The Space Race Report: Who’s Winning And At What Cost?

More than 7,300 rockets have left Earth since Sputnik 1 in 1957. Over half of them launched in the last decade. In 2025, a rocket left the ground somewhere on Earth every 26 hours. The space age isn't slowing down. This report tracks who's launching, who's winning, what it costs, and what it's doing to the orbit we all share.

Contents:

• The New Space Race In Numbers.
• Which Countries Are Dominating?
• Which Companies Are Leading The Charge?
• The Cost Of The Space Race.
• Who's Investing The Most In Future Missions?
• Space Welfare: Is There A Cost To This Experiment?
• Conclusion.
  

The New Space Race In Numbers.

Space activity has followed three distinct phases since 1957.

From 1957 to 2000, launches were slow, expensive, and state-owned. Annual totals peaked at around 130 in the mid-1970s, driven almost entirely by the US-Soviet arms race, before falling as the Cold War ended. From 2000 to 2015, activity stagnated at 60-90 launches a year. Space remained the preserve of governments with large budgets and longer timelines.

That changed in 2015, when SpaceX landed the first orbital rocket booster back on its launch pad, cutting the cost of reaching orbit and triggering a sustained rise in launch activity. Launch totals that had stagnated for two decades started rising: 87 in 2015, 114 in 2018, 186 in 2022, 330 in 2025.

The reliability picture has shifted just as sharply. In the 1950s and 60s, roughly one in five launches failed. By 2024, the global failure rate had fallen to 2.3% (6 failures from 263 attempts). Modern launch vehicles aren't just cheaper than their predecessors, they're far more dependable too.

The 2026 figures in this report cover January 2026 to June 2026 only. 142 successful launches had been recorded by that date, already ahead of the same period in 2025 (136 successful launches to the same point), putting 2026 on pace to match or exceed last year's record total.

Which Countries Are Dominating?

The United States launched 195 successful missions in 2025. China launched 90. Russia managed 17. The figures are more meaningful when grouped by company nationality rather than launch site, crediting rockets to the country where the company is headquartered and funded, not where the ground happens to be.

In 2021, China launched more rockets than the US (53 to 48) on company nationality figures. By 2022 the US had taken the lead and extended it every year since, reaching 195 launches in 2025 against China's 90. Take out Starlink launches and the gap narrows, but the US still leads.

China's 90 launches in 2025 came entirely from Chinese state and commercial operators. Unlike the US figure, China's number is unaffected by the launch site vs company nationality distinction. Every Chinese rocket launched by a Chinese company from Chinese soil.

Russia has declined from 24 launches in 2021 to 17 in 2025. This fall was almost certainly accelerated by post-2022 sanctions. ESA/Europe held steady at 3-7 launches per year, reflecting the slow ramp-up of Ariane 6 following Ariane 5's retirement.

The satellite data tells a similar story.

Active Satellites In Orbit By Country.

The US figure of 12,389 active satellites in orbit is almost entirely one company. Around 10,413 of those satellites are SpaceX Starlink, a single private operator, not a national space programme. Without them, the US total drops to roughly 1,976, which is still the largest individual-operator count but far closer to China (1,331) than the headline number implies. The UK's 763 tells a similar story. The number belongs to OneWeb/Eutelsat, not to any national programme.

China's direction of travel matters more than where it currently sits. Its satellite count grew from 401 in 2020 to 1,331 in 2026 and this is before its two planned mega-constellations, Guowang (12,992 satellites planned) and Qianfan (15,000 planned), have got off the ground. 

Satellite Growth Across The Top 5 Countries From 2015-2026.

Which Companies Are Leading The Charge?

In 2025, SpaceX conducted more than half of all successful orbital launches on Earth, 52.8%, across its Falcon 9, Falcon Heavy, and Starship vehicles. No single company has ever come close to that share in the history of spaceflight.

Four years earlier in 2021, SpaceX's share was 23%, already significant but not yet dominant. That year, CASC, China's state aerospace operator, actually launched more (36.3%) of all global missions across its Long March family. By 2025, the picture had reversed. SpaceX had grown to 52.8%. CASC had fallen to 23%.

SpaceX and CASC have spent the last five years moving in opposite directions. CASC isn't launching less. In fact China launched more rockets in 2025 than in any previous year, with 73 successful CASC missions across 20 different Long March variants. It's just not keeping pace with how fast the rest of the market has expanded. CASC's share actually recovered slightly in 2025 (from 19.8% to 23%) but this reflects a genuine ramp-up in Chinese state launches rather than any reversal of SpaceX's dominance. The global total grew at the same time, keeping CASC's share well below its 2021 peak.

Roscosmos has had the steepest fall of any major operator: from 17.8% of global launches in 2021 to 5.3% in 2025, across all Soyuz-2, Proton-M, and Angara vehicles combined. Rocket Lab has been the most consistent, holding roughly 4-6% every year since 2021. The clear third operator globally, carving out a small-satellite niche that Falcon 9 is too large to serve efficiently.

China's Rising Challengers.

Beneath CASC's state programme sits a separate, commercially active layer. Vehicles from four independent private operators (Galactic Energy, CAS Space, ExPace, and LandSpace) grew from 4 combined launches in 2021 to 13 in 2025, accounting for 4.1% of global launches that year. Add that to CASC's 23%, and China accounts for roughly a quarter of all global launches in 2025, split between a state programme that remains dominant and a commercial sector still finding its feet.

China’s Commercial Launch Operators.

China commercial = independent private operators only (Galactic Energy, ExPace/CASIC, CAS Space, LandSpace). Jielong-3 is classified under CASC, operated by China Rocket Co., a 100% CASC subsidiary. Market share is calculated as a percentage of successful global launches within each year.

The Cost Of The Space Race.

The table below shows how much countries are spending to visit space. Every country runs two space programmes, a civil one and a military one. For the US, that split is public. NASA's $25 billion civil budget sits alongside roughly $55 billion flowing through the Department of Defense, US Space Force, and intelligence agencies.

China publishes no equivalent breakdown. The ~$19.9 billion figure is a Novaspace/Euroconsult estimate combining CNSA's civilian programme with the PLA Strategic Support Force, the unit responsible for military satellites, anti-satellite weapons, and space surveillance. The true total could be higher. The US figure is published, whereas China's is an estimate, which means they are not directly comparable. Russia's ~$3.4 billion was recorded before post-2022 sanctions took full effect, so what the country actually spends on space today is also harder to know.

How Launch Costs Have Collapsed

Getting a kilogram to low Earth orbit cost around $54,000 in the Space Shuttle era. On a reused Falcon 9 today, it costs roughly $2,700-3,000. That is a 90-95% reduction, achieved almost entirely through one company's decision to land and reuse its rocket boosters.

Starship is designed to carry 100-150 tonnes to low Earth orbit at a target cost of $100-200 per kilogram. If that target is reached, missions that are currently too expensive to justify, such as large space stations, in-space manufacturing and crewed Mars missions, could become financially viable. 

Lower costs would also reshape who can afford to go. Missions that currently require billions could come down to hundreds of millions, potentially opening up serious space programmes to countries and institutions that can't currently afford them. SpaceX has not published a commercial price, so we should treat the $100-200/kg figure as a design goal, not an achieved price.

Who’s Investing The Most In Future Missions?

NASA is targeting the first Artemis lunar landing in early 2028, which will be the first time humans will have stood on the Moon since December 1972. That mission is Artemis IV. Artemis III, scheduled for late 2027, is a crewed orbital docking test with commercial lunar landers (an essential rehearsal before anyone descends to the surface). Artemis II already flew in April 2026, sending four astronauts on a lunar flyby, the first crewed flight beyond low Earth orbit since Apollo 17 in 1972.

The missions building towards that landing are already in motion. Firefly's Blue Ghost 2 is scheduled for late 2026, delivering NASA and ESA payloads to the lunar far side and deploying an ESA communications satellite into lunar orbit. This is part of the infrastructure NASA needs in place before astronauts can land. 

The US isn't the only country with its sights on space. India is targeting its first crewed orbital flight with Gaganyaan in 2027, which would make it only the fourth country to independently launch humans to space. It is also planning Chandrayaan-4 for 2027-2028, a lunar sample return that would follow Chandrayaan-3's successful south pole landing in 2023, which would make India one of only a few countries to have returned material from the Moon.

China has announced a crewed lunar landing target of 2030. Both programmes are focused on the same patch of ground near the lunar south pole, where confirmed water ice deposits make long-term habitation possible. NASA's Moon Base programme envisions robotic infrastructure deliveries beginning in 2028, with a permanent crewed outpost targeted by 2030 (the same year China aims to land). Whether China gets there before or after the US is genuinely uncertain, and consequential for how the next decade of space governance plays out.

Mars is also coming into plans. China plans to launch two rockets simultaneously in 2028 for Tianwen-3, land sample-collection hardware on Mars, and return those samples to Earth by 2031. No country has done this before. NASA has been planning a Mars sample return for years and is still working through the funding and logistics. Japan's MMX mission, launching late 2026, will attempt something different by landing on Phobos, one of Mars's two moons, and returning a sample to Earth in 2031.

Not every headline mission is heading to the Moon or Mars. NASA’s Nancy Grace Roman Space Telescope launches in fall 2026 on a Falcon Heavy. It is designed for wide-field surveys of dark energy, exoplanets, and distant galaxies. ESA's PLATO mission follows in December 2026, monitoring 200,000 stars for Earth-like planets in habitable zones.

NASA's Dragonfly, a $3.35 billion nuclear-powered rotorcraft, launches in July 2028 on a Falcon Heavy, bound for Saturn's moon Titan. It won't arrive until 2034. At that point it will become the first rotorcraft ever to fly on another world, covering more ground in its mission than all previous planetary rovers combined.

The Mega-Constellation Arms Race.

The satellite constellation race is the less visible but equally significant side of the space race.

Starlink already has around 10,413 satellites in low Earth orbit, with FCC approval to expand to 15,000 next-generation satellites. China is building two constellations to rival it. Guowang, a state-backed network of around 13,000 satellites filed with the ITU, and Qianfan, a commercial constellation targeting 15,000 by 2030. Combined, the two programmes have around 350 satellites in orbit as of mid-2026. This is a small fraction of their stated goals, but deployment is accelerating fast. 

Amazon Leo (formerly Project Kuiper, rebranded November 2025) is positioning itself as a third major player. It has 365+ satellites in orbit as of 2026, nowhere near the 1,618 required by its FCC deadline of 30 July 2026, but Amazon has requested an extension. The programme has a long way to go before it becomes a usable broadband network.

OneWeb, operated by Eutelsat, is already operational. It completed its first-generation 648 satellite constellation in 2023 and is now in the process of replacing it. 440 next-generation satellites were ordered from Airbus in January 2026, with deliveries beginning late 2026. Phase 2 of OneWeb is no longer dependent on funding, it is already under contract.

Taken together, objects in low Earth orbit could exceed 50,000 by 2035. The FAA's high forecast projects 456 US commercial launches in 2030, compared with 148 recorded in 2024. That figure covers US launches only (China, Russia, and ESA are not included).

Space Welfare: Is There A Cost To This Experiment?

The space race has costs beyond budgets. Debris, carbon emissions, and light pollution are all direct consequences of the launch boom and none are close to being resolved. 

Orbital Debris 

The sheer volume of objects now in orbit is a direct consequence of the launches tracked in this report. Our Space Debris Report takes a deeper look at the debris crisis in detail. ESA's 2025 Space Environment Report concludes that even if every launch stopped today, debris levels would continue growing for over 200 years. The orbit above Earth is filling up faster than it can clear itself. At 28,000 km/h, the speed at which objects travel in low Earth orbit, even a loose screw can carry enough kinetic energy to destroy an operational satellite.

The Carbon Cost

A single Falcon 9 launch produces an estimated 425 tonnes of CO2. With 330 launches in 2025 and the FAA projecting 456 US launches alone by 2030, the trajectory is upward. Standard carbon accounting also almost certainly understates the real impact. Black carbon (soot deposited directly into the stratosphere) may have a warming effect up to 500 times greater than equivalent CO2 at ground level. It's an active area of research, and the numbers are not yet in the industry's favour.

Light Pollution

More than 10,000 Starlink satellites are now in orbit. Astronomers at the Rubin Observatory warned that once that number was reached, satellite streaks would appear in nearly every twilight image captured by wide-field telescopes. The prospect of Guowang, Qianfan, and Amazon Leo adding another 30,000+ satellites has no international coordination mechanism to address it. The IAU's Centre for the Protection of the Dark and Quiet Sky is monitoring the situation and working with satellite operators, but has no enforcement powers. The night sky isn't protected by any enforceable treaty.

Conclusion

The original space race was a contest between two superpowers, driven by geopolitical pride and defence budgets. Launches were government-owned, expensive, and largely secret. The whole thing nearly stopped when the Cold War ended. The difference this time is commercial economics, not national prestige. The cost of reaching orbit has fallen by more than 90% in a decade, and that has opened access to a set of actors including private companies, smaller nations, and commercial constellations that had no role in the original space race.

SpaceX's dominance of the launch market is the clearest expression of this shift. But what’s happening in China may matter more in the long run. In 2025 alone, it made 73 CASC launches, has two mega-constellations under deployment, a crewed lunar landing targeted for 2030, and the world's first Mars sample return mission planned for 2028. It is running its own programme, at its own pace, with its own targets.

At the same time, the orbit above Earth is filling with debris, carbon emissions from launches are rising, and tens of thousands of satellites are being deployed with no binding international coordination framework. The Outer Space Treaty dates from 1967. The night sky has no legal protection, which means nobody is in charge of low Earth orbit.

Whether the US lands on the Moon before China, whether Starship achieves its cost targets, and whether Amazon Leo meets its FCC deadlines are all genuinely uncertain. What isn't uncertain is that the pace of activity is increasing, the number of actors is growing, and the infrastructure decisions being made right now will define what space looks like for the next fifty years. Unlike the original space race, this one has no finish line and no single measure of success.

Methodology

All launch counts (1957-2026) and active satellite counts are sourced from the AEI Space Data Navigator (spacedata.aei.org), downloaded as raw CSV exports on 23 June 2026. The dataset covers 7,330 orbital and deep space launch records. All launch counts reflect successful launches only unless stated otherwise. Government budgets are sourced from Orbital Radar (orbitalradar.com/space-economy/government-space-budgets). All data correct as of 23 June 2026.