclear-powered interplanetary spacecraft targeting a Mars launch before the end of 2028. It will demonstrate nuclear electric propulsion and deploy an Ingenuity-class helicopter payload called Skyfall. The mission is intended to set regulatory precedent for nuclear hardware in space.

How much does the US government spend on space annually?

Total US government space-related spending reaches approximately $60 to $70 billion annually as of FY2026, including approximately $29 billion for the Space Force, $15 to $20 billion for the NRO (estimated), $3 to $4 billion for the SDA, $27 billion for NASA, and additional amounts across NGA, DARPA, MDA, and other agencies. That represents approximately 14 to 15 percent of the global space economy headline figure.

How is the lunar economy structured?

The “lunar economy” in 2025 consisted of approximately $200 to $300 million in NASA payments to commercial contractors through CLPS and related programs. NASA’s Moon Base plan targets 25 missions through 2029, 24 landings through 2032, and 38 tons of cargo annually from 2032 onward. The revenue flowing through the “commercial” lunar sector depends directly on NASA appropriations.

Why is spectrum a structural constraint?

Satellite broadband operators depend on spectrum licenses granted through ITU and enforced by national regulators. SpaceX paid $17 billion to acquire EchoStar’s 2 GHz spectrum in 2025, a transaction that reflects scarcity. Competitors including Amazon Leo, Eutelsat OneWeb, and AST SpaceMobile face varying spectrum constraints that limit their practical capacity expansion.

How has the insurance market behaved as a leading indicator?

The space insurance market absorbed approximately $2 billion in losses across 2023 and 2024. Rates increased 50 to 150 percent. The market began to recover in 2025 before the January 2026 SpainSat loss reversed that trend. Launch coverage costs 10 to 20 percent of insured value, in-orbit premiums 1 to 2 percent.

What happened to Earth observation commercial revenue?

The Earth observation market was $7.04 billion in 2025 per Fortune Business Insights. Maxar led with 21.3 percent share in 2024. Commoditization through Chinese satellite deployment and open-source government data has driven pricing pressure. The 2020-era forecasts that projected $10 to $20 billion in commercial EO revenue by 2030 have not been validated.

What observable events would validate the trillion-dollar thesis?

Successful Starship orbital refueling with sustained cadence above 20 launches per year, direct-to-cell reaching 50 million subscribers with voice capability, in-space manufacturing revenue above $100 million annually, commercial space station non-NASA contracts above $200 million annually, Guowang reaching 3,000 satellites by 2030, SpaceX IPO pricing at or above $1.5 trillion with institutional participation, successful Ignition Core Module procurement, SR-1 Freedom launch on schedule, Axiom reaching operating cash-flow profitability, and Firefly sustaining above $45 through end of 2026.

Appendix: Glossary of Key Terms

Space Economy

A term of art describing total economic activity attributable to space-based systems. Different sources use different definitions, with broader definitions including downstream applications of satellite data and narrower definitions limited to hardware manufacturing, launch, and satellite operations.

Special Purpose Acquisition Company (SPAC)

A publicly listed shell company raised specifically to merge with a private business. SPACs were used extensively between 2020 and 2022 to list NewSpace companies.

Low Earth Orbit (LEO)

An altitude band between approximately 160 kilometers and 2,000 kilometers above Earth’s surface, where the majority of operational satellites now reside.

Geostationary Orbit (GEO)

An orbital altitude of approximately 35,786 kilometers above the equator, where satellites appear stationary relative to points on Earth’s surface.

Kessler Syndrome

A theoretical cascade scenario in which the density of objects in a given orbital band reaches a threshold at which collisions generate enough new debris to trigger further collisions.

Ignition Initiative

A March 24, 2026 NASA realignment that restructured the commercial space station strategy, introduced a government-owned Core Module attached to the ISS, reset the Artemis program, announced the Moon Base plan, and committed to nuclear propulsion with Space Reactor-1 Freedom. Responded to Executive Order 14369.

Core Module

A NASA-procured government-owned module under the Ignition approach, providing propulsion, power, cooling, docking ports, and basic life support. Attaches to the ISS forward port.

Transfer Pricing

The internal price at which a vertically integrated company values transactions between its business units. SpaceX’s Starlink launches occur at approximately $15 million marginal cost internally against a $74 million commercial list price.

In-Space Manufacturing

The production of materials or products in the microgravity environment of low Earth orbit. NASA’s March 2026 assessment states that 25 years of ISS commercial use has produced no scaled product manufacturing.

Chapter 11

A chapter of the US Bankruptcy Code permitting reorganization under court supervision. Virgin Orbit filed Chapter 11 in April 2023.

EBITDA

Earnings before interest, taxes, depreciation, and amortization. Starlink reported $7.2 billion on revenue of $11.4 billion in 2025.

Active Debris Removal (ADR)

A category of space missions that rendezvous with and remove pieces of orbital debris. ClearSpace-1 and Astroscale’s ELSA missions are the most visible examples.

Guowang

A Chinese state-backed broadband megaconstellation with a planned 13,000 satellites. Reached 136 satellites across 17 launch batches by December 2025.

Direct-to-Cell

A satellite-to-smartphone communications service. Starlink launched T-Satellite with T-Mobile on July 23, 2025, and has more than 650 satellites in orbit by April 2026.

Price-to-Sales Multiple

A valuation metric. SpaceX’s targeted 2026 IPO valuation of $1.75 trillion against 2025 revenue of approximately $16 billion implies a multiple of roughly 109 times.

Commercial Lunar Payload Services (CLPS)

A NASA program contracting commercial lunar landers to deliver NASA payloads. Firefly Aerospace’s Blue Ghost Mission 1 became the first fully successful commercial lunar landing on March 2, 2025.

Space Reactor-1 Freedom

A NASA nuclear electric propulsion demonstrator announced at Ignition, targeting Mars before the end of 2028 and intended to set regulatory and launch precedent for future fission power systems.

Private Astronaut Mission (PAM)

A commercial crewed mission to the International Space Station procured through NASA’s partnership framework. Axiom has flown four PAMs since 2022. Under Ignition, NASA plans two PAMs per year with potential commander seat sales.

Down Round

A financing round at a lower valuation than the company’s prior round. Axiom Space’s March 2025 round at $2 billion represented a 23 percent down round from its 2023 Series C at $2.6 billion.