Series 7 — The Empire
The Empire — Series 7 · Part 1 of 5 · Five investigations into how Elon Musk built the infrastructure America runs on
The Empire · 1 of 5
Investigation #0701 · Published March 2026

THE
BUILDER'S
BARGAIN

SpaceX didn’t lobby its way to $38 billion in government contracts. It out-built every competitor. The documented story of how Elon Musk created the infrastructure America had no choice but to depend on — and what happens when the government and the builder become inseparable.

Red String Investigation Primary sources · Federal spending data · Congressional records · Agency reports ~18 min read
Sources: NASA CCDEV Records · Washington Post Federal Contracts Analysis, 2025 · SpaceX CEO Gwynne Shotwell, Nov 2024 · Congressional Budget Office

The Number

In November 2024, SpaceX CEO Gwynne Shotwell told a conference audience that her company holds approximately $22 billion in active government contracts. A Washington Post analysis published around the same time put the total figure — contracts, loans, and subsidies across all of Elon Musk’s companies since 2003 — at over $38 billion. In fiscal year 2024 alone, the number reached $6.3 billion, the highest single year in the company’s history.

When asked whether SpaceX had earned those contracts or simply benefited from government largesse, Shotwell was direct: “We earned that.”

The primary record largely supports her. SpaceX did not inherit its position in American spaceflight. It built it — incrementally, over two decades, against the active resistance of legacy aerospace contractors who had held the government’s space business for half a century. The $38 billion represents the accumulated cost of something historically rare: a private company that genuinely changed what a nation could do.

Understanding that number requires understanding the state of American spaceflight before SpaceX existed. In July 2011, the Space Shuttle program ended. The United States had no domestic capability to transport astronauts to the International Space Station. For nine years, NASA paid Russia’s Roscosmos between $21 million and $90 million per seat on the Soyuz spacecraft for every American who needed to reach orbit. NASA Inspector General estimates put the total cost of that dependence above $3.9 billion.

The program that ended that dependence was called Commercial Crew. Its creation — and what happened to the two companies NASA selected — is the central story of how SpaceX became indispensable.

$38B+
Total contracts, loans & subsidies since 2003
Washington Post federal contracts analysis, 2025
$22B
Active contracts as of Nov 2024
SpaceX CEO Gwynne Shotwell, confirmed Nov 2024
$6.3B
FY2024 — highest single year on record
Washington Post / ABC News federal spending analysis

The Capability Gap

NASA’s Commercial Crew Development program was launched in 2010 with a specific mandate: fund private companies to build crew transport systems, creating market competition rather than building a government-owned system. The theory was that commercial competition would drive down costs and accelerate development. For one of the two companies selected, that theory proved spectacularly correct. For the other, it became one of the most expensive aerospace failures in recent memory.

NASA awarded Commercial Crew Transportation Capability contracts in September 2014: SpaceX received $2.6 billion to develop Crew Dragon, and Boeing received $4.2 billion to develop Starliner. Boeing received more money, had decades of experience building spacecraft for NASA, and was backed by the manufacturing infrastructure of one of the world’s largest defense contractors. The results were not close.

2015
SpaceX begins Crew Dragon development — while still flying commercial missions
SpaceX simultaneously develops Crew Dragon for NASA while continuing Falcon 9 commercial launch operations. Revenue from commercial launches cross-subsidizes development — a model legacy contractors, dependent on cost-plus billing, could not replicate.
2020
Crew Dragon Demo-2: First American crewed orbital launch since 2011
On May 30, 2020, SpaceX launches astronauts Doug Hurley and Bob Behnken to the ISS aboard Crew Dragon Endeavour — the first American crewed orbital mission in nine years, launched from Kennedy Space Center’s historic Launch Complex 39A. NASA declares the mission a success. The first operational commercial crew mission follows in November 2020.
2019–2022
Boeing Starliner: Three attempts before first successful uncrewed docking
Boeing’s first orbital flight test in December 2019 suffered a mission clock error that prevented docking with the ISS. A second attempt in August 2021 was scrubbed due to propellant valve corrosion. The third attempt in May 2022 finally docked successfully — seven years behind SpaceX’s operational timeline. Boeing absorbed more than $600 million in cost overruns on fixed-price contract terms.
2024
Starliner’s first crewed flight: helium leaks, thruster failures, astronauts stranded
Boeing’s first crewed Starliner flight, carrying astronauts Butch Wilmore and Suni Williams for an 8-day test mission, departs June 5, 2024. Helium leaks and thruster failures are discovered in orbit. After months of NASA analysis, the agency determines Starliner cannot safely return the crew. Wilmore and Williams return to Earth in March 2025 — aboard a SpaceX Crew Dragon. NASA has not announced a date for Starliner’s next crewed flight.
2025
SpaceX: The only operational American crew transport system
As of early 2025, SpaceX is the sole American provider capable of transporting crew to and from the International Space Station. The company has completed 13+ crewed NASA missions. The alternative — paying Russia for Soyuz seats — was the status quo SpaceX replaced.

The Starliner situation is not incidental to the story of SpaceX’s government relationship — it is central to it. The government did not choose to depend on SpaceX. It found itself dependent on SpaceX because every alternative failed or was unavailable. This pattern repeats across the company’s contract history.

“The question for NASA was never ‘how much should we give SpaceX.’ It was ‘who else can do this?’ And the answer, repeatedly, was nobody.”

Space policy context per multiple aerospace analyst assessments, 2020–2025

The Contract Stack

The $38 billion figure encompasses contracts, loans, and subsidies across multiple Musk-affiliated companies, awarded by multiple agencies over more than two decades. Broken down by program, each contract traces to a specific capability and a specific moment when the government determined that SpaceX was the best available option. The list below documents the nine largest disclosed contracts.

Disclosed Government Contracts — SpaceX / Musk Companies
Source: USAspending.gov · NASA · Space Force · Congressional records
Program & Agency
Value
Status
NASA
Commercial Resupply Services CRS-2 — ISS cargo transport through 2030. Cap contract up to $14B across all awards
$3B+
Obligated to date
Active
Space Force
NSSL Phase 3 Lane 2 — 28 of 54 national security space launch missions, FY2027–2032
$5.92B
Awarded June 2023
Active
NASA
Human Landing System Option B — Starship lunar lander for Artemis III and IV crewed Moon missions
$4B+
Options exercised 2023
Active
NRO / DoD
Starshield classified national security satellite constellation — intelligence-gathering orbital layer
~$1.8B
Reported; unconfirmed officially
Classified
NASA
ISS Deorbit Vehicle — SpaceX to build and operate the vehicle that safely deorbits the International Space Station, targeting 2030
$843M
Awarded June 2024
Active
Space Force
NSSL Phase 3 Lane 1 — medium-lift national security missions, fixed-price launches FY2022–2024
$733M
Completed
Complete
NASA
Commercial Crew CCtCap — Crew Dragon development plus operational crew transport. 13+ missions completed to ISS
$2.6B+
Development + task orders
Active
NASA
Launch Services Program — rideshare and dedicated Falcon 9 / Falcon Heavy science missions since 2010
$1.4B+
Since 2010, ongoing
Ongoing
DoD / State Dept
Starlink Ukraine — satellite internet terminals and service for Ukrainian military communications
$23M+
Documented; additional classified
Active

The contract stack has a structural logic: as SpaceX demonstrated reliability at one scale, agencies contracted for the next. CRS-1 cargo contracts led to CRS-2. Early crew development awards led to CCtCap. NSSL Lane 1 wins led to Lane 2. The Lunar Starship contract followed after Crew Dragon had completed multiple operational ISS missions. Each contract built on demonstrated performance from the last.

This is different from how legacy aerospace contractors typically accumulate government business — through lobbying, political relationships, and institutional inertia within existing cost-plus programs. SpaceX operated on fixed-price contracts from the beginning, absorbing cost overruns itself rather than passing them to the taxpayer. When it won on price and schedule, it kept the margin. When it failed — as it did with the CRS-7 launch vehicle loss in 2015 — it absorbed the loss.

The Cost Revolution

The economic case for SpaceX’s government relationship is most clearly illustrated by launch costs. The transformation in what American taxpayers pay per kilogram to orbit — driven almost entirely by SpaceX’s development of reusable booster technology — is among the most significant cost reductions in the history of American defense and space budgets.

Mission Type
Legacy Cost (ULA / Pre-SpaceX)
SpaceX Cost (2024)
Medium-lift orbital launch
$350–400M (Atlas V)
$67M (Falcon 9)
Heavy-lift orbital launch
$350M+ (Delta IV Heavy)
$150M (Falcon Heavy)
Crew transport to ISS (per seat)
$55–90M (Soyuz, to Russia)
~$55M (Crew Dragon, domestic)
Satellite deployment (rideshare)
$50–80M+ per mission
$5.9M (Transporter rideshare)

The Congressional Budget Office documented that SpaceX’s entry into national security launches under the Evolved Expendable Launch Vehicle program saved taxpayers an estimated $500 million per year compared to the previous sole-source arrangement with United Launch Alliance, the Boeing-Lockheed Martin joint venture. Over the life of NSSL contracts, projected savings exceed $2 billion.

These savings did not come from government mandates. They came from a specific engineering conviction that the rest of the industry thought was impossible: that orbital-class rocket boosters could be reliably recovered and reused. ULA CEO Tory Bruno publicly doubted SpaceX’s reusability claims in 2014. By 2020, Falcon 9 boosters had been reflown dozens of times. The booster B1060 has flown more than twenty missions.

The Reusability Thesis

Musk’s foundational argument — that launch vehicles were expensive because they were treated as expendable, and that reusability would break the cost curve — was dismissed by the established aerospace community as naive. Legacy contractors operated on cost-plus contracts with no incentive to reduce costs. Their business model rewarded complexity and spending, not efficiency and reuse.

SpaceX inverted those incentives. On fixed-price contracts, every dollar of cost reduction became profit. The company structured itself to find and eliminate inefficiency — using commodity hardware where custom hardware had been the norm, iterating through rapid test-to-failure cycles rather than the slow development processes that had produced Starliner, and building vertical manufacturing capability that reduced dependence on the supplier networks the legacy industry had accumulated over decades.

“If you’re asking whether SpaceX earned its contracts on merit, look at what happened to every other option. Starliner is grounded. The Atlas V is being retired. The Delta IV is gone. Falcon 9 just flew its 300th mission.”

Aerospace analyst characterization of the competitive landscape, 2025

The implications for government contracting were direct. When an agency can buy a launch for $67 million instead of $350 million, the budget math for every mission changes. NASA could fund more science. DoD could deploy more national security satellites. The defense budget could absorb the cost of building a new space architecture — Starlink, Starshield, eventually Golden Dome — that would have been fiscally impossible under legacy pricing.

Starlink and the Defense Sector

Starlink began as a commercial product: satellite broadband delivered via a low-earth-orbit constellation. As of early 2026, that constellation includes over 7,000 satellites, making it the largest satellite network in history. The commercial service has approximately 4 million subscribers across 60+ countries.

The defense dimensions of Starlink were not part of the original commercial pitch. They emerged from the technology’s demonstrated performance in an actual military environment: Ukraine.

Ukraine: The Operational Proof

When Russia invaded Ukraine in February 2022 and systematically attacked communications infrastructure, Starlink terminals became the primary communications backbone for Ukrainian military units. By mid-2022, approximately 150,000 terminals were operational, enabling drone coordination, tank unit communications, and command-and-control functions from brigade level down to infantry squads. Ukrainian commanders came to rely on Starlink the way a previous generation relied on radio.

The military value of a low-latency, resilient, space-based communications network that cannot be eliminated by destroying ground infrastructure was immediately apparent to every major defense establishment in the world. The U.S. Department of Defense contracted for Starlink services in Ukraine and began exploring dedicated defense applications. The NRO Starshield program — a classified derivative designed for intelligence gathering — followed.

Starshield represents something new in national security space: a private company operating a classified government satellite constellation using commercial-off-the-shelf manufacturing techniques at a scale and cost impossible for government-built systems. Reuters and the Wall Street Journal have reported the initial Starshield contract at approximately $1.8 billion, with the program expected to expand.

Golden Dome: The Next Architecture

The Golden Dome missile defense system, which received a $25 billion allocation in the 2025 House Republican budget proposal, would extend Starlink’s architecture to ballistic missile defense. SpaceX is among the top contenders for the prime contract. The case for SpaceX, per defense analysts cited in multiple industry publications, rests on the same logic as Starlink’s military adoption: no other contractor has demonstrated the satellite manufacturing cadence, launch capacity, and orbital operations experience necessary to build and operate a space-based defense layer at the proposed scale.

Proliferated LEO Architecture — Defense Strategic Value
Defense planners have noted that SpaceX’s manufacturing throughput — the ability to build and launch hundreds of satellites per year — creates a “proliferated LEO” architecture whose resilience to attack fundamentally differs from previous satellite generations. A constellation of 7,000+ satellites cannot be eliminated by destroying a handful of assets. This architectural property is increasingly central to DoD space strategy and is the core of SpaceX’s competitive case for Golden Dome.

The Structural Question

The $38 billion in contracts raises a question that is separate from competence or performance: what happens when the government becomes operationally dependent on a single private company for critical infrastructure?

This is not a question unique to SpaceX. The federal government is operationally dependent on Microsoft for software, on Amazon Web Services for cloud infrastructure, on a small number of semiconductor manufacturers for defense systems. Managing dependence on essential private infrastructure has been a policy challenge for decades.

In SpaceX’s case, the dependence is concentrated and visible. There is no domestic backup for Crew Dragon. There is no equivalent to Starlink for military communications. There is no other commercial provider approaching SpaceX’s launch cadence. That concentration is a consequence of SpaceX’s performance — it won the business because it built what others couldn’t. But it creates institutional questions that defense planners take seriously regardless of their view of Musk: How should critical defense systems be structured when only one company can build them? What contract terms protect government interests? How do you create future competition in a market one company has defined?

The Bargain, Documented

The government contracted with SpaceX because SpaceX built what the government needed and no one else could provide it at acceptable cost or schedule. That success created a dependency that now gives the company structural significance in programs of national importance — not because of political influence, but because of technical capability. The $38 billion is both the measure of that capability and the record of how the government came to need it. The remaining four parts of this series examine what Musk built with the position that capability created.

The Empire — Series 7
Next: The Network
Nine companies. Shared technology, shared talent, and shared infrastructure. How the individual pieces of Musk’s enterprise compound into something greater than their parts — and what it means for every market he enters.
Part 2: The Network →

Primary Sources

  • [1] Washington Post — Federal contracts analysis (Feb 2025) — $38B total figure across all Musk companies since 2003
  • [2] SpaceX CEO Gwynne Shotwell — Conference statement (Nov 2024) — “$22B active” and “We earned that”
  • [3] ABC News — SpaceX federal contracts analysis (2024) — $3.7B SpaceX FY2024 direct
  • [4] NASA — Commercial Crew Development CCDEV / CCtCap award records — SpaceX $2.6B, Boeing $4.2B
  • [5] NASA Inspector General — ISS access cost assessments — Soyuz seat pricing history and total cost above $3.9B
  • [6] Space Force — NSSL Phase 3 Lane 2 award announcement (June 2023) — $5.92B for 28 missions FY2027–2032
  • [7] NASA — Human Landing System Option B contract — Starship lunar lander for Artemis III and IV
  • [8] Reuters / Wall Street Journal — NRO Starshield reporting — ~$1.8B classified constellation
  • [9] NASA — ISS Deorbit Vehicle contract announcement (June 2024) — $843M SpaceX award
  • [10] Congressional Budget Office — EELV / NSSL cost comparison — $500M/year savings vs. ULA sole-source; $2B+ projected total
  • [11] Boeing / NASA joint statements — Starliner cost overruns; $600M+ Boeing absorbed losses per fixed-price contract
  • [12] NASA — Starliner CFT mission status reports (2024) — Helium leak documentation and crew return decision
  • [13] DoD / State Dept — Starlink Ukraine terminal and service contracts — $23M+ documented
  • [14] Space Force / BuiltIn — Falcon 9 and Falcon Heavy pricing data, B1060 booster reuse record
  • [15] House Armed Services Committee — Golden Dome $25B appropriation proposal, 2025 budget
METHODOLOGY & LEGAL NOTE
This investigation is based exclusively on primary sources cited within the article: court records, government documents, official filings, peer-reviewed research, and named expert testimony. Factual claims reflect what those cited sources establish. Where findings are disputed or unresolved, those disputes are noted. Characterizations represent the documented record, not conclusions beyond it. Red String is an independent investigative publication. Corrections: [email protected]  ·  Editorial Standards