This publication has already covered space twice — the broad economics of orbital resource access, and the debris risk of putting data centers up there. Neither covered what actually changed the space industry in the past three years: someone figured out how to build satellites the way Detroit builds cars, and the entire sector's economics broke in response.
Five satellites leave SpaceX's Hawthorne, California production line every single day. That number is the actual story — not the asteroid belt, not orbital colonization, not debris litigation. A manufacturing process that used to take years per unit and cost aerospace-grade money now runs on something closer to a consumer-electronics production cadence, and that shift is quietly restructuring an entire industry's economics.
Satellite manufacturing cost has fallen from roughly $100,000 per kilogram to approximately $1,000 per kilogram for Starlink-class satellites — a hundredfold collapse achieved through mass-production technique, not a new physics breakthrough. New Space Economy's own reporting is specific about the mechanism: SpaceX treated satellites "less like aerospace hardware and more like consumer electronics," applying iterative development and high-volume production methods to an object that historically required years of custom engineering per unit. The Starlink constellation crossed 10,000 active spacecraft in March 2026, and the global satellite manufacturing market itself is projected to grow from $21.8 billion in 2025 to $86.7 billion by 2035 — a 14.8% compound annual growth rate driven overwhelmingly by this exact shift in production philosophy, not by any single new application.
The competitive response confirms this is now genuinely an industrial-manufacturing race, not a space race. Amazon has built dedicated manufacturing facilities and supply-chain relationships specifically for high-volume LEO satellite production, targeting 3,236 satellites for its Kuiper constellation on a committed $10 billion budget. Telesat contracted 14 separate SpaceX Falcon 9 launches, each carrying roughly 15 satellites, after completing a 180,000-square-foot dedicated manufacturing facility. MDA Space, Eutelsat/OneWeb, and AST SpaceMobile are all scaling comparable production infrastructure. This is a genuine factory-floor arms race, fought on manufacturing cadence and unit cost, not on orbital ambition.
The other half of this story is the bill nobody sends directly. The European Space Agency tracks roughly 40,000 objects in orbit, of which about 11,000 are active payloads — against an estimated 1.2 million pieces of debris larger than 1cm that current tracking cannot fully catalogue. Fortune Business Insights' own market analysis is direct about the resulting cost structure: orbital congestion now drives a longer, more expensive checklist for every operator — extra maneuver planning, fuel reserves for collision avoidance, conjunction screening, deorbit planning, tighter insurance scrutiny, and launch-licensing compliance — with smaller companies deploying large constellations facing disproportionately higher relative costs than incumbents who can spread that overhead across thousands of units. This publication's own Letter 142 (The Orbital Debt) priced the debris-collision-risk side of this equation directly for orbital data centers specifically; this letter is pricing the same underlying congestion cost as it applies to the commercial satellite manufacturing race broadly — a distinct but related bill coming due from the same physical reality.
The satellite industry didn't get cheaper by inventing something new in orbit. It got cheaper by importing a production philosophy from an entirely different industry — and every competitor now has to match that manufacturing cadence or accept a permanent cost disadvantage.
This publication's thesis (Section 02: Space Colonisation) makes the broader case for space economics generally — falling launch costs, asteroid-belt resource value, the multi-decade colonization thesis. Letter 142 covers the specific debris-collision risk of putting AI data centers in orbit. Neither covers what this letter is pricing: the industrialization of satellite manufacturing itself as a distinct, current, investable trend, separate from what any given satellite is actually used for once it's built. A reader who wants the full picture should treat these as three layers of the same stack — colonization economics (thesis), applications and their risks (Letter 142), and now, manufacturing (this letter) — rather than three attempts at the same argument.
The Manufacturing Layer Itself. This is a genuinely difficult layer to access directly — SpaceX, Amazon (Kuiper), and Telesat's satellite manufacturing operations are either private or bundled inside much larger public companies where the segment doesn't move the share price meaningfully. MDA Space (TSX: MDA) is the cleanest pure-play public exposure to the actual manufacturing-and-production side of this story specifically, rather than the launch or application layers this publication has covered elsewhere.
The Congestion-Compliance Layer. As orbital traffic-management requirements tighten, companies providing conjunction-screening, tracking, and space-situational-awareness services capture real, recurring revenue independent of which satellite manufacturer wins the production race — a picks-and-shovels position on the congestion cost this letter documents directly, similar in spirit to this publication's own framework in Letter 149 (Pricing the Impossible) for backing infrastructure that pays off across multiple competing outcomes.
The real story in space right now is not colonization and it is not debris litigation — both real, both already covered by this publication elsewhere. It is that satellite manufacturing has industrialized, with unit costs falling roughly a hundredfold and production cadence reaching five spacecraft per day at the category leader, forcing every competitor (Amazon, Telesat, Eutelsat, AST SpaceMobile) into a genuine factory-floor arms race measured in manufacturing capacity, not launch ambition. MDA Space is the cleanest available public exposure to that specific layer. The congestion cost this manufacturing race is generating — tracked directly by ESA, priced indirectly through rising compliance and insurance overhead — is the same underlying physical reality this publication already named in Letter 142, now showing up as a broader industry cost rather than a single-application risk. The risk to this thesis: if orbital traffic-management regulation tightens faster than expected, smaller manufacturers may be squeezed out entirely, consolidating the entire industry around two or three vertically-integrated players before the broader market this letter describes has a chance to diversify.
Founder, NextGen Economics · Bangalore, India · July 2026
Sources: New Space Economy ("The Satellite Manufacturing Market After Starlink," Apr 2026) · Fortune Business Insights (LEO Satellite Market Report, Satellite Mega Constellations Market Report, 2026) · SpaceNexus Blog ("5 Space Industry Trends Reshaping the Market in 2026") · Nova.space ("Space Industry Outlook 2026") · Via Satellite ("The Coming Wave of Competition in LEO Constellations," Mar 2026) · European Space Agency (tracked-object statistics) · New Market Pitch (Space Economy Market Size 2026). Builds on this publication's own Letter 142 (The Orbital Debt) and thesis.html Section 02 (Space Colonisation).
Not investment advice. This letter evaluates a geoeconomic policy trend; it does not constitute a recommendation regarding any security.