The International Space Station is in its last full operating years. NASA plans to retire it at the end of 2030 and has contracted SpaceX, for a potential $843 million excluding launch, to build the US Deorbit Vehicle that will bring its 420-tonne structure down over the remote Pacific. Whether the United States and its partners keep a permanent human presence in low Earth orbit after that date depends on commercial stations that do not yet exist. In October 2026 no commercial station has flown. The one closest to launch, Vast's Haven-1, is a single-module demonstrator now targeted for no earlier than the first quarter of 2027, and it is designed to host four short crew visits over a three-year life rather than continuous occupation.
This study assesses whether commercial space stations are feasible as infrastructure: technically, commercially and institutionally. It covers the NASA transition and its procurement, the demand for station time beyond NASA, the competing programmes, the technology, the capital and delivery model, the cost of capacity, the risks, the regulatory frame and the environmental and governance issues, and it closes with a decision framework for the stakeholders who will fund, use, supply or regulate these stations. Figures are as of 20 October 2026 and are sourced in the text. Where a figure is a Setcoin assessment (readiness scores, the risk register, the demand and coverage scenarios), it is labelled as such.
The verdict in brief
Commercial space stations are technically feasible, institutionally necessary and commercially unproven.
The technology largely exists. Pressurised modules, life support, power, thermal control, communications and docking have more than two decades of flight heritage on the ISS, and before it on Mir, Skylab and Salyut. The open questions are integration at a much lower cost point, qualification within a compressed schedule, and dependence on a small number of launch vehicles. None of these is a question of physics.
The institutional case is strong. NASA, its international partners and a growing list of sovereign agencies need somewhere to fly astronauts and run experiments after 2030. China's Tiangong will be operating whatever the West decides, and it is opening to foreign astronauts. The US Congress has written into its appropriations report that it expects at least one commercial destination to be available when ISS operations end.
The commercial case is the weak link. Demand beyond government is real but thin. Private astronaut missions, sovereign astronaut programmes, microgravity research, in-space manufacturing and media each exist, and each has paying customers, but none yet has the depth to fill a station. In most credible scenarios NASA and other agencies remain the majority customer for the first decade of commercial operations. The commercial station market of the 2030s is better described as a government services market with a commercial tail than as a commercial market with a government anchor.
Ten findings
- The deadline is fixed and close. ISS retirement at the end of 2030 leaves about four years to launch, commission and certify at least one replacement. Every programme has already slipped by one to three years against its first public target, and the deadline they are slipping towards has not moved.
- NASA's procurement has changed shape four times in fourteen months: a directive in August 2025, a pause in January 2026, a government-owned core module proposed in March 2026 and withdrawn on 1 June, and a final request for proposals on 9 October 2026. Proposals are due on 8 December 2026 and awards are expected in spring 2027. Each change consumed schedule that the deadline does not return.
- The final procurement is a firm-fixed-price, multi-award contract under the Federal Acquisition Regulation: early development contracts to at least two providers, then a competition among them for final design, test, certification and services. NASA is buying a service, not a station, and providers bear the cost of overruns.
- NASA's direct development funding to date is under $600 million across all providers, against an estimated $2–4 billion to build each station. Private capital, more than $3 billion so far, is carrying most of the development cost: about five dollars of private money for every NASA dollar. On commercial crew the ratio ran the other way.
- Capital raised in 2026 alone exceeds $1 billion. Axiom's round closed at more than $525 million, with investors including Type One Ventures, the Qatar Investment Authority and MUFG, and Vast raised $500 million in March ($300 million of equity and $200 million of debt). Debt has entered the capital stack for the first time.
- Starlab passed its commercial critical design review with NASA in February 2026 and is designed to launch whole, on a single Starship, in 2029. It is the furthest advanced on design maturity at system level, and the most exposed to a single launch vehicle.
- Axiom has re-sequenced at NASA's request. Its first module, a payload, power and thermal module with eight research racks, is targeted for early 2028 at the ISS, followed by a habitat in 2029; the two then fly free as a four-person station before the ISS retires.
- Orbital Reef, the Blue Origin and Sierra Space concept, has no current launch date, and reporting indicates the partners are weighing the future of the partnership as both prioritise other programmes.
- Demand beyond NASA exists, but at today's cadence it fills little of a station. Setcoin's demand model fills about 28% of a four-person station's 1,460 annual crew-days with two NASA increments, two private missions and two sovereign missions a year. Covering operating costs needs either continuous NASA presence or several non-NASA segments growing at once.
- China's Tiangong, permanently crewed since 2022, is expected to host its first foreign astronaut, a Pakistani payload specialist, in late 2026, and on 22 June 2026 China announced an expansion to a six-module station of about 180 tonnes. A gap in Western presence after 2030 would leave Tiangong as the only permanently crewed station in orbit.
Six contrarian findings
The study challenges six common narratives about the commercial station market.
- "NASA's anchor tenancy de-risks the business." It does not. At the outline's estimate of $40–50 million a year per station, NASA covers 20–30% of operating cost. The anchor is necessary for certification and credibility but insufficient for viability: the study calls this the government anchor tenant mirage.
- "Four stations are competing for a growing market." The market, at NASA's revised requirement of two four-person, 30-day increments a year, supports one, possibly two, NASA-funded stations. The competition is for a small number of slots, and the likeliest outcome is consolidation.
- "Starship makes stations cheap." It lowers the cost of launching modules and supplies sharply, but most of the cost of using a station is crew transport, priced by NASA at about $325 million for a four-seat flight in 2030, and qualification and certification, which do not fall with launch prices.
- "Tourism will fill the gap." Private astronaut missions have flown at about one a year to the ISS. Even at three times that cadence they fill a small fraction of a station's crew-days, and most of each mission's price pays for transport.
- "In-space manufacturing needs crewed stations." The leading in-space manufacturing company, Varda, flies uncrewed capsules: it has completed six launch and re-entry missions and raised $598 million in total. Some of the manufacturing demand once attributed to stations may never need a crew.
- "The ISS can simply be extended." Its age, a long-running leak in the Russian segment and a contracted deorbit make an extension unlikely and an early retirement more plausible than a late one.
What would change the verdict
Three developments would move the commercial case from conditional to credible. The first is a NASA services contract funded at a level that covers a meaningful share of operating cost for at least two providers: the House Appropriations Committee's proposal of at least $400 million for commercial LEO development in fiscal 2027, against a request of $299.7 million, points in that direction but is not yet enacted. The second is Starship flying routinely, which gates Starlab and Vast's Haven-2 core module and lowers the cost of building and supplying every station. The third is one non-government segment, most plausibly sovereign astronaut missions or pharmaceutical research, contracting for multi-year capacity before launch. The study tracks each as a signal in its final chapter.
How to read this study
Chapters 2 and 6 cover the NASA transition and procurement in detail; chapters 3 and 8 the demand and the economics of operating a station; chapters 4 and 5 the competing programmes and their technology; chapter 7 the geopolitical dimension; chapters 9 to 11 the capital, the delivery model and the cost of capacity; chapters 12 to 14 risk, regulation and sustainability; chapter 15 the decision framework. Interactive models let the reader test schedule slips, demand, operating economics, funding shares and programme readiness against their own assumptions, save scenarios, share them by link and download the Excel versions.