NASA's commercial crew program is the agency's lowest-cost human spacecraft development effort in nearly 60 years.
According to data reported by NASA and analyzed by The Planetary Society, the space agency spent $6.2 billion ($6.7 billion adjusted for inflation to 2019 dollars) on commercial crew since fiscal year 2011. That $6.2 billion resulted in 2 new human-capable spacecraft: the Crew Dragon and the Starliner.
In NASA's history, the only other crewed spacecraft project to have spent a comparable amount of money during its development (after adjusting for inflation) was Mercury between 1959 and 1961:
|Spacecraft Development Costs (NASA portion, inflation adj)|
|Apollo CSM||$30.9 billion|
|Shuttle Orbiter||$27.4 billion|
|Crew Dragon||$1.7 billion|
Inflation adjusted using NASA's New Start Index. Spacecraft development costs are approximations and include only direct costs through the fiscal year of their first test flight. Orion assumes a crewed test flight in FY 2023. Raw data, methods, and sources are in this Google Sheet.
In other words, if Dragon and Starliner work as intended, they could be some of the best deals in NASA's history.
The public-private partnership approach to developing new crew vehicles, in which NASA committed a fixed sum of money while taking a more hands-off approach to management, was a risk. When proposed in 2010, companies like SpaceX were only just beginning to demonstrate their capability. The model had never been attempted on something as complex as human spaceflight.
But NASA officials argued that a new approach was needed: previous efforts to develop new human spacecraft were plagued by cost overruns. Fixed-price commercial partnerships, it was argued, would incentivize partners to reduce costs and provide them the means to kickstart a new market for human access to low-Earth orbit.
While a new market for human spaceflight has yet to materialize, the public-private model resulted in significant cost savings compared to other human spaceflight programs. Remarkably so.
|Per-Seat Cost (millions, inflation adj)|
|Apollo (to LEO)||$390 million|
|Shuttle Orbiter||$170 million|
|Crew Dragon||$60 - $67 million|
|Starliner||$91 - $99 million|
Per-seat costs are calculated using the total costs of program operations divided by the number of astronauts flown during that period. Shuttle costs are calculated from the final 6 years of the program. Apollo uses the nominal costs of Apollo 7 as its baseline. Starliner and Crew Dragon per-seat costs use the total contract value for operations divided by the maximum 24 seats available. The upper range reflects the inclusion of NASA's program overhead. Soyuz costs are from the seat NASA purchased in 2020. Orion costs use estimated program costs in 2024 and 2025 assuming one flight of 4 astronauts per year. Raw data, methods, and sources are in this Google Sheet.
Despite the stunning difference in price tags, keep in mind that making direct comparisons between NASA's human spacecraft projects is fraught with caveats.
Obviously: Apollo. It was designed to go the Moon, not just low-Earth orbit. The Space Shuttle was the first experiment in reusability, launched massive amounts of cargo, and could carry a 7-person crew into space. Orion is designed for deep space missions that last weeks. Mercury and Gemini provide the closest comparisons, but they were experimental, transitory programs at the dawn of human spaceflight.
Commercial crew should be more affordable than other NASA programs because its ambitions are far more modest: provide regular ferry services to the space station. The very idea of turning over responsibility for this capability to commercial entities is predicated upon the condition that such capabilities are no longer experimental. The low-cost outcome is not a surprise. But the degree of savings is.
|Total Program Development Costs|
(NASA portion, inflation adj)
|Commercial Cargo & Crew||$7.6 billion|
Inflation adjusted using NASA's New Start Index. Total program costs are approximations and include development costs of spacecraft, related launch vehicles, engines, and management overhead through the fiscal year of the first test flight. Raw data, methods, and sources are in this Google Sheet.
After nearly 3 years of delays that, in any other situation, would have significantly increased NASA's costs, the agency's fixed-priced contracts limited its financial exposure. The contracts with SpaceX and Boeing remain within 3% of their original amounts. And, should both spacecraft be successful, NASA will pay less, per seat, to send its astronauts to the ISS than during the space shuttle era, and less, overall, than for seats on Russia's Soyuz spacecraft.
The budgetary success (followed very soon, hopefully, by the programmatic success!) of commercial crew paves the way for new experiments in public-private partnerships. The question now is what level of risk NASA and Congress are willing to take in applying them more broadly.
Already, NASA is aggressively pursuing fixed-price partnerships for components of its Artemis program. And, already, NASA is getting push back from Congress in doing so. Given the degree of cost-savings demonstrated by the commercial crew experiment, the potential savings for future projects should be weighted strongly against the risk. Even a failure to deliver would be instructive—NASA would learn the extent to which the lessons of commercial crew apply. And if it succeeds, NASA would once again reap the rewards.