Has a Second Space Age Turned Space Into Industrial Infrastructure?

Key Takeaways

• Space is shifting from a government domain toward industrial infrastructure tied to terrestrial markets.
• Lower launch costs opened access, but capital, spectrum, regulation, and demand now shape expansion.
• Commercial space networks increasingly connect economic power with national security and state policy.

A Second Space Age Is Already an Industrial Story

On June 15, 2026, SpaceX completed its initial public offering, issuing 638,888,888 Class A shares at $135 per share and generating $85.675 billion in net proceeds after underwriting commissions and offering costs. The transaction placed one of the world’s largest launch, satellite communications, and space technology companies directly inside public capital markets on a scale rarely associated with aerospace companies.

That event captures much of the argument made by the August 12, 2026 Goldman Sachs Global Institute paper, The Second Space Age: How Markets, Technology, and Power Are Reshaping the Final Frontier. Goldman argues that space activity is moving beyond an era dominated by government exploration programs and becoming part of the industrial economy. Launch capacity, satellite production, communications networks, navigation systems, Earth observation, ground infrastructure, capital markets, military requirements, data services, and regulation now interact in ways that increasingly resemble other large industrial systems.

The argument becomes stronger when space is viewed through what happens on Earth rather than through the number of spacecraft in orbit. Satellite communications support broadband, television, aviation, maritime operations, mobile networks, and government communications. Positioning, navigation, and timing systems support transportation, financial transactions, electrical grids, telecommunications networks, precision agriculture, and logistics. Earth observation feeds weather forecasting, environmental monitoring, insurance, agriculture, intelligence, infrastructure management, and commodity analysis.

That wider definition also explains why estimates of the space economy extend far beyond the revenues of rocket manufacturers and satellite operators. A 2024 McKinsey and World Economic Forum analysis estimated that the global space economy could reach $1.8 trillion by 2035, compared with $630 billion in 2023. Its definition includes both direct space activities and economic value generated through space-enabled services in terrestrial industries. The figure remains a forecast rather than an established outcome, and differences in market definitions mean it should not be treated as interchangeable with narrower space-industry revenue estimates.

This distinction matters because launch revenue can remain small relative to the economic activity enabled by what gets launched. A navigation satellite can support economic activity far exceeding the cost of the spacecraft and its launch. The same relationship applies to weather satellites, communications constellations, geospatial data systems, and military warning networks.

New Space Economy’s August 2026 explanation of how the space economy works makes a related distinction between orbital infrastructure and the terrestrial industries that consume its output. That framing helps explain why the economic significance of space cannot be measured through launch counts or spacecraft manufacturing revenue alone.

The more consequential change is institutional. Governments remain enormous customers, regulators, investors, technology sponsors, and operators, yet private companies increasingly own assets that governments depend upon. Financial markets now finance satellite networks and launch infrastructure. Telecommunications companies integrate satellite links into terrestrial services. Defense organizations buy commercial imagery, communications, and launch capacity. The dividing line between a commercial space company and a strategic infrastructure provider has become harder to draw.

Falling Launch Costs Changed the Economics of Access

The technological basis of the commercial model begins with access to orbit. Goldman’s analysis traces a long reduction in the cost of launching payload mass into low Earth orbit and connects that decline to reusable launch vehicles, higher launch cadence, improved manufacturing, and the ability to spread fixed costs across repeated missions.

Launch-cost comparisons require care because published values may represent advertised prices, estimated provider costs, dedicated mission prices, rideshare prices, or calculations adjusted for inflation. The Center for Strategic and International Studies launch-cost dataset compares dedicated-launch cost per kilogram and identifies vehicle characteristics that affect those calculations. The dataset was last updated in September 2022, so it should be treated as a historical comparison rather than a current price list. Its data nevertheless document a pronounced long-term decline in estimated launch cost per kilogram, particularly after the mid-2000s.

Lower prices altered spacecraft design. When every kilogram was exceptionally expensive, operators had strong incentives to build long-lived satellites containing redundant systems and specialized components. Smaller spacecraft became more practical as electronics shrank, commercial components improved, software became more capable, and launch opportunities became easier to obtain.

That shift produced a manufacturing model closer to industrial production than traditional bespoke spacecraft development. Constellation operators can build satellites in batches, revise designs between production runs, distribute functions across many spacecraft, and replace individual units rather than designing every spacecraft for decades of service. This does not make satellites inexpensive in ordinary terms, but it changes the relationship among spacecraft price, expected life, production volume, and network design.

Reusable launch vehicles amplify the effect because launch capacity can support internal demand as well as external customers. SpaceX reported in its June 2026 quarterly filing that it conducted 77 Falcon launches during the six months ending June 30. Of those, 17 were customer launches and 60 were internal launches. The filing explains that SpaceX allocates substantial launch capacity to its Connectivity segment, meaning launch operations can build company-owned infrastructure instead of functioning solely as a stand-alone service sold to external customers.

This creates a different competitive structure from a launch company whose business depends almost entirely on external payload customers. An integrated operator can use rockets to build its own communications network, derive recurring revenue from that network, and use the resulting demand to support launch cadence. New Space Economy’s examination of vertical integration in space companies explains why this model can create manufacturing, scheduling, learning, and cost advantages without establishing that every competitor should reproduce it.

Lower launch prices have also changed which business ideas can receive serious consideration. Satellite servicing, orbital transfer vehicles, commercial stations, microgravity manufacturing, large constellations, lunar logistics, and orbital computing all become more plausible when transportation represents a smaller share of total project cost.

Transportation economics alone cannot create a market. A company can obtain inexpensive launch capacity and still fail because its spacecraft costs too much, customers will not pay, regulatory approval takes too long, spectrum is unavailable, insurance is expensive, financing disappears, or operations prove unreliable. Falling launch cost is better understood as an enabling condition than as a guarantee of commercial success.

Low Earth Orbit Is Becoming Part of the Digital Economy

Low Earth orbit, commonly abbreviated LEO, has become the busiest commercial testing ground for this industrial model. Its proximity to Earth reduces communications latency compared with geostationary orbit and allows smaller spacecraft and user terminals to maintain useful links. The tradeoff is that a LEO network requires many satellites moving continuously relative to users on the ground.

The scale reached by Starlink shows how far that model has progressed. SpaceX reported that, as of June 30, 2026, its connectivity network was powered by more than 10,200 Starlink satellites and served customers across 167 countries, territories, and other markets. The same filing reported 12.0 million Starlink subscription service lines, compared with 6.0 million one year earlier. SpaceX defines that measure as unique service lines rather than unique individuals, devices, or end users, so the figure should not be described as 12 million individual customers.

These numbers turn satellite broadband from a specialized remote-connectivity product into a significant telecommunications platform. The commercial questions now concern network capacity, pricing, terminal economics, spectrum access, competition, customer retention, geographic expansion, and integration with terrestrial networks.

Direct-to-device and direct-to-cell services extend the same logic by connecting conventional mobile devices to satellite networks without requiring a separate broadband terminal. That approach can make satellite connectivity an extension of terrestrial mobile services rather than an isolated consumer product. Agreements between satellite operators and mobile carriers can also provide access to existing subscriber bases without requiring the satellite company to build a terrestrial retail network from scratch.

The communications sector illustrates a broader pattern in the space economy. Customers usually do not purchase space as an end product. They purchase connectivity, location information, imagery, weather information, timing, security, analytics, or data. Space infrastructure competes with terrestrial alternatives on service quality, coverage, reliability, latency, price, and availability.

A New Space Economy review of megaconstellations connects the rise of large LEO systems to reusable launch, satellite manufacturing, spectrum access, and network economics. These systems also reveal the industrial concentration emphasized by Goldman because a large constellation combines spacecraft production, launch access, spectrum, user equipment, ground networks, software, financing, and continuous operations.

That concentration can produce economies of scale, but it can also create dependency. Airlines, maritime operators, military units, disaster-response agencies, businesses, and households may come to depend on services controlled by a small group of operators. Governments then face policy questions that resemble those associated with telecommunications networks, cloud computing, semiconductor supply chains, or energy infrastructure.

LEO communications also demonstrate why the space economy can expand without consumers thinking of themselves as space customers. A passenger purchasing in-flight internet, a shipping company monitoring vessels, or a mobile subscriber receiving satellite coverage may never interact directly with a satellite operator. The space component becomes part of an ordinary digital service.

Economic expansion may consequently come less from exotic activities conducted far from Earth and more from making orbital infrastructure an ordinary component of communications, logistics, data, defense, financial services, and industrial operations.

Capital Markets Are Changing the Scale of Commercial Space

Technology explains only part of the commercial expansion. Large space systems require substantial amounts of capital, often years before the assets generate stable revenue. Manufacturing plants, launch sites, rockets, satellite constellations, ground networks, data centers, user terminals, testing facilities, and regulatory programs all require financing.

Private investment in the sector has grown considerably since the 2010s, though annual and quarterly totals fluctuate with interest rates, market sentiment, acquisition activity, company valuations, and large financing events. Space Capital reported by July 2026 that its manually curated database tracked $488 billion invested across the space economy since 2009. Its second-quarter 2026 assessment described the quarter as a turning point for public-market access, led by the SpaceX IPO and increased merger and acquisition activity.

SpaceX’s June IPO demonstrated how dramatically public markets can change the financing ceiling. The company issued 638,888,888 Class A shares at $135 each and reported $85.675 billion in net proceeds. Its Form 10-Q states that those proceeds are intended to support growth that includes AI computing infrastructure, launch infrastructure and vehicles, expansion of satellite constellations, and general corporate purposes.

That transaction connects businesses once discussed separately. Launch infrastructure, satellite communications, artificial intelligence computing, software, and capital markets now sit inside the finances of one publicly traded company. SpaceX’s June filing identifies three reportable segments: Space, Connectivity, and AI. The structure illustrates how a space company can become an integrated technology and infrastructure platform rather than remaining a narrowly defined aerospace contractor.

This structure supports Goldman’s contention that access to capital can itself become a source of competitive advantage. Companies with lower financing costs can build factories earlier, buy long-lead components, absorb development failures, expand launch capacity, deploy networks before rivals, pursue acquisitions, and survive longer periods before new business lines mature.

The opposite dynamic matters too. Space companies remain exposed to capital cycles. A technically sound business can encounter difficulty if funding costs rise or investors become unwilling to finance long development schedules. Markets such as commercial stations, lunar logistics, orbital manufacturing, and servicing may require several rounds of financing before demand becomes predictable.

Public markets add another discipline. Companies must disclose revenue, expenses, capital requirements, operating metrics, risks, and business concentration in forms that private companies can often keep confidential. Greater disclosure can make it easier for investors to distinguish operating businesses from projects built primarily around forecasts.

Public listings do not mean space has become a conventional asset class in every sense. Launch failures, spacecraft losses, regulatory decisions, program delays, government procurement changes, technical redesigns, and geopolitical events can produce unusually large changes in value. Space infrastructure also tends to require substantial spending before customer demand is fully known.

Capital nonetheless changes what companies can attempt. A sector financed mainly through government contracts and venture funding faces a different scale limit from one that can use public equity, debt markets, government procurement, strategic investment, export finance, and recurring service revenue together.

Commercial Space Power Is Becoming National Power

Government involvement did not disappear when private investment increased. The relationship changed. Governments now purchase services from commercial networks, finance technology development, regulate launch and spectrum, contract for imagery and communications, support scientific missions, and treat domestic space capability as an element of economic and military security.

NASA’s Artemis II mission provided a visible demonstration of renewed government ambition. Artemis II launched on April 1, 2026, carrying NASA astronauts Reid Wiseman, Victor Glover, and Christina Koch with Canadian Space Agency astronaut Jeremy Hansen on a voyage around the Moon. The crew splashed down on April 10 in the Pacific Ocean off San Diego after a mission lasting 9 days, 1 hour, and 32 minutes. NASA identifies Artemis II as the initial crewed Artemis mission and the initial crewed flight of Orion.

The U.S. policy framework extends beyond that mission. Executive Order 14369, signed on December 18, 2025, sets policy goals that include returning Americans to the Moon through Artemis by 2028 and establishing initial elements of a permanent lunar outpost by 2030. As of August 18, 2026, these remain policy objectives rather than completed outcomes, and their execution depends on appropriations, spacecraft readiness, launch systems, contractor performance, and later government decisions.

Commercial providers increasingly sit inside state strategy. Launch services, satellite communications, remote sensing, space domain awareness, spacecraft manufacturing, and data analysis can serve both commercial and government customers. An asset developed for civilian broadband can acquire military relevance because resilient connectivity and secure communications matter during conflict.

The resulting relationship does not fit a simple government-versus-private-sector model. Governments can become anchor customers that make commercial investment possible. Companies can own infrastructure that states regard as strategically important. Investors can finance capabilities later purchased by defense organizations. National policy can influence which suppliers gain scale.

New Space Economy’s analysis of sovereign space capability treats sovereignty as a question of control rights rather than a requirement that every country own an entire domestic space industry. That approach is relevant because most countries cannot economically reproduce every part of a space supply chain. Trusted partnerships, assured access, diversified suppliers, domestic control of selected systems, and contractual rights can provide alternatives to complete national ownership.

The geopolitical dimension extends to rules and coalitions. NASA established the Artemis Accords with the U.S. Department of State and an initial group of international partners in 2020. Mauritius became the 70th signatory on July 17, 2026. NASA’s signatory list continued to show 70 participating nations as of August 18, 2026. The accords set principles for civil space exploration and reinforce commitments associated with existing space treaties and responsible behavior.

Commercial competition and geopolitical competition can reinforce each other. A government may support domestic launch capability because it wants dependable national access. That support can give domestic companies enough demand to improve technology and lower costs, which can make them stronger commercial competitors.

The same dynamic works in reverse. A commercially successful company may develop infrastructure that later becomes strategically valuable to its home government. Space policy, industrial policy, defense procurement, telecommunications regulation, investment policy, and national security policy are becoming more closely connected.

Orbital Congestion and Governance Can Restrain Expansion

Industrial growth brings physical constraints that financial markets cannot remove. Orbital regions have finite practical capacity, radio-frequency spectrum requires coordination, spacecraft must avoid collisions, debris can remain in orbit for long periods, and operators depend on shared rules for registration, communications, maneuver coordination, disposal, and interference management.

Goldman’s analysis places governance beside capital and technology because many newer activities do not fit neatly into regulatory systems created during an earlier era. Space traffic management, debris removal, lunar resources, in-space manufacturing, proximity operations, and new spacecraft categories can cross national agencies or lack a single established regulatory pathway.

International space law still rests heavily on agreements negotiated during the Cold War. Article VI of the United Nations Outer Space Treaty makes states internationally responsible for national space activities and requires authorization and continuing supervision of nongovernmental space activities. Commercial expansion consequently does not remove government responsibility from the legal structure.

Spectrum and orbital assignments introduce another constraint. Describing the International Telecommunication Union process simply as ownership of orbital slots on a first-come basis is incomplete. The ITU Space Services Department administers regulatory and technical procedures associated with satellite systems and Earth stations, including frequency-assignment processing, coordination requirements, and conformity with the Radio Regulations. The ITU describes satellite-frequency registration as a cooperative international process intended to prevent harmful interference and preserve equitable access to limited spectrum and orbital resources.

National regulators are adjusting their own systems. On July 22, 2026, the U.S. Federal Communications Commission adopted its Space Modernization for the 21st Century Report and Order and Further Notice of Proposed Rulemaking, released July 23. The order created a new Part 100 licensing framework that the commission describes as a licensing assembly line intended to provide greater speed, predictability, flexibility, and regulatory certainty for space operators.

Collision risk may prove harder to solve because it is governed by physics rather than administrative procedure. The European Space Agency’s 2026 Space Environment Report documents the continuing growth of objects in Earth orbit and the long-term risks created by debris-generating collisions. The report’s modeling shows why future debris conditions depend on more than future launch activity: collisions among objects already in orbit can generate additional debris and increase environmental risk.

This creates an economic cost that grows with activity. Operators must invest in tracking, maneuver planning, propulsion, fuel reserves, communications, coordination software, insurance, spacecraft identification, disposal systems, and operational staffing. A more congested orbit can make every new satellite somewhat more expensive to operate safely.

The issue also complicates the assumption that lower launch prices automatically produce lower system costs. If inexpensive launch encourages much larger constellations, some savings can be offset by greater spending on network management, spectrum coordination, debris mitigation, replenishment, and collision avoidance.

Space industry expansion consequently depends on institutions as much as machinery. Regulators need enough predictability for companies to invest, yet permissive rules that ignore congestion or interference can impose costs on other operators. Sustainable expansion requires balancing access with preservation of shared orbital and spectrum resources.

Orbital Industrialization Is Still a Portfolio of Experiments

Goldman’s most ambitious theme concerns orbital industrialization. The paper considers in-space servicing, assembly and manufacturing, microgravity production, refueling, orbital computing, lunar infrastructure, resource extraction, energy production, and related activities that could move selected industrial functions away from Earth’s surface.

Evidence is strongest for activities that extend existing spacecraft operations. NASA’s in-space servicing, assembly, and manufacturing program develops technologies for servicing, assembling, and manufacturing space systems after launch. These capabilities could change spacecraft design by allowing operators to maintain, reposition, assemble, or modify assets after they reach orbit.

NASA is also preparing the Fly Foundational Robots mission, which remained scheduled for launch in late 2027 as of August 18, 2026. The technology demonstration is designed to deploy and operate a commercial robotic arm in low Earth orbit, creating an orbital platform for robotic demonstrations. NASA identifies Rocket Lab Robotics, Astro Digital, Arkysis, SpaceWorks, iBoss, and PickNik among the mission partners.

The commercial logic is straightforward in principle. A satellite that can be refueled, inspected, repaired, moved, or upgraded might remain productive longer than a spacecraft designed for disposal when one subsystem fails. Servicing can also change spacecraft design by reducing the need to carry every spare component or all lifetime propellant from launch.

The in-space economy already contains activity in station services, lunar delivery, satellite life extension, cargo transportation, orbital transfer, research services, inspection, and technology demonstrations. Those markets sit at different stages of maturity. Cargo delivery and satellite life extension have operational precedents, whereas broad orbital manufacturing and resource extraction remain much less proven commercially.

Orbital computing has attracted attention for a different reason. Terrestrial artificial intelligence development requires growing quantities of electrical power, cooling infrastructure, grid connections, data centers, networking, and advanced processors. Solar energy is abundant in orbit, but that advantage comes with difficult engineering requirements involving heat rejection, radiation exposure, maintenance, data movement, launch cost, hardware replacement, and communications latency.

An orbital data center architecture could eventually allow spacecraft to share computing, storage, networking, and mission-management resources rather than carrying all processing hardware independently. Such designs depend on high-capacity optical communications, reliable orbital networking, autonomous operations, power management, thermal control, and serviceable computing hardware.

Microgravity manufacturing faces a different commercial test. NASA’s In-Space Manufacturing Portfolio Plan describes research and technology development for producing materials, components, and systems in space. Commercial success requires more than proving that a manufacturing process works in microgravity. A material, pharmaceutical product, semiconductor, biological sample, or specialty component must provide enough additional value to offset launch, spacecraft, operations, insurance, processing, validation, and return costs.

Lunar commerce carries the same distinction between technical feasibility and customer demand. Governments can purchase transportation, communications, navigation, power, construction experiments, and scientific services for lunar missions. Those contracts can sustain early providers, but a self-supporting lunar market requires customers whose demand extends beyond a small number of public exploration programs.

Goldman’s concept of orbital industrialization is consequently most useful as a framework for evaluating markets rather than as a prediction that large portions of terrestrial industry will soon migrate into space. Communications already demonstrate industrial-scale demand. Servicing and logistics are moving toward operational markets. Manufacturing, computing, resource extraction, and permanent lunar industry require much more evidence.

Summary

The defining feature of a second space age is not a return to the Moon, a record number of launches, or the appearance of another constellation by itself. It is the growing dependence of ordinary economic and government activity on infrastructure that operates beyond Earth.

Goldman Sachs Global Institute’s framing captures that structural change. Launch vehicles are becoming transportation infrastructure. Satellite constellations operate as communications networks. Earth observation feeds commercial data services. Positioning and timing systems support financial and industrial systems. Capital markets finance infrastructure once funded almost exclusively by governments.

The commercial model remains uneven. Satellite broadband has reached millions of service lines, yet many in-space manufacturing concepts remain demonstrations. Public markets can provide extraordinary amounts of financing, yet expensive financing can still eliminate companies that lack recurring demand. Lower launch prices make more projects technically possible, but they do not solve spectrum scarcity, congestion, regulation, insurance, customer acquisition, or weak economics.

National strategy is becoming inseparable from these commercial developments. Governments increasingly depend on privately operated launch vehicles, satellite networks, imagery, software, and data. Companies depend on government contracts, licensing, spectrum coordination, research funding, infrastructure, and legal protections. Neither side operates independently.

The most useful measure of progress may consequently be dependence rather than novelty. A technology becomes economically significant when other industries begin to assume that it will be available, affordable, and reliable. Satellite navigation crossed that threshold decades ago. Satellite communications are moving further in that direction as LEO networks expand. Earth observation and space-derived analytics continue to enter terrestrial decision systems.

Orbital servicing, manufacturing, computing, lunar logistics, and resource-related activity may follow, but each will need repeat customers and economics that survive without permanent dependence on demonstration funding. The industrial age of space will be established less by what companies can place in orbit than by what customers repeatedly pay them to do once they get there.